|
The following papers were presented in
WSEAS ICONEMC 2002, ICOSMO 2002, ICOSSIP 2002, ICOMIV 2002 and ICRODIC 2002.
(Data Base in MicroSoft Access 2002).
|
| 73 Traffic Analysis for Voice in Wireless IP Networks |
pp. 731-737 |
|
|
by : Toni Janevski, Boris Spasenovski
Abstract: - In this paper we provide an extensive analysis of voice service in wireless IP networks. Voice
traffic is serviced with priority over the rest of the traffic. Hence, we concentrate on the voice traffic in the
analysis. We model the packetized voice traffic with Markov Modulated Poisson Process (MMPP), and we
propose an analytical framework for its analysis in wireless networks. Also, we created a simulator in
Matlab, with capability for QoS analysis in different network scenarios, considering user call intensity,
voice-encoding rate, link capacity and buffer sizes. The observed QoS parameters are packet loss and
delay. Voice traffic is very sensitive to delay, while some low losses may be tolerated. We present
overwhelming QoS analysis of IP telephony traffic at different network setups and give a concept for
dimensioning wireless links for IP telephony under given constraints on the QoS parameters. |
| 104 Multiwavelength optical fibre source |
pp. 1041-1045 |
|
|
by : Jan Lamperski, Piotr Stepczak
Abstract: - In this paper we described a structure of an original multiwavelength optical fiber source for DWDM systems. In order to obtain a several dozen of optical carriers with highly stable channel spacing of 4.5GHz a single laser source and acousto optic frequency shifter were used. Results of the analysis of the structure properties were presented. |
| 108 Bandwidth tolerances of cascaded filters in WDM optical networks |
pp. 1081-1083 |
|
|
by : Mario Lima, Antonio Teixeira, Paulo Andre, Jose Rocha
Abstract: - In this contribution we develop a theoretical statistical approach to predict the acceptable bandwidth tolerances for optical filter cascades in wavelength division multiplexing (WDM) networks. We also establish tolerances for the center frequency misalignments considering a statistical approach presented in a previous work. The optical filters analyzed have a trapezoidal shape, and different bandwidths are considered for the study. The analytical results are validated with system simulations. |
| 109 Phonon Spectra in Crystalline Nanostructures |
pp. 1091-1096 |
|
|
by : J. P. Setrajcic, S. K. Jacimovski, D. Rakovic, D. I. Ilic
Abstract: - Considering the models of spatially bounded structures and structures with broken translational symmetry (ultrathin films, quantum wires and dots), we have found the phonon spectra using the method of two-time commutator Green's functions. The phonon spectra are distinctly discrete and possess gaps (which are the consequence of the spatial boundedness of the system). Using the definition of the activation temperature of the phonon subsystem we connected these results with better superconductive properties of micro samples, which can be a possible explanation of high critical temperature of HTC ceramics. |
| 110 Need of Knowing Fiber Non-linear Coefficient in Optical Networks |
pp. 1101-1106 |
|
|
by : Bostjan Batagelj
Abstract: - This paper describes the need of knowing fiber non-linear coefficient in global optical networks. The basic Kerr non-linearities that appear in silica-based optical fibers have been summarized, together with their implications for optical communication systems. In this paper, the non-linear coefficient has been discussed and its measurement techniques reviewed. The brief description of each method is given. This paper focuses on the comparison of different interferometric and non-interferometric measurement schemes in terms of their simplicity and versatility of testing various types of fibers. Accurate determination of fiber non-linearities is an important issue in the design of optical systems. |
| 111 Exploiting enhanced reliability features in MPLS traffic engineered optical speed networks |
pp. 1111-1116 |
|
|
by : Francesco Palmieri
Abstract: - The demand for multimedia applications along with rapid growth, the need for quality of service, reliability, and resilience have made traffic engineering an essential consideration in the design and operation of large public Internet backbone networks. In order for carriers to minimize the financial impact of inevitable link and network component failures, backbone protection solutions must be implemented which can rapidly reroute traffic across redundant links to minimize/eliminate service disruption. This paper details traffic engineering strategies for carriers to provide rapid, cost-effective and physical layer independent mechanisms, based on the latest MPLS traffic-engineering features to protect against link failures in the network core. Such mechanisms have been experimentally investigated, from the performance point of view, on a state-of-the-art high performance optical network, to evaluate how they can improve the reliability and resilience of today's Internet, and identify opportunities for further improvement, and development of new fault-recovery mechanisms for future networks. |
| 118 A Particle Model for Wigner Transport through Tunneling Structures |
pp. 1181-1183 |
|
|
by : Hans Kosina, Mihail Nedjalkov, Siegfried Selberherr
Abstract: - The Wigner equation taking into account both the non-local potential operator and a scattering operator represents a suitable physical model for carrier transport in nano structures. In this paper a new Monte Carlo method is proposed for the solution of the stationary Wigner equation. Scattering can be included at the level of the semi-classical Boltzmann scattering operator, whereas coherent effects are treated without simplifying approximations. As opposed to the Monte Carlo method for semi-classical transport, in the quantum case the weight of a particle can take on positive and negative values. |
| 122 Soliton Dynamics in Dispersion-Varying Compensating Fibers (DVCF) |
pp. 1221-1224 |
|
|
by : Alessandro M. Melo, M. C. Gouveia, Henrique J. A. da Silva
Abstract: - We investigated analytically and by numerical simulations the possibility of dispersion managed solitons in links with dispersion varying compensating fibers. We present an analysis where the conventional DCF (Dispersion Compensating Fiber) is replaced by a fiber with varying dispersion. We have studied the case of DDF (Dispersion Decreasing Fiber) and DIF (Dispersion Increasing Fiber), for the DVCF dispersion profiles. Results show that the soliton dynamics depends on the DVCF profile providing an evidence of system performance improvement. |
| 123 Planning Reliable UMTS Access Networks |
pp. 1231-1236 |
|
|
by : Attila SzlovencsÜk, JÜnos Szigeti
Abstract: - The access network of the Universal Mobile Telecommunication System (UMTS) forms a tree--topology network, however, the great amount of traffic carried in it requires a more reliable network structure. That way, new methods are needed to plan a minimum cost extension of the classical topology. In order to satify the reliability expectation, expressed as a limitation for the traffic loss in the network, we insert additional links, while taking into account several topological constaints. In the current paper, we introduce a genetic algorithm to solve the specific problem, and we compare the results of it with the ones produced by a greedy heuristic. |
| 124 A distributed fairness mechanism for slotted WDM rings |
pp. 1241-1247 |
|
|
by : John Angelopoulos, Catherine Leligou, Harris Linardakis, Alexandros Stavdas
Abstract: - The use of Metropolitan rings based on WDM technology is spreading in response to strong demand for IP traffic. In slotted rings employing spatial re-use the problem of fairness inherent in the ring topology is quite significant. To combat this and allow all nodes around the ring to equally share the available bandwidth a novel mechanism is proposed based on access credits. The mechanism is simple and executed in a distributed way. Its performance is evaluated by simulation to show its effectiveness even under highly asymmetric loading. |
| 125 Molecular Nanotechnology: Golden Mean as a Driving Force of Self-Assembly |
pp. 1251-1256 |
|
|
by : Djuro Koruga
Abstract: - In this paper we are considering self-assembly approach from nanobiology to molecular nanotechnology. Because the genetic code, as a most complex biomolecular system, is determined by Golden mean, we used it knowledge to study approaches to nanotechnology. Understanding protein self-assembly driving force and structure of clathrin and microtubules based on Golden mean we have found that fullerene C60 and nanotubes could be very useful materials for molecular nanotechnology. One possible solution as a sample is given. |
| 127 Nanophtonics and Consciousness: An Anesthetics-Luciferase Appruoch |
pp. 1271-1274 |
|
|
by : Jovana Simic-Krstic, Magdalena Ivetic1, Djuro Koruga
Abstract: - In this paper we present results of experimental research of luciferase-anesthetic interaction in vitro. To characterise secondary structure of luciferase we used circular dichroism (CD), while to characterise luciferase-anesthetic interaction we have used scanning tunneling microscope (STM). Understanding molecular mechanism of anesthesia may illuminate biological consciousness. Because the week, non-chemical process based on charge-dipole and dipole-dipole interactions are responsible for photonic swich we believe that nanophotonics and molecular optical networks under determined conditions, in future may lead to new type of information machines based on artificial consciousness. |
| 140 Adapting Content In a Portal Of Academic Information Based On XML and PHP |
pp. 1401-1403 |
|
|
by : E. Carrillo, S. Guerero, G. Pabon, J. J. Martinez-Dura, J. J Samper
Abstract: - With the consolidation of Internet as a means to share information in the new society and the need to access such information from mobile devices, such as cellular telephones, PDAs, WebTVs, Playstations, and others, we are now facing the problem of making some extra work in projects that generate contents for the Internet and that are to be visualized in traditional browsers and in devices with a small size and memory available. The present paper deals with the problem of content generation for traditional browsers taking an academic point of view. The paper also tries to explain the strategy used and the technologies of public dominion that gave support to the work that has been carried out in the last year for the development of a dynamic application that can be executed from PCs with Web browsers and from mobile devices. The framework that was being designed is based on the different models of cellular telephones with Internet capabilities available in Colombia |
| 141 Development of a Solution for Editing and Compiling of Wireless Internet Applications |
pp. 1411-1412 |
|
|
by : E. Carrillo, J. J. Samper, J. J. Martinez-Dura, V. Ramon Tomas
Abstract: - In this paper a summary of an academic experience related to the design and development of a tool for the editing and compilation of dynamic WAP (Wireless Application Protocol) applications is presented. in which it was included support for WML (Wireless Markup Languaje) version 2.0 and WMLScript, as well as the possibility of interaction with different technologies for interaction with databases in the server-side as: Java Server Pages, Hipertext Proccessor and Active.server Pages. |
| 159 A model for soliton charge transport through microtubules |
pp. 1591-1594 |
|
|
by : Goran Kekovic, Dejan Rakovic, Miljko Sataric, Djuro Koruga
Abstract: - In this paper the theoretical model for charge transport through microtubules (MTs) by soliton mechanism is developed, and analyzed by methods of nonequilibrium statistical physics. Thus obtained electrical current through MT is estimated as ~ 0.1 nA. |
| 166 Transmission Coefficient Estimation for High-k Gate Stack Evaluation |
pp. 1661-1664 |
|
|
by : Andreas Gehring, Hans Kosina, Siegfried Selberherr
Abstract: - We investigate methods to estimate the transmission coefficient of high-k gate stacks. Based on the commonly used Tsu-Esaki model, we approximate the WKB and Gundlach methods to be feasible for device simulators. Comparisons with a rigorous solution using the transfer-matrix method show good qualitative agreement. We further use the models to analyze the trade-off between barrier height and permittivity in different high-k dielectrics and identify ZrO2, Al2O3 and possibly Ta2O5 as the materials with appropriate gate current blocking ability. |
| 167 PNNI-based Routing Algorithm in ATM Network |
pp. 1671-1675 |
|
|
by : HyunSoon Shin, Sang-Ho Lee
Abstract: - In this paper, we propose an implementation scheme of source routing function, one of the characteristics, in applying PNNI protocol to ATM switching system. The proposed substance is a PNNI routing software structure design and source routing algorithm. The software structure guarantee maximal independence among blocks by keeping a function manager, composing a handling block separately, working together between those. As the path selection scheme has characteristics to choose real-time path and establish a path satisfying QoS, the source routing algorithm is desirable that pre-calculated scheme together with on-demand scheme is used. |
| 170 Proposition of a new MAC layer for wireless networks |
pp. 1701-1708 |
|
|
by : Val Thierry, Mercier Jean-Jacques
Abstract: - This paper details an original access method that can be used for all types of network using broadcast media (ad-hoc WLAN for example). This RNET protocol is a major development of the COMB method while bringing uniformity in the choice of codes and priorities. This method offers numerous advantages with respect to existing MAC layers: the determinist aspect, the guarantee of minimum bandwidth and resource reservations with the prior knowledge of the state of the network. This method is mainly interesting for home wireless LAN, for topologies based on a single cell, and/or multi-cell cover, for which it has also been adapted |
| 174 Analyzing the effects of ElectroMagnetic Interference on state-of-the-art copper twisted pair Ethernet cabling |
pp. 1741-1746 |
|
|
by : Francesco Palmieri, Angelo Violetta
Abstract: - With the dramatic increases in desktop computing power driven by Intranet-based operations and the growing demands for time-sensitive delivery between networked users, office and production-level Ethernet LAN technology has rapidly evolved from a coaxial cable running at 10 megabits per second to twisted pair running data rates up to 1 gigabit per second. As a side effect of this evolution, the higher the transmission rate grows the more sensitive the communication system will be to external noise inducted by Electromagnetic Interference (EMI). Consequentially, the designers, installers and users of Ethernet networks must consider the effects of EMI as a key factor in cabling design, setup and operation. Operating and environmental conditions in terms of Electromagnetic compatibility, should be seriously taken into account when choosing materials and designing cable routing layouts especially in industrial plants and high density computer rooms. In this paper we performed a detailed analysis on the disruptive effects of EMI on the most common state-of-the-art copper twisted pair Ethernet cabling technologies. Transmission parameters, noise/disturbance levels and network errors have been extensively examined and compared under varying EMC environmental condition to provide clear and up-to-date installation practice guidelines in terms choice of materials and separation requirements from suspected sources of interference, aiming to optimize communication channel characteristics and yield the maximum possible throughput. |
| 176 Modification of the surface coat of nanometer-sized magnetoliposomes by phosphatidylethanolamines. |
pp. 1761-1764 |
|
|
by : Marcel De Cuyper
Abstract: - Using high-gradient magnetophoresis, the non protein-mediated transfer of dimyristoyl-phosphatidylethanolamine (DC14:0PE) and dipalmitoylphosphatidylethanolamine (DC16:0PE) between sonicated phospholipid vesicles and magnetoliposomes was followed. The latter structures consisted of nanometer-sized magnetite cores which are surrounded by a phospholipid bilayer. In accordance with the 'aqueous transfer model', it is shown that the transferability strongly depends on the fatty acyl side chain composition of the PE entity used. With DC14:0PE a halftime for transfer of 912 min is found, while, using DC16:0PE no transport occurred within the time scale of the experiment (3 days). The relevance of these observations to the production of colloids of biomedical importance is briefly discussed. |
| 178 Optical Fiber Devices and Their Applications |
pp. 1781-1786 |
|
|
by : Yutaka Sasaki
Abstract: - Recent progress in research on optical fiber devices, especially optical fiber couplers and optical fiber grating couplers, by a new fabrication method, that is, a CO2 laser irradiation method is reviewed. Furthermore, applications of optical fiber grating couplers are also described. |
| 179 Bragg grating synthesis by Fourier transform |
pp. 1791-1795 |
|
|
by : J. C. Aguado, P. Fernandez, J. Blas, I. de Miguel, F. Gonzalez, J. Duran, R. M. Lorenzo, E. J. Abril, M. Lopez
Abstract: - We present a method for the synthesis of Bragg gratings using a Fourier Transform-based algorithm. The algorithm has an easy implementation, yields to reasonably accurate results and is fast due to the use of FFTs. Moreover, it resolves partially some of the problems faced by other Fourier transform methods when applied to high reflection problems. |
| 181 Simulation and Design Tool for Spectral Characterization of Fiber Bragg Gratings |
pp. 1811-1815 |
|
|
by : Patricia Fernandez, Juan Carlos Aguado, Juan Blas, Fernando Gonzalez, Ignacio De Miguel, Javier Duran, Ruben M. Lorenzo, Evaristo J. Abril, Miguel Lopez
Abstract: - The optimization of apodized linearly chirped fiber Bragg gratings in the field of chromatic dispersion compensation is discussed in terms of group delay response and pulse recompression. To develop this study two prototype scenarios are considered to show the different behavior of the compensating devices in function of the link length. Simulation results for 20 ps gaussian pulses transmitted and recompressed with different grating designs over the both links are presented. The maximum deviation error and the regression slope parameters are introduced to compare the quality of the group delay response to establish an optimum design of the apodization strength. |
| 183 Water-Soluble Fullerene C60(OH)24 Modulates Growth and Proliferation of K562 Human Erythroleukemia Cell Line |
pp. 1831-1836 |
|
|
by : Dimitar Jakimov, Gordana Bogdanovic, Jasminka Mrdanovic, Aleksandar Djordjevic, Slavica Tursijan, Vladimir Galic, Mirjana Vojinovic-Miloradov
Abstract: - Fullerenes and their water-soluble derivatives have recently become a topic of interest in biomedicine. Here we report on growth and proliferation activity of fullerenol C60(OH)24 on K562 human erythroleukemia cells. Capability of K562 cells to synthesize a DNA by incorporation of radiolabeled thymidine, their mitotic activity and cell cycle distribution has been evaluated 3, 24 and 48 hours after fullerenol treatment. Viability of fullerenol treated K562 cells was up to 5% lower in all samples in comparison to control, regardless C60(OH)24 concentration or incubation period of cells with fullerenol. No significant differences were found in morphological appearance between control and fullerenol treated cells analyzed by light microscopy. A mitotic index of fullerenol treated cells was decreased in comparison to control regardless the period of cell incubation with fullerenol. Synthesis of DNA of fullerene treated K562 cells was inhibited, but only at the highest fullerenol concentration. The cell cycle distribution pattern of K562 cells treated with fullerenol at concentration of 0.041umol/l was comparable to that of the control but increasing number of cells in G2M phase was observed during the treatment period. Water-soluble [C60] fullerene derivative, C60(OH)24, modulates cell cycle of the K562 cell influencing both synthetic and mitotic cell cycle phase |
| 184 Two-Dimensional Modeling of Quantum Mechanical Effects in Ultra-Short CMOS Devices |
pp. 1841-1844 |
|
|
by : Klaus Dragosits, Vassil Palankovski, Siegfried Selberherr
Abstract: - Quantum mechanical analysis of the quantum confinement of ultrashort CMOS is numerically very expensive. In this paper we present a macroscopic model, which includes a new approach to match the vertical carrier profile and combines it with a classical model in lateral direction. The simulation results show a significant improvement concerning the accuracy of the carrier profile and the C/V characteristics. |
| 187 A Compact Carrier Quantization Model for Nanoscale MOSFETs Simulation |
pp. 1871-1876 |
|
|
by : Yiming Li, Yen-Yu Cho, Chuan-Sheng Wang, Pu Chen, Ting-Wei Tang
Abstract: - Due to the advanced fabrication technology, the oxide thickness is now in the regime of 1.0 nm for nanoscale MOSFETs. The quantization effects and inversion charge density displacement away from the interface of oxide and silicon can not be neglected. The most accurate model for such problem is solving the Schrodinger-Poisson (SP) equations with proper boundary condition in 1D structure. However, for 2D device characteristics and circuit dynamics, solution of the SP model encounters numerical difficulty and is not ready for practical applications. Various quantum correction models have been proposed for theoretical exploration and verification. In this paper we study the quantization effects and for the first time develop a corresponding charge analytical model in terms of oxide thickness and applied voltage for ultrathin oxide MOSFETs. Based on a comprehensive investigation of charge peak location, peak value, averaged charge displacement, and charge density, the successfully derived compact model accounting for the quantization effects enables fast and accurate characterization of the effective charge density in nanoscale MOSFETs. This new model has computational superiority and can be directly applied for nanoscale device and circuit simulation without solving the SP equations. Compared with the measured C-V data of an ultrathin N-MOSFET, our simulation results demonstrate the model accuracy. |
| 192 EMC and LV Directive Enforcement in Finland – an Analysis Using the EFQM Model |
pp. 1921-1926 |
|
|
by : Jyri Rajamaki, Seppo Tuominen, Veli-Pekka Nurmi, Reijo Mattinen, Hannu Mattila
Abstract: - - The EMC and the safety enforcement of electrical equipment in Finland changed when Finland joined the EU. Today, enforcement is based on international obligations with The Safety Technology Authority of Finland (TUKES) being responsible for maintaining it in Finland. This paper examines the methods, effects and foremost results of TUKES’s market surveillance, especially with regard to the EMC and Low Voltage Directives. The study is a piece of comparative qualitative research which has capitalised on the European Foundation for Quality Management (EFQM) model. |
| 194 Optimization of Microwave Power Absorption in Biological Tissues |
pp. 1941-1944 |
|
|
by : D. Razansky, D. F. Soldea, and P. D.
Eintziger
Abstract: - The phenomenon of microwave power absorption has recently become of increased public and research interest in various areas including such research topics as an EM power absorption by human head with closely coupled hand-held mobile phone and the problem of a perfect matched absorbing layers. Herein, we focus on a simplified 1-D transmission-line model. This model, being directly associated with the problem a normal incidence of the plane-wave generated by infinite source sheet upon a lossy half-space, establishes a tight upper bound on the maximal (optimal) power absorption in a realistic mobile phone - human head configuration. Furthermore, the source impedance and the absorption/radiation efficiencies are obtained via an explicit closed-form expressions, leading to an explicit optimal power absorption criteria for highly lossy tissues. |
| 197 Evaluation of E-business applications performance |
pp. 1971-1976 |
|
|
by : Ines Brosso,Graca Bressan,Wilson Ruggiero
Abstract: - This paper proposes the use of XML Petri Nets (PNML), a kind of high-level Petri Nets, for modeling, development, management and performance Web-based systems, e-Activities like e-Business, Business-to-Business (B2B), e-Commerce and mobile applications wireless in the Web or ad-hoc networks. The combination of Petri Nets and XML (Extensible Markup Language) files are emerging as an important approach for the specification of complex processing business tasks in distributed heterogeneous systems, allowing the follow-up of electronic business documents and data interchange process. In a Case Study, the proposed methodology is applied in the development of an e-business application, the customer surfs in the catalogue of products and adds desired items to its basket of purchases, the system verifies the authorization of the credit card and confirms the purchase immediately and soon sends an email to follow contends an identification code so that the customer follows its purchase. |
| 102 What might go wrong when trying to globalize e-business |
pp. 1021-1025 |
|
|
by : Zeljko Panian
Abstract: - Enterprises in a small country, like Croatia, can enjoy many benefits from e-business globalization, but they can also suffer from it if they are not able to find the right way to accept new rules of their business behavior. A vast majority of Croatian executives have no doubt any more whether their e-business should be globalized or not. Today, they are not asking something like “Why globalize”; usually their principle question is “How do we globalize?” This shift in perspective signals a growing commitment to a fact that globalization has become a critical component of e-business, as important as content management, personalization, or customer relationship management – to name only a few of the current e-business initiatives on the enterprise agenda. During the second half of 2001, the globalization efforts of about forty Croatian companies were surveyed, and some common mistakes were identified that had hampered them and continue to plague them today. The most frequently observed mistakes were: assuming that the Internet is inherently global; equating globalization with translation; deploying web code that is not internationalized; divorcing the globalization plans from revenue objectives; thinking of globalization as the last step in the process; and trying to manage globalization using only a Content Management System (CMS). Sources and possible consequences of these mistakes are discussed in the paper. |
| 108 Reaction neuron network model for image perception |
pp. 1081-1086 |
|
|
by : Han Wang, Weichi Sun, Xuelin Cheng and Bingzhe Liu
Abstract: - A new neuron network model, as an assistance of image perception, is proposed in this paper. It is based on the retina. And it is shown that this neuron network model can both describe the transfer property of retina well and neglect subordinate details when processing an image. Specially, we also describe how to implement this model and make some experiments on it. |
| 112 Safety relevant design of a fuzzy controller |
pp. 1121-1128 |
|
|
by : Gerhard H. Schildt, Daniela Kahn
Abstract: - After an introduction into safety terms a fuzzy controller for safety related process control will be presented. One can show that the number of necessary rules is relatively small. Thus, there exists a real chance for verification and validation of software due to the fact that the whole software can be structured into standard fuzzy software ( like fuzzyfication, inference algorithms, and defuzzyfication ), real-time operating system software, and the contents of the rule base. Furthermore, there is an excellent advantage due to real-time behaviour, because program execution time is much more planable than for conventional PID-controller software. Additionally, up to now special know-how does exist to prove stability of fuzzy controller. Hardware design has been done due to fundamental principles of safety technique like watch dog function, dynamization principle, and quiescent current principle. |
| 114 Improvement to the Smart Data Server with SOAP |
pp. 1141-1145 |
|
|
by : Wanjun Huang, Uwe Roth, Christoph Meinel
Abstract: - As a distributed computing middleware, the Smart Data Server (SDS) provides a general frame for easy-to-build environment-independent modules with distributed functionality and a secure mechanism for invocation. But there are still some problems in the widespread cooperation with other distributed computing system and client terminal. Nevertheless, SOAP, as a new promising simple object access protocol, provides a mechanism based on XML for the structured message exchanging in the distributed environment. Here we take SOAP as the protocol for message transport between the remote client and the smart data server instead of Information Package Transfer Protocol (IPTP), and the SDS gains some new advantages: the remote client could be in the any platforms and implemented with any languages; it can cooperate with other distributed computing solutions, such as CORBA, DCOM, RMI and so on; the SDS can be applied more widely for the SOAP message transferred through HTTP, a wide and most common used protocol in the Internet. |
| 115 An Engineering Model of Coils and Heat Exchangers for HVAC System Simulation and Optimization |
pp. 1151-1156 |
|
|
by : Yaowen Wang, Wenjian Cai, Yengchai Soh, Shujiang Li and Lihua Xie
Abstract: - This paper presents a technique for developing a simple yet practicable model that can accurately replicate equipment performance of cooling coil units and heat exchangers. The modeling technique is based on energy balance and heat transfer principles. Only three or two parameters need to be fit in the developed model. Catalog or commissioning information is used to estimate the best value of the model parameters by least squares methods. The technique is illustrated for the heat transfer properties of a specific set of chilled water coiling coils. It is shown that the method is robust and gives a good match to the real performances over the entire operating range compared with the existing methods. This model is expected to have wide applications in real-time control and optimization of Heating, Ventilating, and Air-Conditioning (HVAC) systems. |
| 116 Wavelet-Based Solution to Elliptic Two-Point Boundary Value Problems with Non-Periodic Boundary Conditions |
pp. 1161-1169 |
|
|
by : Hira N. Narang, Rajiv K. Nekkanti
Abstract: - The Wavelet solution for boundary-value problems is relatively new and has been mainly restricted to the solutions in data compression, image processing and recently to the solution of differential equations with periodic boundary conditions. This paper is concerned with the wavelet-based Galerkin solution to two-point boundary-value problems involving elliptical problems with non-periodic boundary conditions. The wavelet method can offer several advantages in solving the boundary-value problems than the traditional methods such as Fourier series, Finite Differences and Finite Elements by reducing the computational time near singularities because of its multi-resolution character. In order to demonstrate the wavelet technique to non-periodic boundary value problems, we extended our prior research of solution of parabolic problems to two elliptical problems, the one-dimensional Helmholtz Equation and a two-dimensional elliptical equation. The results of the wavelet solutions are examined and they are found to compare favorably to the exact solutions. This paper on the whole indicates that the wavelet technique is a strong contender for solving two point boundary value problems with non-periodic conditions involving elliptical problems. |
| 122 Simulation Analysis of Scheduling Rules in MEMS Manufacturing |
pp. 1221-1226 |
|
|
by : Lixin Wang
Abstract: - This paper focuses on the production scheduling in MEMS (microelectromechanical system) manufacturing. The whole MEMS production process can be organized as three sub-processes, that is, the front-end process, the wafer cap process and the back-end process. As the total processing time of the front-end process is significantly longer that that of the wafer cap process, the coordination between the releases of wafers to these two sub-processes is an important issue. In this paper, three synchronization rules are developed and they are evaluated together with six dispatching rules under Poisson input. The performance measures considered are cycle times of wafers. A visual interactive simulation model is constructed to imitate the production line. The simulation results indicate that both synchronization rules and dispatching rules have significant impacts on the performance of MEMS manufacturing. |
| 124 Adjective Text Estimation Method: Linking 2d and 3d Space Database |
pp. 1241-1248 |
|
|
by : Tatsuya Shibata,Toshikazu Kato,Motofumi Suzuki
Abstract: - This paper shows an adjective text estimation method to map from still-images and 3d models to adjective texts because adjective texts can be used for different multimedia data to link texts, images, 3d object models, etc. The method automatically obtains adjective scores and labels adjective texts for all the images and 3d models in the database based on the mapping model. To construct the mapping, we have analyzed the relationship between adjectives and graphical parameters in office images. We found that color, edge direction, frequent pattern and position are strongly related to adjectives. The adjective text estimation method is examined by an experiment and we found that the method is valid. |
| 126 Retrieval Image by Region Classification |
pp. 1261-1263 |
|
|
by : Ryszard S. Choras
Abstract: - The retrieval of images from a large database of images is an important and emerging area of research. High-resolution images in the physical database are decomposed into sets of image features which are stored in the logical database. Currently a few image retrieval systems combining color and texture as features to search images. In this paper 2D image is processed using a set of Gabor filter to derive a feature vector representing texture in the image. Color information in an image is represented by color histogram. This method is useful for processing large collections of image data. |
| 129 A New Introduction to Global Optimization over Polyhedrons |
pp. 1291-1296 |
|
|
by : Mojca Indihar Stemberger, Hossein Arsham
Abstract: - For some kinds of linearly constrained optimization problems with unique optimal solution, such as linear and convex problems, the single local optimum is also global. However, there are a broad variety of problems in which the property of unique solution cannot be simply postulated or verified. The paper presents an effective approach for the global linearly constrained optimization problem with continuous objective function. With the help of a parametric representation of the feasible region an equivalent unconstrained problem is constructed which is much easier to solve. Our aim is to propose a new introduction to global optimization, the design of a general solution algorithm that always finds the solution and provides useful information such as bounding of the objective function. The algorithm and its applications are presented in the context of a numerical example. |
| 130 Evaluation Of Alternative E-Business Models By Business Process Simulation Modeling |
pp. 1301-1306 |
|
|
by : Jurij Jaklic, Mojca Indihar Stemberger, Andrej Kovacic, Vesna Bosilj-Vuksic
Abstract: - Business process modeling, simulation, and an evaluation of different alternative scenarios for improvement has been proven to be promising tools for decreasing the risk related to business process reengineering projects. Business renovation as the highest level of strategy for managing change in an organization is the key element of electronic business orientation. Business process simulation is investigated as a tool for deriving an additional in-depth understanding of how the process is executed, an identification of the sources of the problems observed during the process execution, etc. An example of how simulations can be used for understanding a process and evaluation of how the proposed improvements and introduction of electronic business will influence the system performance is presented. |
| 131 Circular dynamic stereo and its image processing |
pp. 1311-1315 |
|
|
by : Kikuhito Kawasue, Yuichiro Oya
Abstract: - A new approach to the measurement of three-dimensional movement of particles is introduced. A single TV camera with an apparatus to rotate the image enables us to record the three-dimensional movement of particles as spiral streaks on a single image. Every shape of the spiral streak on the image plane is related to the position and the velocity of the individual particle. The information about three-dimensional movement of particles is extracted from the image using an image processing technique. We applied the technique to the measurement of three-dimensional water flow field and obtained satisfactory results. |
| 132 Iterative Algorithm for Computing the Eigenvalues |
pp. 1321-1325 |
|
|
by : Liljana Ferbar
Abstract: - We consider the eigenvalue problem H*x=lambda*x, where H is a real symmetric matrix. When eigenvalues of symmetric matrices are computed it is generally expected that they will be computed with an error bound proportional to the product of machine precision and the norm of the matrix. In particular, tiny eigenvalues are usually not computed to high relative accuracy. We propose an iterative refinement of the eigensolution computed by a fast method. Our algorithm usually gives the eigensolution with high relative accuracy and it is more efficient than accurate Jacobi type methods. |
| 134 Educational Hypermedia System ALICE: an Evaluation of Adaptive Features |
pp. 1341-1346 |
|
|
by : Alenka Kavcic, Marko Privosnik, Matija Marolt, Sasa Divjak
Abstract: - This paper presents a Web-based educational hypermedia system, called ALICE, and focuses on the evaluation of its adaptive features. The system builds on the concept domain model and includes the elements of knowledge uncertainty in the process of user modelling. The two models are used for adapting system's navigation support to an individual user; link annotation, direct guidance and link insertion are implemented in the system. The system has been evaluated in a real environment on the course Introduction to Java as a testing teaching domain. The results of the experiments have shown that the presented educational system and its adaptation techniques provide a valuable, easy-to-use tool, which positively affects user knowledge acquisition and thus leads to better learning results. |
| 135 An Approach to Decomposition of Muscle and Nerve Signals |
pp. 1351-1356 |
|
|
by : Damjan Zazula, Eric Plevin
Abstract: - This paper considers the decomposition of surface electromyograms (SEMG) using higher-order statistics (HOS). Modelling surface EMG by a MIMO system whose inputs in the form of innervation pulse trains are considered independent identically distributed (i.i.d.) random white noise, the system identification methods based on higher-order statistics may be introduced. We disclose how a two-phase procedure lead to SEMG decomposition using HOS, firstly, obtaining a coarse estimation of the EMG building blocks, i.e., the motor-unit action potentials (MUAPs) via a simpler and less complex cumulant-based identification, and, secondly, refining the outcomes of the first phase by a more demanding optimisation method. In our experiments on synthetic SEMG signals, we used multivariate C(q,k) in the first phase, and non-linear LMS optimisation of third-order cumulants in the second phase. The simulation results in rather noisy case with 10 dB additive white Gaussian noise prove the robustness and efficiency of the proposed approach. |
| 137 Force and torque measuring devices for drilling and milling simulation |
pp. 1371-1374 |
|
|
by : Antonios Baldoukas, Georgios Demosthenous, Dimitrios Sofianos, Nikolaos Vlachakis, Konstantinos Kondourgiotis
Abstract: - Cutting force measurement during material working processes on machine tools is the reference point for the precise phenomenon and also for the parameters control which affect it. In this paper construction details were presented relating to the two prototyped cutting force measuring instruments adequate for drilling and milling operations. Their low construction cost and their experimental measuring reliability could be adopted as a productive tool in manufacturing processes and used in factories regardless of size. |
| 138 An Effective Signal Processing Procedure for the Detection of Road Damages using Ground Penetrating Radar |
pp. 1381-1386 |
|
|
by : A. Benedetto f. Benedetto M.R. De Blasiis G. Giunta
Abstract: - Pavement damage is actually one of the most crucial problems in roads. Agencies need first to localize the damage, second to identify the causes. Indeed, the rehabilitation can be compromised, if the cause is not removed. The ground penetrating radar (GPR) technique is used by many agencies involved in roads management. In fact, it is nondestructive and is promising for soil characteristics interpretation, such as moisture or density. A preliminary detection and a subsequent classification of the pavement damage, based on an automatic GPR analysis, have been performed and experimentally validated. An optimum detection procedure is performed. It implements the classical Neyman-Pearson radar test, which is based on a constant false alarm rate strategy. All the setting needed by the optimum procedure have been estimated from a training set of measures. The overall performance has been evaluated by looking at the usual receiverÔs operating characteristic (ROC), obtained by processing a different (from the training phase) set of traces. The obtained results evidence that a reasonable performance has been achieved by a suited analysis of GPR images exploiting the spatio-temporal correlation properties of the received signal. Although a generalization is not reliable, this study shows that an automatic GPR-based evaluation of subgrade characteristics is feasible. |
| 144 least-connection algorithm based on variable weight for multimedia transmission |
pp. 1441-1445 |
|
|
by : Yu Shengsheng, Yang Lihui, Lu Song, Zhou Jingli
Abstract: - A least-connection algorithm based on variable weight is presented in this paper in order to satisfy the requirement of multimedia transmission, which is based on the analysis of the existing cluster architecture, algorithms of the load distribution and balancing of network nodes. A validating trial has been performed and the results show that our algorithm has effective load balancing in one central control node scenario. |
| 145 Construction of cooperative behavior in multi-agent systems |
pp. 1451-1453 |
|
|
by : Marko Privosnik, Matija Marolt, Alenka Kavcic, Sasa Divjak
Abstract: - Massive multi-agent systems with emergent global behavior represent a new paradigm in designing and building program systems for handling complex problems. Emergence, nonlinearity and self-organization of global behavior underpin robustness, flexibility and autonomy of massive multi-agent systems. On the other hand, this makes the traditional design approach difficult, if not impossible. In this paper we will focus on the construction of cooperative behavior in reactive multi-agent systems using evolutionary computation. We will present how emergence can be utilized to design global behavior of the system. |
| 146 MATLAB programs for generating orthonormal wavelets |
pp. 1461-1465 |
|
|
by : B.G. Sherlock and Y.P. Kakad
Abstract: - This paper presents MATLAB programs for generating the coefficients of the lowpass analysis filter corresponding to orthonormal wavelet analyses. One of the programs generates the famous Daubechies maxflat wavelets, and a second generates the Daubechies complex symmetric orthonormal wavelets. The remaining two programs generate the space of all orthonormal wavelets in terms of parameterizations whereby the space of wavelets of a given length 2N is generated by N parameters. This software should prove useful where it is desired to perform an optimization to obtain the best wavelet for a given application. |
| 147 Web-based Multimedia Content Management System for Effective News Personalization on Interactive Broadcasting |
pp. 1471-1479 |
|
|
by : S.N.Cheong, Azhar K.M. , M.Hanmandlu
Abstract: - This paper describes the design and implementation of an effective web-based multimedia content management system (MCMS) for organizing, integrating and composing personalized digital news for interactive broadcasting. A 5-layered architecture is proposed for the MCMS and is implemented using the Java 2 Enterprise Edition (J2EE). A prototype based on our framework has been implemented in the News On Demand KIOSK Network (NODKN). This system enables news publisher to manage personalized news that is structured, profiled and streamed to end-users. It examines the user’s profile to identify what kind of news to include in the news program and what kind of styles to provide to the users. The MCMS generates SMIL documents, which are the backbone for digital news broadcasting, using XML and XSLT techniques. The MCMS has three essential features. (1) It allows an effective collaboration among contributors to the content production process. (2) It provides an easy way to produce personalized digital news, and (3) it provides an easy method to create different presentations of the digital news that is catered to users with different needs. |
| 149 A multi-agent model for the estimation of passenger waiting time in public transportation networks |
pp. 1491-1497 |
|
|
by : Berdai Abdelhamid, Gruer Juan Pablo, Hilaire Vincent, Koukam Abderrafiaa
Abstract: - Several efforts have been focused on the improvement of new services of an urban transport network in order to face the increasing of the demand. In this paper, we present a methodology based on a meta-model to design and simulate multi-agent systems. This methodology allows to build a multi-agent model of public transport network in order to estimate the passenger waiting time at connection stops. This will provide the basis for minimization of the waiting time when transferring between buses. |
| 151 A matrix-oriented approach for analysis and optimisation of block digital filters |
pp. 1511-1515 |
|
|
by : Gilles Burel
Abstract: - Block Digital Filtering (BDF) is a well known method for fast digital filtering. Indeed, it decomposes the input signal into blocks and takes profit of computationality reduction provided by the Fast Fourier Transform algorithm to considerably reduce the computational cost. In this paper, we propose a simple and efficient matrix-oriented approach to compute and visualize the frequency response of a BDF, as well as to optimise the BDF coefficients. This approach can be considered as an interesting alternative to the traditional overlap-save method, because it provides better frequency responses. |
| 152 Wireless Adhoc Networks in the Metropolitan/Wide Area: Concepts and First Results |
pp. 1521-1526 |
|
|
by : Arasch Honarbacht, Anton Kummert
Abstract: - This paper is dedicated to on-going research activities in the field of wireless metropolitan and wide area networks with Quality-of-Service (QoS) guarantees. We focus on self-organizing systems, so-called adhoc networks, which can operate without any wired infrastructure (e.g. cellular base stations) - except for some interface points providing interconnectivity to other networks like the Public Switched Telephone Network (PSTN) or the Internet. Obviously, the lack of a wired back-bone network implicates that routing, signalling and other management functions must be handled completely over the air in a self-organizing manner. |
| 153 Hardware implementation for a face recognition algorithm using template matching in frequency domain |
pp. 1531-1535 |
|
|
by : Peter Rauschert, Marco Krips, Anton Kummert
Abstract: - Based on video images, face identification gets more and more interesting to be implemented in new products. By the means of a very large scale integration (VLSI) based solution, this very challenging task can be solved. As target platform a field programmable gate array (FPGA) is used. Thus, the identification cycle time can be reduced by parallel processing. The realisable frame rate is very high and can be used easily for real-time applications. For identification purposes every picture is transformed using a fast fourier transformation (FFT). Working in fourier space offers 2 main advantages. By concentration on most significant frequency bands a conciderable amount of memory can be saved and the system performance will be much better. Finally, identification is done by the use of compressed data and an euclidean classificator. All classification results are associated with confidence probabilities. Special parameters can be adapted in order to affect the required conditions. |
| 154 Statistical Analysis of the Smallest Singular Value in MIMO Transmission Systems |
pp. 1541-1546 |
|
|
by : Gilles Burel
Abstract: - Multi-Input Multi-Output (MIMO) transmission systems is a topic of great interest for a few years due to the huge spectral efficiency gain they can provide over rich scattering transmission channels, such as indoor (e.g. wireless local area networks) or urban outdoor (e.g. mobile wireless communications). MIMO transmission channels are usually modelled by random matrices. In this paper, we use results from random matrices theory to derive the statistics of the smallest singular value of a MIMO channel. Indeed, the smallest singular value is of crucial importance for the performances of the transmission system because it determines the minimum distance between the received vectors. |
| 159 A teaching environment to model and simulate computer processors |
pp. 1591-1596 |
|
|
by : Sebastiano Pizzutilo and Filippo Tangorra
Abstract: - The paper describes a system to design computer processors and to simulate their behaviour during the execution of assembly user programs. The system, called APE (Architecture Prototyping Environment), is based on a dynamic object oriented definition and use of processor components. After the user choice of the architecture components, the system builds a processor simulator allowing users to study the processor behaviour. The iterative process of modelling the computer architecture, prototyping the corresponding simulator and simulating its behaviour, makes the system particularly useful in the activity of teaching computer architectures. |
| 160 A graphical user interface for the initial path generation of a robotic manipulator for an arc welding system |
pp. 1601-1607 |
|
|
by : Theodore Pachidis,John Lygouras, Philippos Tsalidis
Abstract: - In this paper a novel, graphical user interface, is presented. This interface can be used to generate a desired path of the end effector of a robotic manipulator. The path is selected by a static scene of the robot environment, manually either on line using an image capturing system or off line using a stored pair of images or a complex image. On line the system captures a pair of images using a stereovision system with one or two cameras or our pseudo stereovision system. A desired path could be generated by an edge or part of it of the scene image, a line manually designed to the image or a combination of lines of the previous cases. A user can initially process a pair of images selecting from pull down menus a variety of filters, edge detection methods and operations. Then the desired path as a combination of lines is selected from images. Applying our correspondence algorithm, corresponding edges can be found. Finally, a successive number of path points are calculated by means of the stereo system equations, the camera calibration parameters and the hand-eye transformation. In on line operation the capturing system mounted on the end effector can capture images with the desired best view of a scene by moving or rotating, using push buttons, the end effector of the robotic manipulator PUMA 761. Other facilities of the above system are the selection of a variety of colors and shapes, histogram view, magnification, automatic execution of user selected operations and system information. The interface is developed in Visual C++, it runs in a personal computer and communicate with the robot PUMA 761 via ALTER communication port. |
| 161 Pseudo-Articulatory Representations and the Use of Syllable Structure for Speech Recognition |
pp. 1611-1616 |
|
|
by : Li Zhang, William Edmondson
Abstract: - The alternative approach for speech recognition proposed here is based on pseudo-articulatory representations (PARs), which can be described as approximation of distinctive features, and aims to establish a mapping between them and their acoustic specifications. This mapping which is used as the basis for recognition is first done for vowels. It is obtained using multiple regression analysis after all the vowels have been described in terms of phonetic features and an average cepstral vector has been calculated for each of them. Based on this vowel model, the PARs values are calculated for consonants. At this point recognition is performed using a brute search mechanism to derive PAR trajectories. Subsequently we'll show how a model of syllable articulation can be used with PAR trajectories to computationally provide a general articulatory transcription of speech without phonetic labeling. This will form the basis of a speech recognition system. Finally the recovered syllable patterns are used to obtain a phone sequence. The results are very promising taking into account the preliminary nature of the work and the novelty of the approach. |
| 162 Ubiquitous interaction: adapting to the user in context |
pp. 1621-1626 |
|
|
by : Berardina De Carolis and Sebastiano Pizzutilo
Abstract: - Context-awareness is increasing of importance in achieving an effective communication of informa-tion and provision of services. Then, besides taking into account “classical” user-related features, other context-dependent factors (i.e. users’ location, activity, emotional state and technical characteristics of the used device) have to be considered. In this paper, we present the Personalization component of a multiagent infrastructure that we have developed for supporting interaction between users and environments. This component establishes which information to present, how to organize it and how to set up the output layout according to “user in context” fea-tures. |
| 164 Implementation and Computer Simulation of Robot Joint Error Maximum Mutual Compensation |
pp. 1641-1646 |
|
|
by : Yauheni Veryha and Jerzy Kurek
Abstract: - The method of end-effector pose accuracy improvement using joint error mutual compensation for robotic manipulators with rotation joints was presented. The computer simulation and experiments of the developed method showed that it was possible to perform the technological operations with a higher accuracy in the special areas of the robot working space using the developed approach. The method provides the basis for an industrial application of joint error mutual compensation in the conventional robotic manipulators. The practical areas and typical robotic systems where the developed framework of joint error mutual compensation could be applied were presented. |
| 165 An algorithm for corrugated paper cutting |
pp. 1651-1655 |
|
|
by : Damir Kalpic, Vedran Mornar, Kresimir Fertalj
Abstract: - An automated machine cuts the rolls of corrugated paper longitudinally and splits the paper stripe into multiple conveyors, where in each of them a different equidistant lateral cut can be applied. There is a choice of input rolls of infinite length but different widths. The market requirement for large series of different rectangle-shaped articles has to be met. Upper limits for the articles also exist. The minimum material consumption is the objective of optimisation. A recursive function to generate all the possible cutting schemas is written. It provides for formulation of a linear programming model. The minimisation of machine set-up costs cannot be practically solved by binary variables because of the prohibitive problem size. Instead, an iterative navigation around the achieved optimum solution, using the dual activity values is devised. |
| 166 A rational interpolation technique to approximate the time-dependent matrix exponential |
pp. 1661-1666 |
|
|
by : Stephen Stubberud
Abstract: - While the computation of the matrix exponential is known to be dubious, the computed value is used widely in system analysis. This is especially true for the time-dependent matrix exponential. As the systems that we model become more complex, the order of the systems, and thus the order of the matrix exponential become larger. Even with today?s increases in processing power, the computation of numerous values to visualize or analyze the system can tax computational output to point where a degradation in performance can be seen by the user. Often great accuracy is not needed especially in the visualization and interpolations can be used. In this paper, we present a rational interpolation method to estimate the time-dependent matrix exponential that can also estimate the error bound of the interpolation. We clearly note that, while this method utilizes significantly less computational resources than actual computation, it is dependent on the actual computational method chosen for its performance. |
| 167 A variable step-size LMS algorithm |
pp. 1671-1675 |
|
|
by : Bozo Krstajic, Ljubisa Stankovic, Zdravko Uskokovic
Abstract: - The paper proposes a new adaptive VS LMS algorithm, obtained by combining LMS algorithms with different step sizes without calculating their weighting coefficients. As a criterion for choosing the VS LMS algorithm step size, we take the ratio between the weighting coefficients' bias and variance. Identification of an unknown system in nonstationary noisy environment is performed and simulations with the proposed and other VS LMS algorithms are presented. Simulation results confirm the favorable properties of the proposed algorithm in nonstationary environment with abrupt changes of unknown system parameters. |
| 171 Velocity estimation and adaptive clutter filtering for color flow imaging |
pp. 1711-1716 |
|
|
by : Ch. Kargel, G. Hoebenreich, G. Plevnik, B. Trummer, M.F. Insana
Abstract: - Preliminary results are presented for a color-flow imaging system being developed to detect early cancer by imaging tissue elasticity and blood flow simultaneously. The principal challenge is to separate blood velocities from movement of the surrounding tissues even though the corresponding Doppler spectra overlap. This report examines a broad range of digital filter designs using pulse-echo measurements from tissue-like materials. Adaptive eigenfilters are most able to separate the sources of physiological motion. When used with an improved autocorrelation technique (2-D autocorrelator), accurate and precise velocity estimates were obtained for slow pulsatile flow in 3-mm-diameter vessels with a peak velocity of 30 mm/s. |
| 172 Hardware Implementation of the Data Encryption Standard (DES) |
pp. 1721-1724 |
|
|
by : Waiyawut Sanayha, Yuttapong Rangsanseri
Abstract: - This paper presents hardware implementation of the data encryption and decryption of the DES standard algorithm. The design consists of the S-Box circuit Exclusive-OR circuit, Circular Left Shift circuit and the control unit. The S-Box circuit implemented using 32 PLDs (Programmable Logic Device) chip that hold substitution a 48-bits input data and yields a 32-bits output data. The other circuits implemented using hi-speed CMOS integrated circuits and a PC printer port control the whole operation of the DES. The prototype hardware implementation can produce encrypt and decrypt data up to 30 Mbytes/sec. |
| 173 Analysis of Internet Multi-Agent Based System for Zeus and SkeletonAgent Frameworks |
pp. 1731-1736 |
|
|
by : David Camacho,Ricardo Aler,Cesar Castro,Jose M. Molina
Abstract: - Because of the evolution of the World Wide Web, the fields of intelligent agents and multi-agent systems have become a focus of interest. This intensive research has originated several frameworks, or toolkits, that are used by the researchers to develop Internet agent-based systems. Their aim is to help the designers and engineers to build complex systems based on the agent concept. This paper presents a brief description of two of those frameworks: ZEUS and SkeletonAgent. The aim of this paper is to compare them in a common domain: to search for news in several electronic newspapers. Because every one of those multi-agent toolkits has different features, the effort to design and build systems with them, as well as their behavior is expected to be different. The paper will measure the software effort to deploy the first agent and the global multi-agent system for each technology. It also studies the software reusability and the empirical performance evaluation. |
| 175 on the steganography effects in digital images |
pp. 1751-1756 |
|
|
by : Francescomaria Marino, Giuseppe Mastronardi
Abstract: - In this paper the effects of steganography in different image formats (BMP, GIF, JPEG and DWT-coded) are studied. With respect to these formats, we try to give an answer to the following questions: “how many bits of noise (i.e. the textual secret message) can be injected without perceptually deteriorating the quality of the image?” and “how and where to inject these bits in order to achieve the best trade-off in terms of length of the textual message and preserved quality of the image?”. |
| 176 A Selective Attention Based Method for Target Detection |
pp. 1761-1766 |
|
|
by : Kyungjoo Cheoi,Yillbyung Lee
Abstract: - This paper describes a new selective attention based method for wide range of target detection tasks, without using any a-priori knowledge about the target. In proposed system, several basic features are extracted directly from visual stimuli, and these features are integrated based on their local competitive relations and statistical information. Through integration process, unnecessary features for detecting the target are spontaneously decreased, while useful features are enhanced. The performance of the system is evaluated over various synthetic images which were typically presented in psychophysical studies of visual search, and complex real images of natural environment taken from different domains, including noisy images. |
| 177 Hidden Markov Models for Greek Sign Language Recognition |
pp. 1771-1776 |
|
|
by : Vassilia Pashaloudi, Konstantinos Margaritis
Abstract: - Sign languages are the basic means of communication between hearing impaired people. A translator is usually needed when a deaf person wants to communicate with persons that do not speak sign language. The work presented in this paper aims at developing a system that could recognize and translate isolated and continuous Greek Sign Language (GSL) sentences. Input is obtained by using a feature extracting method from two-dimensional images. The feature vectors produced by the method are then presented as input to Hidden Markov Models (HMMs) for recognition. We use a vocabulary of 26 greek words (nouns, pronouns, adjectives, verbs) and achieve a recognition rate of 98% for isolated recognition and 85.7% for continuous recognition. |
| 178 Adaptive Colour Segmentation in Digital Video Images by means of Artificial Neural Networks and Morphological Operations |
pp. 1781-1786 |
|
|
by : Marco Krips, Anton Kummert
Abstract: - There are many industrial products that use digital video image processing for orientation control or object detection. Often object shape is used for classification. However, if detection task has to be performed very fast and there are many different candidate shapes for an object, other characteristic features have to be used. One of the most relevant object features can be colour. For example, the detection of human hands is a challenging problem, where a shape-based approach is unpromising, if detection has to be done in real time. The presented approach proves that a robust detection can be obtained by applying a neural network to the RGB values of digital video images and subsequent use of morphological operations. Furthermore, the claim for very small processing times is considered in the presented solution. |
| 179 Elements of an object-oriented FEM program for multibody systems analysis |
pp. 1791-1797 |
|
|
by : Kromer Valerie,Dufosse Francois,Gueury Michel
Abstract: - The principal objective of this paper is to give elements of the object-oriented architecture of a finite element code dedicated to multibody systems analysis. Emphasis is placed on the adequacy between object-oriented programming and the finite element method used for the treatment of three-dimensional multibody flexible mechanisms with large rotations and large strains. Numerical examples are given in order to demonstrate the software capabilities. |
| 180 Analysis of foundation-layered soil interaction using propagating wave-mode analysis |
pp. 1801-1806 |
|
|
by : Igor Spacapan, Miroslav Premrov
Abstract: - In the engineering analysis of the response of the foundation in layered soil, subjected to direct or seismic excitation, parametric analysis is the most desirable type of analysis because it has potentials for better optimal design. We are presenting a computational approach, which yields dynamic displacements of a foundation and wave motion in layered soil. The computational approach yields wave-modes and their amplitudes as parameters, and the influence of each wave mode on the vibration of the foundation and on the spreading of waves into the surroundings. Computation is accomplished in the frequency domain. It uses the finite element method where the radiating conditions on the fictive boundary are satisfied exactly. We are presenting a brief outline of the key formulas of the computational approach. Numerical results are partially compared to the exact ones, suggesting the efficiency of the approach. |
| 181 Robot Pageantry: Issues in Assessing Performance of Social Robots |
pp. 1811-1816 |
|
|
by : Brian Duffy, Gina.Joue
Abstract: - How can the performance of a social robot be measured? Are existing metrics and techniques
appropriate for physically and socially embodied autonomous mobile robots? With the rapid expansion of
robust robotic systems from strong industrial applications to domestic and entertainment domains which
naturally involve social interactions, a requirement to gauge a system’s abilities either individually or
relatively has emerged. This paper discusses the implications that arise in aiming to assess the performance of
robots engaged in a degree of social interaction with people. |
| 182 Statistical Surface Analysis as a High Speed Process |
pp. 1821-1826 |
|
|
by : Arnulf Schiller, Paul O'Leary
Abstract: - In the course of an industrial research project a high-speed optical surface defect detection system was developed. The material to inspect is piston ring wire. Main condition was to continuously inspect the whole surface without slowing down the production speed of 0.5 to 1.5 meters per second. The minimum defect size was defined by 100 mm extension in one dimension at wire perimeters up to 40 millimeters. This leads to an amount of up to 120 megabyte of image data to acquire and process in a second. To meet this requirements on one hand fast hardware components (cameras, frame grabber) had to be utilized for grabbing the image data. On the other hand high-speed image processing algorithms had to be developed to compute features which can indicate defects out of the image data. The developed image processing algorithm is based on the computation of higher statistical moments, which turned out to be well suitable to distinguish between the gaussian distribution of the image data from faultless surface and disturbed statistical distributions in the presence of defects. The adaptation of this concept to line scan image data and the implementation on on-board processors necessitates some modifications, which result in increasing processing speed, better localization of the defects and higher sensitivity. |
| 185 Massively Parallel Priority Queue for High-Speed Switches and Routers |
pp. 1851-1856 |
|
|
by : Kari Seppanen
Abstract: - The need for better quality of service is growing as new quality sensitive services are becoming more and more important in data networks. The key element in providing quality-of-service or grade-of-service in packet networks is deciding when or in what order the incoming packets should depart the system. In other words the departing packets have to be somehow scheduled. Implementing this scheduling becomes harder and harder as line-speeds and number items to schedule increases. One way to implement the scheduling function is to use a priority queue that can be realised by using simple and scalable heap data structure. The problem is that traditional sequential heap cannot be made fast enough. In this paper a massively parallel system based on set of heaps is presented. It will be shown that such system can be scaled to support very large priority queues at STM-256 speeds and beyond. |
| 186 Comparisons of meta-heuristics for the layered digragh drawing problem |
pp. 1861-1865 |
|
|
by : Bruno Pinaud, Pascale Kuntz, Remi Lehn
Abstract: - In this paper we present a Genetic Algorithm (GA) with problem-specific operators for the layered digraph drawing problem and compare it with two very different meta-heuristics: Tabu search and Multi-start descents. Tabu search has previously proved to be better than the classical deterministic local heuristic often employed in graph drawing. Here, we show on a set of 282 medium-sized graphs that GAs lead to similar results to Tabu for graphs of relatively high density and better results for graphs with medium density. Comparisons with multi-start descents tend to confirm the good quality of the solutions found by using GAs. |
| 187 Roof Types Visualization for Human Habitats Indicator |
pp. 1871-1876 |
|
|
by : Minoru Ueda
Abstract: - The objective of this paper is to show that the visualization of an urban landscape by 3D computer graphics can be used as indicators of human habitats on a worldwide basis when we think of a sustainable world in future. In the 1900s, 2D digital urban maps became popular as commercial products such as automobile navigation map with GPS in Japan. They adopted the traditional map making technique consisting of map symbols and polylines. On the other hand, 3D CG urban map making has not yet established agreed principle or guidlines how to create. The simplest methods is to create a virtual house and a building as they are, using 3D CG architectural CAD system. This method demands enourmous amount of labor so that we can not create an urban map with thousands of houses and buildings. The second method is to express houses and buildings as boxes vertically projected on the closed polygon representing their building site. This method can show a geographical distribution of building sizes only. The database volume of this second method becomes ten times larger than the counterpart 2D digital map of the same area. The auther proposes the third method. The actual buildings and houses have different each other but there are something in common such as roof shapes. In the world, we know there are about 30 roof types. Assuming that a house or a building has one roof and prepare 30 different prototype building models. We call them as 3D template analogous to library parts used in CAD systems. A 3D CG urban map created by the third method has an advantage over the first and the second method, not only from the database size viewpoint but also indicators of human habitats in the world. Shortly, (1) flat roof dominates in dry region. (2) cone shape roof in sem-dry region such as some part of Africa. (3) short eaves gable roof is dominat in humid region such as Western Europe and North America. (4) deap eaves and Asian mansard roofs are dominant in Monsoon Asia. (5) a modern high rise expressed by a tall box. This proposed method needs as large database volume as one by the second method but can show the cultural differences of cities. The auther describes the usability of the third method as a tool for environmental planning. |
| 189 Calibration Method for Light Sectioning Measurement Systems |
pp. 1891-1898 |
|
|
by : Ewald Fauster, Peter Schalk, Mark Tratnig
Abstract: - This paper presents a new approach in calibrating a light sectioning measurement system. The developed method permits the calibration of such a system without previous knowledge of the exact positions of camera and laser. The basic concepts of this new approach and its advantages compared to common methods are discussed. Further, a possible implementation in an industrial process is given. |
| 190 Recognition and rejection performance in wordspotting systems using support vector machines |
pp. 1901-1906 |
|
|
by : Yassine Ben ayed, Dominique Fohr, Jean Paul Haton, Gerard Chollet
Abstract: - Support Vector Machines (SVM) is one such machine learning technique that learns the decision surface through a process of discrimination and has a good generalization capacity. SVMs have been proven to be successful classifiers on several classical pattern recogntion problems. In this paper, one of the first applications of Support Vector Machines (SVM) technique for the problem of keyword spotting is presented. It classifies the correct and the incorrect keywords by using linear and Radial Basis Function kernels. This is a first work proposed to use SVM in keyword spotting in order to improve recognition and rejection accuracy. The obtained results are very promising. The Equal Error Rate (EER) for the linear kernel is about 16,34% compared to 15,23% obtained by the radial basis function kernel. |
| 191 A new concept of numerical object on partially solving method (PSM) for a system of linear equations |
pp. 1911-1916 |
|
|
by : Minetada Osano and Mitumori Tanimoto
Abstract: - System of linear equations appear in a wide range of application fields. The Partially Solving Method(PSM) which was proposed with us is one of the methods of solving a linear equation. So far, PSM can be solved without require knowledge of the entire system at any time. In this paper we introduce a decomposition approach that significantly enhances the efficiency of the PSM. In this decomposition approach of system analysis, many PSSs ( Partially Solving System) are generated at arbitrary place simultaneously, in random or as parallel processes without knowledge of the total system, and are merged using an assembling approach. and a final solution is achieved at last process of merging of PSSs. Next, we proposed a new concept of numerical object created with each PSSs as same as Object-Oriented Paradigm on software science. Finally, the numerical objects have unsolved variables as boundary conditions. The merging process means to extend the boundary conditions on merged numerical objects. |
| 192 Significance of non-linearity and component-internal vibrations in an exhaust system |
pp. 1921-1926 |
|
|
by : Thomas Englund, Johan Wall, Kjell Ahlin, Goran Broman
Abstract: - To facilitate overall lay-out optimisation inexpensive dynamics simulation of automobile exhaust systems is desired. Identification of possible non-linearity as well as finding simplified component models is then important. A flexible joint is used between the manifold and the catalyst to allow for the motion of the engine and to reduce the transmission of vibrations to the rest of the exhaust system. This joint is significantly non-linear due to internal friction, which makes some kind of non-linear analysis necessary for the complete exhaust system. To investigate the significance of non-linearity and internal vibrations of other components a theoretical and experimental modal analysis of the part of a typical exhaust system that is downstream the flexible joint is performed. It is shown that non-linearity in this part is negligible. It is also shown that shell vibrations of the catalyst and mufflers as well as ovalling of the pipes are negligible in the frequency interval of interest. The results implies, for further dynamics studies, that the complete system could be idealised into a linear sub-system that is excited via the non-linear flexible joint, that the pipes could be modelled with beam elements and that the other components within the linear sub-system could also be modelled in a simplified way. Such simplified component models are suggested. The agreement between theoretical and experimental results is very good, which indicates the validity of the simplified modelling. |
| 193 Automated updating of simplified component models for exhaust system dynamics simulations |
pp. 1931-1936 |
|
|
by : Thomas Englund, Johan Wall, Kjell Ahlin, Goran Broman
Abstract: - To facilitate overall lay-out optimisation simplified component models for dynamics simulations of automobile exhaust systems are desired. Such optimisation could otherwise be computationally expensive, especially when non-linear analyses are necessary. Suggestions of simplified models of the mufflers and the catalyst are given. To account for the flexibility at the connections between those components and the pipes short beam elements with individual properties are introduced at these locations. An automated updating procedure is developed to determine the properties of these beam elements. Results from an experimental modal analysis are used as the reference. The theoretical model of the exhaust system is built in the finite element software ABAQUS. The updating procedure uses the sequential quadratic programming algorithm included in the Optimization Toolbox of the software MATLAB to minimise the sum of the differences between experimentally and theoretically obtained natural frequencies. Constraints are used on the correlation between the experimentally and theoretically obtained mode shapes by considering the MAC-matrix. Communication between the two software packages is established by an in-house MATLAB script. The correlation between results from the updated theoretical model and the experimental results is very good, which indicates that the updating procedure works well. |
| 194 Secure Video-on-Demand Server Project: Requirements and Solutions |
pp. 1941-1946 |
|
|
by : Josep Pegueroles Valles and Francisco Rico-Novella
Abstract: - When cryptographic techniques are added to advanced network services such as Multimedia-on-demand many aspects have to be considered. Especially, when commercial services are being offered, aspects related to secure payments, server fraud protection and client rights have to be studied. The Security Module in SSADE (Secure System for the Access and Efficient Distribution of Multimedia-on-Demand Services - CICYT TEL99-0822) has centered his investigations in two main items: Secure and Anonymous Payment and Ciphering Techniques for protecting data from eavesdroppers. This paper will focus in the solutions adopted in this project in order to achieve secure access and payment to the Multimedia-on-Demand service and privacy for content distribution. |
| 195 Block Diagram Model of Lathe Machine |
pp. 1951-1956 |
|
|
by : M. Ebrahimi, W. Moughith and J. Victory
Abstract: - This paper presents a computer aided method for the analyses and simulation of a non-linear mathematical model of a CNC Lathe machine tool. The derived model consist of the controller, axis drive and the spindle drive of a CNC turning machine and their connection through the cutting process. The overall model uses a modular multi-model approach to prototype the machine tool and its process. |
| 196 Application of Entropy-Constrained Vector Quantization to Subband Images |
pp. 1961-1965 |
|
|
by : S. Hechaichi , E. Ademovic and A. Merigot
Abstract: - Vector quantization is a powerful way to compress images, rarely used in present compression schemes. We present here the use of this technique on subband images, by means on an entropy contrained quantization algorithm. The codebooks are predetermined on a set of images to speed up transmission. The outline of the method is presented, as well as some experimental results. |
| 198 Signal pre-pocessing and speech quality validation: xm2vtsdb testing |
pp. 1981-1985 |
|
|
by : J.Jorda, E. Bailly-Baliere, B. Ruiz, A. Garcia Crespo O. Monterrubio
Abstract: - In this paper we give a detailed description of the progress achieved in the techniques for feature extraction, signal pre-processing, parametric representation and robustness against noise necessary to implement a speaker verification system useful in a real environment. We also present the completly experimentation environment and the results obtained. |
| 199 Real-time regularized iterative enhancement of low-resolution video |
pp. 1991-1996 |
|
|
by : Sungjin Kim, Joonki Paik, Jae Woo Kim, and Jong Ho Paik
Abstract: - Regularized iterative image restoration is proven to be a successful technique in restoring degraded images. However, its application is limited to still images or off-line video enhancement due to its slow convergence. In order to enable this iterative restoration algorithm enhance the quality of video in real-time, each frame of video is considered as the constant input and the processed previous frame is considered as the previous iterative solution. Each frame of a video is segmented into two regions: still background and moving objects. These two regions are processed differently by using a segmentation-based spatially adaptive restoration and a background generation algorithms. The proposed framework enables real-time video enhancement at the cost of image quality only in the moving object area of dynamic shots, which is relatively insensitive to the human visual system. |
| 200 Simulation-based Comparisons of Reno, Vegas and Westwood+ TCP |
pp. 2001-2006 |
|
|
by : Luigi Alfredo Grieco, Saverio Mascolo
Abstract: - During the last 20 years, several congestion control algorithms have been proposed to achieve network stability, fair bandwidth allocation and high resource utilization. This paper aims at comparing three well-known control schemes that are Reno, Vegas and Westwood+ TCP. To the purpose, an extensive set of ns-2 simulation results has been collected. In particular, single and multi bottleneck scenarios with link capacities ranging from 1Mbps to 100Mbps and in the presence of homogeneous and heterogeneous traffic sources have been considered. The following main results have been found: (1) Westwood+ TCP fairly behaves when interacting with Reno TCP; (2) Westwood+ TCP improves the fairness in bandwidth sharing with respect to Reno TCP; (3) Vegas TCP is not able to grab its own bandwidth share when interacting with Reno TCP. |
| 203 Formalized symbolic execution |
pp. 2031-2033 |
|
|
by : Zafirios Karaiskos, Constantine Lazos
Abstract: - Formalized Symbolic Execution Zafirios Karaiskos Constantine Lazos Department of Informatics Aristotle University of Thessaloniki 55006 Thessaloniki GREECE clazos@csd.auth.gr Abstract: In this paper we present a semantics-directed description of symbolic-execution of programs. In the field of program analysis, most of the methods have been applied on optimizing compilers. It is our belief that there is enough theory that could support the application of program analysis methods on software construction. Embedded and safety-critical systems but also large and complex software demand sophisticated tools for design, testing and maintenance of software. Most program analysis methods are syntax-directed, thus, they should base the description of the analysis on some form of semantics, which is syntax-directed, such as the denotational semantics. The description, then, should be augmented with information specific to the method, which finally would be implemented by a tool. Tools built from semantic descriptions of programming language’s constructs serve two purposes: make the semantics useful and popular. Symbolic execution is a program analysis method used to reconstruct logic and computations along a program path by executing the path with symbolic, rather than actual values of data. The results obtained by symbolic execution are the symbolic formulae, representing the values of the program variables, and the path condition for each path examined. The symbolic execution approaches, traditionally act upon a control flow graph in which nodes represent program segments, which are to be executed linearly, and the only control flow structures are (conditional) branches represented by arcs. The control flow graph is normally structured by using an intermediate representation of the program, such as triples or quadtruples. This is a “low level” representation of the program. We follow the syntax-directed approach, which is considered to be “high level”. It acts upon a program representation, called abstract syntax tree (AST), which includes all of the high-level control structures present in the source program. First, we define the abstract syntax for an example language. Then we give a semantic description of symbolic execution, augmented with information specific to the method, in order to be able to maintain the symbolic formulae and path conditions. |
| 205 Regenerative estimation variants of response times in closed networks of queues |
pp. 2051-2056 |
|
|
by : Panajotis Katsaros and Constantine Lazos
Abstract: - In this paper, we present a comparison of the possible regenerative estimation variants of response times, in multivariate simulations of closed queuing networks. The underlying stochastic framework of the techniques under study is first described and the applicability of each one of them is discussed. An appropriate sequential control procedure has been selected, in order to produce confidence intervals of the same nominal level and similar width, for the response times of interest. The first experimental results exhibit improved coverage of the corresponding analytic solutions, when a marked job based method is used, instead of an indirect estimation, by simulating a single regeneration sequence of the common number-in-queue process (usually used when estimating other characteristics like throughputs, utilizations, queue lengths etc). This finding clearly implies the simultaneous use of more than one regeneration sequence for multivariate studies that include response time characteristics. |
| 206 Joint Source and Channel Coding with Multiple Description Codes |
pp. 2061-2066 |
|
|
by : Holger Hutzelmann, Robert Klinski, Rudi Knorr
Abstract: - Multiple description (MD) codes have been proposed as a mechanism to provide robustness against loss of data in the context of packet based networks. Therefore correlation is inserted between the descriptions produced by the MD code. In this paper the focus is on the transmission of multiple descriptions over noisy channels. We investigate the usability of the correlation between the descriptions to combat channel impairments. A MD transform coder (MDTC) is combined with an a priori/a posteriori soft-output Viterbi algorithm (APRI-SOVA) which uses the correlation generated by the MDTC for better decoding. We demonstrate that the correlation between the descriptions can be used for improved decoding at the receiver. Hence multiple description codes may also be viewed as joint source channel codes. |
| 207 A new numerical approach to calculate the eigensystem of a matrix that appear in the problem of canonical correlation |
pp. 2071-2079 |
|
|
by : Raimundo Sampaio, Jorge Leludak, Marco Candido
Abstract: - In several many situations we need to calculate the eigenvalues and eigenvectors of a matrix which is given as a product of many others matrices, that is, we need to calculate the eigenvalue and eigenvectors of a matrix M which is given as $M=m_{1}m_{2}...m_{k}$. Even without considering the complexity of the multiplication of matrices, many times $M$ become very ill-conditioning as soon as eigenvector is the concern. Actually this happens even when $M$ is a symmetric matrix, which will be the our only concern here. |
| 216 Auditory Based Feature Vectors for Speech Recognition Systems |
pp. 2161-2166 |
|
|
by : Waleed H. Abdulla
Abstract: - Signal processing front end for extracting the feature set is an important stage in any speech recognition system. The optimum feature set is still not yet decided though the vast efforts of researchers. There are many types of features, which are derived differently and have good impact on the recognition rate. This paper presents one more successful technique to extract the feature set from a speech signal, which can be used in speech recognition systems. Our technique based on the human auditory system characteristics. It relies on the Gammatone filterbank to emulate the cochlea frequency resolution. Compared to the standard mel frequency cepstral coefficients and the perceptual linear prediction analysis front ends, our auditory based front-end yielded higher recognition rate when embedded in a continuous hidden Markov model based automatic speech recognition (ASR) system. Also our feature set has better classification characteristics when considering the F-ratio as a figure of merit. |
| 220 Attractor reconstruction for chaotic epileptic signals |
pp. 2201-2206 |
|
|
by : H. G Gonzalez-Hernandez, M. Montiel-Castellanos
Abstract: - Some techniques for reconstructing attractors from time series are shown in this paper. First, the time delay for obtaining the extra coordinates used for the reconstruction is selected using the Average Mutual Information (AMI); second, the embedding dimension of the attractor is obtained determining the False Nearest Neighbours (FNN). An important feature of this reconstruction algorithm is that it only needs one state variable measurement for reconstructing the attractor. Finally, some attractor reconstructions are shown for two different cases: Chua's circuit and certain epileptic data measured during a seizure, which shows random-like behaviour. |
| 221 Using Previous Knowledge for Stock Market Prediction Based on Fundamentalist Analysis with Fuzzy-Neural Networks |
pp. 2211-2216 |
|
|
by : Renato de C. T. Raposo, Adriano J. De O. Cruz, Sueli Mendes
Abstract: - The purposes of this paper are: i) show how to use the previous knowledge of any economic analyst and represent this knowledge using frames [7]; ii) discuss the application of a combination of Neural Networks and Fuzzy Logic to predict the evolution of stock prices of Brazilian companies traded on the Sao Paulo Stock Exchange; iii) present the obtained results. The network indicates if a trader would have to keep, sell or buy a stock using a combination of information extracted from balance sheets (released every three months) and market indicators. The results show that the network combining the previous knowledge of the economic analyst deliver good results depending on the quality of the available data and others factors. |
| 223 Text Categorization Approach For Chat Room Monitoring |
pp. 2231-2236 |
|
|
by : Eiman Elnahrawy
Abstract: - The Internet has been utilized in several real life aspects such as online searching, and chatting. On the other hand, the Internet has been misused in communication of crime related matters. Monitoring of such communication would aid in crime detection or even crime prevention. This paper presents a text categorization approach for automatic monitoring of chat conversations since the current monitoring techniques are basically manual, which is tedious, costly, and time consuming. This paper presents the results of a cross method comparison between the Naive Bayes, the K-nearest neighbor, and the Support Vector Machine classifiers. The objective is to determine the most suitable method for the data of chat conversations that would automate the chat room monitoring task. |
| 224 Design and simulation of adaptive measurement-based admission control algorithms for controlled-load service |
pp. 2241-2248 |
|
|
by : Kwan Yong Sim
Abstract: - The purpose of call admission control in Controlled-load Service network is to offer a Quality of Service (QoS) commitment to guarantee that QoS bounds are not violated. The traditional way of achieving this goal by declaring the worse case traffic descriptors for the incoming calls results in poor bandwidth utilization. In recent years, measurement-based admission control has become an appealing alternative to improve bandwidth utilization. It exploits the statistical gain and offers adaptivity to changing traffic condition. The measurement-based admission control first proposed by Jamin et. al in [1] was then improved by Casetti et. al in [2] with an adaptive measurement window in the algorithm. Building on the work of [2], this paper examines the algorithm design and simulation of adaptive sampling time in the algorithm. The proposed algorithms are tested through simulations under difference traffic scenarios and proven to produce higher level of utilization without violating the delay-based QoS guarantees. The simulation results are compared with the algorithm described in [2]. |
| 228 A systematic approach to cleaning fiscal data in Local Public Administration |
pp. 2281-2285 |
|
|
by : Mario Bochicchio, Antonella Longo
Abstract: - - In the paper we present a systematic approach in managing fiscal data in Local Public Administration. The novel way has been developed to deal with the “avalanche effect”, due to the mismatch of fiscal data in adjacent years, and other similar problems, mainly related to data dirtiness and to the lack of standardization in fiscal data. The application of the approach is discussed referring to the preliminary results of the Taviano Project, a real case in the Southeast of Italy |
| 229 Results Merging in a Distributed Search System |
pp. 2291-2296 |
|
|
by : Vitaliy Kluev
Abstract: - Mechanisms of results merging in a distributed search system have been discussed in this paper. Such mechanisms’ aims are to select the most relevant documents retrieved by different servers and put them on the top of the list returned to the end user. In our calculation, we use the clustering technique to divide retrieved results into several homogeneous groups and a metric on the base of the vector space model to arrange items inside each group. The main thrust of our approach is both the proposed clustering metric and the ranking metric. Our mechanisms have been implemented in the OASIS system, a system which belongs to the distributed search systems in the Internet. |
| 230 Blind Channel Estimation in Presence of Carrier Offsets for DS/CDMA |
pp. 2301-2306 |
|
|
by : Serguei Burykh, Karim Abed-Meraim, A. Rahim Leyman
Abstract: - Most blind channel estimation methods for Direct-Sequence Code-Division Multiple-Access (DS/CDMA) systems does not take into account the carrier offset resulting from imperfect carrier recovery. The method presented here estimates jointly the carrier offset and propagation channel thus allowing for the carrier offset compensation. Our technique combines the subspace decomposition with a Jacobi-like Constant Modulus Algorithm (CMA) and exploits the fact that CM criterion is invariant to the presence of residual carriers. Channel identifiability is proved for the case when different users have different carrier offsets. Numerical studies illustrate the efficiency of the proposed algorithm. |
| 231 Automatic Speech Recognition In Noisy Environments Using Wavelet Transform |
pp. 2311-2316 |
|
|
by : Weaam Alkhaldi, Waleed Fakhr and Nadder Hamdy
Abstract: - The performance of speech recognition systems is mainly determined by the used acoustic feature extraction technique. Two techniques are known, namely the full-band approach and the multi-band approach using filter banks. Systems using either approach usually suffer from performance degradation in the presence of noise. In this paper, the multi-band approach using Wavelet transform is suggested for speaker-independent isolated word recognition in noisy environments. Moreover, it has been found that combining the acoustic features obtained using both the full-band approach and the Wavelet transform- based multi-band approach has led to an improvement in the achievable recognition rates especially under mismatched conditions at low signal-to-noise ratio situations. |
| 233 An Agent-based Simulation for Modeling Intelligent Munitions |
pp. 2331-2336 |
|
|
by : Karl Altenburg, Joseph Schlecht, Kendall Nygard
Abstract: - An agent-based emergent intelligence methodology is utilized to assemble agents into teams and assign distributed tasks to them in a dynamic environment. The application concerns the management of military missions that utilize intelligent munitions that are capable of searching, detecting, identifying, and attacking targets in a battlefield. The simulator follows a reactive, behavior-based design philosophy. Agents are implemented as asynchronous threads that employ polymorphic signal-based communication. Results to date indicate that the simulation framework is effective in evaluating ways for emergent teams to quickly solve dynamic teaming and task allocation problems. |
| 234 Dynamic Multi-optimal Learning Rates For Neural Network |
pp. 2341-2343 |
|
|
by : Han-Leih Liu, Seng Kah Phooi
Abstract: - This paper presents a method called dynamic multi-optimal learning rates for neural network (NN) with backpropagation (BP) training. The stability analysis of the learning rates for a 3-layer NN to minimize the total square error is included. The optimal learning rates can be obtained by using proper numerical method. These optimal learning rates are then applied to BP training to tune the corresponding weighting factors in each layer so that the total square error is minimized rapidly. A simulation example is performed for a nonlinear first-order system identification. The results have indicated that the training or convergence speed is faster compared with the standard BP with fixed learning rates. |
| 236 Adaptive Pixel Difference Classification, an Efficient and Cost Effective Algorithm for Motion Estimation |
pp. 2361-2363 |
|
|
by : Hamid-reza pourreza, Mohammad rahmati, Farid Behazin
Abstract: - In this paper we present a simple adaptive block-matching algorithm for video compression. Our method is an adaptive approach for pixel difference classification (PDC) matching criterion. A constant threshold is used in original PDC algorithm. The optimum threshold varies from sequence to sequence. In contrast to the original PDC algorithm, adaptive PDC (APDC) employs a time variant threshold, where the threshold is obtained from parametric classification of each frame. Proper threshold obtained for each frame is used for subsequent frame. Experimental results shown that this method can yield further 99% of optimum performance. |
| 237 Pitch Estimation Using Music Algorithm Based on the Sinusoidal Speech Model |
pp. 2371-2374 |
|
|
by : Mehdi Hosseinpour , Hamidreza Amindavar
Abstract: - In this paper we present a new method of pitch estimation in speech signals. The existing methods in pitch estimation do not provide an accurate estimate of the pitch value, in presence of sinusoidal noise.Thus these methods have large errors in presence of sinusoidal noise and some other noise characteristics. The method we presented in this paper is based upon the MUSIC algorithm has been widely used to estimate the frequency and angle of arrivals for narrow(wide) band signals. We estimates the pitch frequency, in the presence of sinusoidal noise and like other methods in the precense of Gaussian noise. |
| 238 Software for Synthesis of Radiation Patterns by Linear Antenna Arrays |
pp. 2381-2386 |
|
|
by : Jose Luis Ramos Q., Martin J. Martinez S., M. Susana Ruiz P.
Abstract: - This paper presents the characteristics of a software developed on MatLab Scripts to obtain the synthesis of radiation patterns, using linear antenna arrays. Several classical techniques have been considered such as Schelkunoff method, Fourier series and transform methods, Woodward-Lawson method, Taylor line-source methods. In all those cases, algorithms were developed and the results agree with the literature. In order to improve some characteristics of the radiation pattern, an algorithm based on the gradient method has been included for optimization purposes. |
| 243 Lyapunov Theory-based Fuzzy Neural Network With MOGA And Its Application To Nonlinear Time Series Prediction |
pp. 2431-2433 |
|
|
by : Seng Kah Phooi, Kai Ming Tse, Zhihong Man, H.R.Wu
Abstract: - In this paper the authors present a Lyapunov theory-based FNN with MOGA for application to nonlinear time series prediction. The architecture employs FNN structure and the learning algorithms are the combination of MOGA and LAF. The application considered is the nonlinear time series prediction. Simulation results are obtained using the MATLAB for the nonlinear sunspot data prediction. The work not only demonstrates the advantage of the neurofuzzy approach but it also highlights the advantages of the fusion of MOGA and LAF. |
| 244 Synchronization of Video Streams in the Implementation of Web-Based E-Learning Courses |
pp. 2441-2446 |
|
|
by : Karl Kurbel, Alexei Pakhomov
Abstract: - E-learning courses based on video-streaming technology are a promising form of transferring lectures to the Internet. Though not applicable to all types of courses, lecturing is a well-tried educational mode suitable in many cases and proven for transporting content. In this paper an approach is discussed in which course material is divided into separate parts ƒ{ one silent video file containing visual demonstrations, and another video with sound, with the instructor's explanations. We discuss approaches to achieve synchronization among two separate videos streams. Based on Microsoft Windows Media technology this problem can be solved but the solution is not straightforward. The solution is based on integration of two Windows Media Player ActiveX controls in one HTML page. Problems encountered and the final solutions to those problems are outlined. |
| 245 An overview of major satellite systems |
pp. 2451-2455 |
|
|
by : Nikos Panagiotarakis, Ilias Maglogiannis, George Kormentzas
Abstract: - A satellite communication system is distinguished by its global coverage, inherent broadcast capability, ability to support mobility and bandwidth-on-demand flexibility. Taking into account this emerging communication role of the satellite systems, the paper provides an overview of major narrowband GEO, MEO and LEO satellite systems. It presents some typical commercial examples of such satellite systems and discusses their basic features consisting of used frequency bands, supported applications and terminals and critical performance issues. |
| 246 Pareto Neural Model for Finding Task Allocations |
pp. 2461-2466 |
|
|
by : Jerzy Balicki, Zygmunt Kitowski
Abstract: - In this paper, the Hopfield model of artificial neural networks called HANNs for finding some software task allocations in multiple computer systems have been proposed. A multiobjective optimisation problem with two criteria has been considered. Resource constraints have been assumed. Both the cost of parallel program execution and the cost of computers have been minimised. Two models of neural networks for minimisation of the computer cost and for minimisation of the cost of parallel program execution have been proposed. HANN for finding local Pareto-optimal solutions has been considered. Some simulation results related to minimisation of the energy function for constructed neural networks have been included. A trajectory of energy function obtained during finding Pareto-optimal task allocation has been presented |
| 252 Dynamic modelling and simulation of a flexible rectangular isotropic plate structure using finite difference methods |
pp. 2521-2526 |
|
|
by : Intan Z. Mat Darus, M. Osman Tokhi
Abstract: - An investigation into the dynamic characterisation of a rectangular flexible plate structure is presented and a simulation algorithm characterising the behaviour of the plate is developed using finite difference methods. A thin, rectangular plate, with all edges clamped, is considered. The investigation is accomplished by varying the width over the length ratio of the plate. The algorithm is implemented within the Matlab environment and allows application and sensing of a disturbance signal at any mesh point on the plate. Such a provision is desirable for the design and development of active vibration control techniques for the system. The performance of the developed algorithm shows that it converges faster than previously reported work. The simulation algorithm thus developed and validated forms a suitable test and verification platform for development of active vibration control strategies for flexible rectangular plate structures of various width over length ratios. |
| 253 The effects of spatial correlation and the LOS component on the capacity of broadband MIMO channels |
pp. 2531-2537 |
|
|
by : Pedro Tejera, Holger Hutzelmann, Rudi Knorr
Abstract: - The effects of spatial correlation and LOS (Line of Sight) component on capacity of broadband MIMO channels are investigated. For this purpose a flexible, parameterised model for the MIMO channel is proposed. The capacity gain that can be achieved with the use of CSI (Channel State Information) on the transmitter side is examined for a variety of different channel parameters. Channel knowledge at the transmitter is difficult to attain and will normally require some kind of feedback from the receiver. Complexity and bandwidth are the costs of having CSI at the transmitter. The channel conditions under which capacity gain could compensate for these additional costs are determined. Relative phases of the LOS components of the respective SISO channels turn out to have an enormous impact on both capactiy and capacity gain. This fact allows to extract some conclusions about the importance of antenna placement for fix receiver and transmitter and the significance of CSI at the transmitter when mobility is considered. |
| 254 Sound Source Tracking Using Microphone Arrays |
pp. 2541-2545 |
|
|
by : Peng Wang, Wee Ser
Abstract: - The capability of being able to track moving sound sources is one of the most important characteristics for microphone array systems. In this paper, a simple source tracking scheme is presented. Performance analysis and simulation results are also presented. |
| 255 A parallel processing engine for motion estimation in MPEG-4 multimedia applications |
pp. 2551-2556 |
|
|
by : Rui Gao, Donglai Xu and John P. Bentley
Abstract: - In this paper, a parallel VLSI processing architecture for MPEG-4 standard motion estimation implementation is proposed. It possesses the characteristics of low memory bandwidth and low clock rate requirements, thus primarily aiming at 3G mobile applications. Based on one dimensional tree architecture, the processing core employs dual-register/buffer technique to reduce the preload and alignment cycles. As an example, a full-search block matching algorithm has been mapped on this architecture using a 16-elements PE array, which has the ability to calculate motion vectors of QCIF video sequences in real time at 8.2 MHz clock rate and using 15.5 Mbytes/s memory bandwidth. |
| 259 Using hidden markov models in segmentation of speaker-independent connected-digits corpus |
pp. 2591-2595 |
|
|
by : Francisco Dºaz, Manuel Rubio, Pedro G?mez, Victor Nieto, Victoria Rodellar
Abstract: - The first task to be accomplished in speech recognition is the segmentation and labeling of records. Regarding speech, this is a very complicated and costly procedure, although of most importance because at the present time many available speech corpora are not segmented. This paper proposes a semi-automatic segmentation method in order to reduce the manual segmentation burden of a very large corpus. First, Hidden Markov Models are created with a reduced set of records. Afterwards they are used to perform an automatic segmentation on the rest. Recursively, new more robust models are created and used to create new segmentations. The method consists in three main steps: (1) Initial Reduced Segmentation, (2) Recursive-Extended Segmentation and (3) Post-processing of the labels. This method was evaluated in the segmentation of the TIDIGITS corpus with two independent initial manual segmentations. Finally the method was able to label correctly 96.18 % and 95.72 % of the corpus records, respectively. |
| 260 On-line state observers for baker's yeast fermentation |
pp. 2601-2606 |
|
|
by : Celina P. Leao, Filomena O. Soares
Abstract: - Product quality and productivity improvement of a fermentation process implies, usually, reliable on-line measurements. Due to the difficulties, high costs and inefficient technology to obtain on-line measurements, several methodologies were developed to overcome these problems. In this work, algorithms for on-line state observation and kinetics parameters estimation applied to fed-batch bakerÔs yeast fermentation process are presented. The algorithms, initially proposed by [12] are used for a brief theoretical study and validated with experimental data. Only two on-line state variables measurements, oxygen and carbon dioxide, are used to estimate all the state and the kinetics parameters. The method shows to be very sensitive to measurements precision, in particular, in the oxygen transfer rate evaluation. A corrective parameter is then introduced and consequently a better performance of the estimators is obtained. |
| 262 The performance analysis of B-WLL system using pre-equalization techniques with fast adaptive algorithms under the Ka-Band channel |
pp. 2621-2623 |
|
|
by : Byungchul Kim,Yunseok Choi,Kyunhyon Tchah
Abstract: - In this paper, we propose the pre-equalization technique for the uplink burst transmission under the intersymbol interference (ISI) channel and the rain attenuation channel existed in the Ka-Band (20~30GHz) of the B-WLL system. We compare the mean-square-error (MSE) convergence properties of standard least-mean-square (SLMS) and discrete cosine transform (DCT)-based transform-domain least-mean-square (TRLMS) algorithms, analyze the BER performance of pre-equalization and post-equalization using TRLMS and SLMS algorithms. Pre-equalization, which is a technique to enhance the BER performance by avoiding the noise enhancement of receiver, needs the post-equalization as an initial step to extract the tap coefficients of pre-equalizer in the transmitter. The simulation results show that TRLMS as a post-equalization algorithm provides much faster MSE convergence rate and lower steady-state MSE properties than SLMS does under the ISI and rain attenuation channel. From the above MSE results, it is preferable to use the TRLMS as an adaptive filter algorithm of pre-equalizer system rather than the SLMS to obtain better BER performance. |
| 263 Computer Semantics of Russian |
pp. 2631-2636 |
|
|
by : Vitaly Tuzov
Abstract: - The brief summary of solving the problem of the computer semantic analysis of Russian texts is presented here. The essence of the analysis is the translation from Russian into the formal semantic language. The semantic dictionary which is the computer analogue of a learnerfs dictionary (replete with explanation) is used for translation. The dictionary contains more than a hundred thousands lexical units. Its basis consists of nearly twenty thousands basic concepts (a language thesaurus) divided into 1450 classes and a small collection of about two dozens basic functions. |
| 264 Effective design and measurements of switched current circuits |
pp. 2641-2645 |
|
|
by : Mariusz Jankowski, Zygmunt Ciota, Andrzej Napieralski, Malgorzata Napieralska
Abstract: - Switched current circuits cover such important areas of application like filtering systems, neural networks, A/D and D/A conversion. In this paper very simple and effective design method and practical realization of switched current filters has been presented. Current mode circuits, because of their special structure, need appropriate methodology during design preparation. This paper is an overview of most common problems related with laboratory measurements of such circuits. Very efficient laboratory station for measurements of integrated circuits has been presented. The control computer permits to obtain in a simply way precision characteristics of current mode circuits. |
| 272 A Mathematical Modeling and a Computer Simulation of an Epidemic Model |
pp. 2721-2724 |
|
|
by : Ponidi, Heru Cahyadi
Abstract: - In this paper we will discuss a mathematical modeling process and a computer simulation of an epidemic problem. The problem is modeled by control optimal, and then analyze by the maximal principle of Pontryagin with bang-bang control. The result will be simulated using MATLAB. As the result is the schedule of effective medical treatments can be determined provided that the values of parameters are known. |
| 273 A Two-Level Learning Hierarchy for Constructing Incremental Projection Generalizing Neural Networks and Its Application in Artificial Odor Discrimination System |
pp. 2731-2736 |
|
|
by : Hendri Murfi, Benyamin Kusumoputro
Abstract: - One of incremental learning-based neural networks that theoretically guarantees the optimal generalization capability and provides exactly the same generalization capability as that obtained by batch learning is incremental projection generalizing neural networks. This paper will describe a two-level learning hierarchy for constructing the networks. An incremental projection learning in neural networks algorithm is employed at the lower level to construct the network while the learning parameters, the orders of the reproducing kernel Hilbert space, are optimized using a genetic algorithm at the upper level. The networks produced by this learning hierarchy will be used as subsystem of the artificial odor discrimination system to approximate percentage of alcohol. |
| 281 Adaptive speech separation using hybrid approach |
pp. 2811-2816 |
|
|
by : Yan Li
Abstract: - A hybrid iterative learning algorithm for recurrent neural networks based on higher-order statistics to blind signal separation is presented in this paper. Fourth-order statistics are used as the separation criterion to train an RNN to perform the separation. Some simulation results for both artificially convoluted audio signals and real recordings demonstrate that the proposed approach is promising. |
| 284 Boundary between CCM and DCM in DC/DC PWM converters |
pp. 2841-2845 |
|
|
by : Elena Niculescu and Eugen Iancu
Abstract: - It is presented a modality to compute the boundary between the discontinuous (DCM) and continuous (CCM) operating modes in the DC/DC fourth-order PWM converters with parasitic resistances. To reach this goal, an averaged model of the converter has been used. The boundary between DCM and CCM, considered to be the critical load resistance Rcrit, or, equivalently, the critical value of the inductor conduction parameter kctrit, is found as the positive root of a second-degree equation with unknown Rcrit or kctrit. The algorithm carried out and implemented with MATLAB environment allows studying the effect of the parasitic resistances and the coupling of inductors over the boundary Rcrit or ktcrit, the steady-state properties and the external characteristics of the converter. The simulation results show that the parasitic resistances and the inductor coupling change the boundary between the discontinuous and continuous conduction modes of a fourth-order PWM converter. |
| 285 An Approach to Usability Evaluation of an Intelligent Tutoring System |
pp. 2851-2856 |
|
|
by : Andrina Granic and Vlado Glavinic
Abstract: - Intelligent tutoring systems are a generation of computerized educational systems, which are intended to improve the process of learning and teaching by attempting to mimic the capabilities of human teachers. The development of application technology in the field of computerized educational systems has not been accompanied by an adequate usable, user-centered interface design, possibly resulting in scarce, unusable support for interaction and indeed in the decrease of interface’s effectiveness and efficiency. In order to improve the usability of an intelligent tutoring system, nowadays perhaps the most important goal of current research in the field of human-computer interaction in general, scenario-based usability evaluation is advocated as a formal approach for intelligent tutoring system evaluation. |
| 286 Regions of the accepted tolerance for the recognition of binary imprints in robotics |
pp. 2861-2865 |
|
|
by : Dorina Mioara Purcaru
Abstract: - The tactile matrix sensors often generate the binary imprints of the explored shapes in robotics. For the shape recognition, their vectors of the characteristic features define both the prototypes and the unknown shapes. The similarity between two shapes is estimated with different distances. The paper recommends the reference distances and defines the regions of the accepted tolerance, associated with a prototype. The shapes of these regions in the parameter plane and 3D-parameter space are presented and analyzed for various distances. |
| 289 Proposed Principles for Context Development of e-Learning Training Courses |
pp. 2891-2897 |
|
|
by : Stavros Cotsakis, Andreas Generalis, Maria Loumou, Elena Kayafa, Vassilis Loumos and Eleytherios Kayafas
Abstract: - Growing evidence suggests that the success of a learning system may critically depend not only on the nature of the environment in which the learner is placed, but also on the methodology for context development and for learners' evaluation. This paper is based on the learning factors and on the theories of classical instruction and proposes a methodology for context development of e-learning training courses. |
| 290 Internet Traffic Generation for Large Simulations Scenarios |
pp. 2901-2906 |
|
|
by : Kai Below, Ulrich Killat
Abstract: - The simulation of realistic networks is an important aspect in the development of new protocols or algorithms for communication networks. The traditional approach for such a simulation is to connect a certain amount of sources to network nodes and to mea |
| 291 An Intelligent Agent based Approach for Service Discovery in Wireless Ad-hoc Networks |
pp. 2911-2914 |
|
|
by : Minas Pertselakis, Nicolas Tsapatsoulis
Abstract: - In this paper we propose an architecture for modeling the service discovery functionality of computing devices operating in pervasive computing environments. Forming wireless ad hoc networks requires service discovery protocols that support both proximity-based detection and abstract level service matching. In the proposed architecture, services, provided by sensors and actuators, are indirectly located through polling while composite service matching is based on the intelligent agent philosophy. In this way, our service discovery approach exploits the spread of proximity based device detection and the dynamics of the multi-agent environments. |
| 292 Using Acme to Specify the Software Architecture of the OpenH323 Open Source Code |
pp. 2921-2926 |
|
|
by : Fernando Nascimento, Antonio Costa, Virginia Paula, Guido Souza
Abstract: - This paper presents an Acme specification of the software architecture of the OpenH323 open source code. OpenH323 is a collaborative initiative that aims to produce a public implementation of the ITU-T recommendation H.323. Providing this architectural description, we intended to make easy the maintenance and reuse of this open source code. It is well known that the implementation of a videoconference system requires the conception of a very complex code. One of the approaches to implement such type of system is to reuse code. Considering this, we highlight, as the main contributions of our work, to facilitate and to accelerate the implementation of videoconference systems compatible with the H.323 recommendation through the reuse of the OpenH323 code. |
| 293 Optimization of Signaling Traffic in Centralized Conferences using SIP |
pp. 2931-2936 |
|
|
by : Igor Miladinovic, Johannes Stadler
Abstract: - Multiparty conferencing is becoming an important topic in the next generation networks. An essential part of multiparty conferencing is the signaling protocol, which has to establish, modify, update and terminate conferencing sessions. This paper focuses on signaling for closed centralized conferencing using Session Initiation Protocol (SIP) and proposes an approach that achieves an optimization of signaling traffic on the conference server. Presented approach also improves fault tolerance of centralized conferences by preventing the conference server to be a single point of failure. The results obtained with this optimization are also present. |
| 294 On-line controller autocalibration based on parameter predictors: a case study |
pp. 2941-2945 |
|
|
by : M. Eduarda Silva, Teresa Mendonca, Isabel Silva, Hugo Magalhaes
Abstract: - In this paper, a method for the on-line autocalibration of a controller, based on statistical regression techniques, is presented. The control of neuromuscular blockade is used as a case study. The method relies on the information about the patients dynamics, inferred from the response induced by the initial bolus dose given in the beginning of anaesthesia. A new approach to the characterization of the bolus response based on Walsh-Fourier spectral analysis is considered. A detailed comparison of alternative descriptions is presented. |
| 295 Computation of Motion Activity Descriptors in Video Segments |
pp. 2951-2956 |
|
|
by : Junghwan Oh and Praveen Sankuratri
Abstract: - Recently, motion activity which is defined as amount of motion in a video sequence has been included as a descriptor in MPEG-7 standard. The motion activity descriptors (MADs) which describe this motion activity need to enable efficient content analyzing, indexing, browsing, and querying of video data. To address this issue, first, we propose a novel technique for automatic measurement of motion activity using accumulation of quantized pixel differences among the frames of given video segment. As a result, accumulated motions of shot are represented as a two dimensional matrix. Also, we investigate an efficient and scalable technique to compare these matrices and generate MADs that are representing various motions of shots effectively. Not only the degrees(amounts) but also the locations of motions are computed and presented accurately. Our preliminary experimental studies indicate that the proposed techniques are effective in capturing and comparing motion activities. |
| 296 Shadow volumes for mpeg-4 applications |
pp. 2961-2966 |
|
|
by : Helge Drumm Stephan Kussmaul
Abstract: - The method to create shadows with shadow volumes for real-time applications, was introduced by Crow and improved by Heidmann in „Real Shadows Real Time“. Heidmann used the Stencil Buffer for the calculation of the inner geometry of shadow volumes. The fundamental algorithm is based on utilization of point lights, supported by OpenGL. Shadow volumes normally only can be applied on polygonal geometry. In recent time, McCool published another approach in „Shadow Volume Reconstruction from Depth Maps“. This algorithm slightly changed the basic algorithm, so that it can be used for arbitrary objects. This caused an enormous improvement. Nowadays, shadow volumes for real-time rendering are the most practicably and used method for shadow creation. The content of this paper describes the results of recent action researches for applications of shadow volumes in MPEG-4 environment. The research project is called „Interactive AudioVisual Application Systems” (IAVAS) and executed on the technical university Ilmenau (Germany) institute for media technology (IMT). Principle purpose of work was the standardisation of shadow volumes as node in MPEG-4. Even for this where important aspects for instance performance and improvement of visual quality for shadow volumes. The results of our work shows one possibility of standardisation for the shadow volume node in connection with proposals for robust implementations. |
| 297 Real-time Non-Invasive Eyetracking and Gaze-point Determination for Human-Computer Interaction and Biomedicine |
pp. 2971-2979 |
|
|
by : Ashit Talukder, John-Michael Morookian, S. Monacos, R. Lam, C. LeBaw, A. Bond
Abstract: - Eyetracking is one of the latest technologies that has shown potential in several areas including human-computer interaction for people with and without disabilities, and for noninvasive monitoring, detection, and even diagnosis of physiological and neurological problems in individuals. Current non-invasive eyetracking methods achieve a 30 Hz rate with possibly low accuracy in gaze estimation, that is insufficient for many applications. We propose a new non-invasive visual eyetracking system that is capable of operating at speeds as high as 6-12 KHz. A new CCD video camera and hardware architecture is used, and a novel fast image processing algorithm leverages specific features of the input CCD camera to yield a real-time eyetracking system. A field programmable gate array (FPGA) is used to control the CCD camera and execute the image processing operations. Initial results show the excellent performance of our system under severe head motion and low contrast conditions. |
| 298 Control of a Public Transport Network by the Max-Plus Algebra: case of a system constrained by maximal connection times |
pp. 2981-2983 |
|
|
by : A. Nait Sidi Moh, M.-A. Manier, H. Manier and A El Moudni
Abstract: - The theory of linear systems in the max-plus algebra is developed for the analysis of discrete event systems. The timed event graphs are one of the tools used for modelling these systems, and their behaviour can be described by (max, plus )-linear equations. This paper proposes a control structure for a public transport system. We mainly based the research of this control on the Residuation theory. This control aims at conceiving a timetable of buses such that: the connection time at interchange points respects an upper bound corresponding to a tolerance for passengers, and it limits the number of buses required to ensure the connections. |
| 303 A new genetic encoding for edge point stereo matching |
pp. 3031-3036 |
|
|
by : Issa Hazem, Ruichek Yassine, Postaire Jack-Gerard
Abstract: - Stereo vision is a well-known technique for obtaining depth information from two or more video images of a scene. The heavily investigated problem in this approach is how to find the corresponding features extracted from the stereo images. In this paper, we present a genetic scheme to the stereo correspondence problem where a new solution encoding is proposed. The problem is formulated in terms of finding pairs of true matches that satisfy local constraints and global ones. Local constraints are used to encode only compatible matches, whereas global constraints are used to define the fitness function such that best correspondences correspond to its minima. To demonstrate its effectiveness, the proposed genetic stereo matching is applied for extracting depth information of a scene seen by a linear stereoscopic sensor. |
| 308 Hidden Markov models suitable for text generation |
pp. 3081-3084 |
|
|
by : Grzegorz Szymanski, Zygmunt Ciota
Abstract: - The paper presents the application of Hidden Markov Models to text generation in Polish language. A program generating text, taking advantage of Hidden Markov Models was developed. The program uses a reference text to learn the possible sequences of letters. The results of text processing have been also discussed. The presented approach can be also helpful in speech recognition process. |
| 312 Choice of method for General One-Dimensional Cutting Stock Problem |
pp. 3121-3125 |
|
|
by : Peter Trkman, Miro Gradisar
Abstract: - Problems in operations research are often NP complete and optimal solution for larger problems cannot be obtained, therefore heuristic methods have to be used. The boundary between the size of the problem that can be solved optimally and those where heuristic methods should be used is not always clear. To solve this problem we propose a method based on decision trees. The method can be also used to determine which variables have the largest influence on complexity of the problem. The proposed method was tested on General One-Dimensional Cutting Stock Problem. The result is reduced trim loss. |
| 313 A simulation study of the influence of vector and scalar filters on the matching precision of stereo images |
pp. 3131-3136 |
|
|
by : Jerzy Siuzdak, Tomasz Czarnecki
Abstract: - Investigated is the problem of points matching in stereo images (both color and monochrome). The correlation method is applied to the areas around the points being matched. The influence of noise on the matching errors and the error reduction by vector and scalar filters are analyzed. The obtained results indicate that there is only a small difference between the matching precision of color and monochrome imapges as well as that only two of several analyzed filters are suitable for the precision increase. These filters are an adaptive filter and an impulse noise filter designed by the authors. |
| 314 Robust Voice Activity Detector under the Noise Environment in G.723.1 Vocoder |
pp. 3141-3143 |
|
|
by : KyungA Jang, SoYeon Min, Myung Jin Bae
Abstract: - Generally the one of problems in Voice Activity Detection (VAD) is speech region detection in noise environment. Therefore, this paper proposes the new method using the energy, lsp varation. Processing time and speech quality of the proposed algorithm as a result, the processing time is reduced due to the accurate detection of inactive period, and there is almost no difference in the subjective speech quality test. Proposed algorithm measures the number of VAD=1 and the results shows predominant reduction of bit rate as SNR of the noisy speech is low (about 5~10dB). |
| 315 The pitch extraction method through spectrum flattening |
pp. 3151-3154 |
|
|
by : SoYeon Min, KyungA Jang, MyungJin Bae
Abstract: - The exact pitch (fundamental frequency) extraction is important in speech signal processing like speech recognition, speech analysis and synthesis. However the exact pitch extraction from speech signal is very difficult due to the effect of formant and transitional amplitude. So in this paper, the pitch is detected after the elimination of formant ingredients by flattening the spectrum in frequency region. The effect of the transition and change of phoneme is low in frequency region. In this paper we proposed the new flattening method of log spectrum and the performance was compared with LPC method and Cepstrum method. The results show the proposed method is better than conventional method. |
| 316 Neuro-Fuzzy Hybrid System Applied to the Optimal Reactive Power Flow Problem |
pp. 3161-3169 |
|
|
by : Maury Meirelles Gouvea Jr., Luiz Danilo Barbosa Terra
Abstract: - In this paper, the integration of Artificial Intelligence and sensitivity analysis is described for the formulation and resolution of the optimal reactive load flow problem. Instead of formulating of a scalar optimization problem, the development of a fuzzy multi-objective programming is proposed. This treatment considers all the objectives and constraints in a satisfaction degree. Additionally, an Artificial Neural Network is used to supply the voltage collapse margin of the weakest bus, also considered an additional objective in the problem. The several steps and advantages involving this approach applied to the multi-objective problem are discussed. The developed algorithm makes it possible to create a fuzzy decision function that can optimize the reactive power flow, while eliminating voltage violations. The efficiency of the method is investigated in case studies using the standard IEEE-30 bus bar system. |
| 319 Development of a Distributed System on Internet/Intranet Environment and Its Application to Spatial Multi-source Data Integration |
pp. 3191-3194 |
|
|
by : Margareth Simoes P. Meirelles, Oscar Farias
Abstract: - This paper relates our experience in developing DISI2E - a distributed system on an Internet/Intranet environment; built from modules the same way children play LEGO. These modules can be any kind of file visualized by a browser: pictures, html files, audio files, animations files and other types of files visualized via a plug-in, like executable programs (CGI), applets and queries to databases. The user through an interface very similar to the Windows Explorer selects these modules. Then, these modules, following the object-oriented paradigm, are related basically in part-of relations (containment relationships, like regions on a map) or some of them – simpler – are taken as attributes of more complex ones. All the user needs to operate DISI2E are http (hyper text transfer protocol) servers – to publish the distributed information – and browsers, enabling users to navigate through the distributed system visualizing non-structured information, querying databases and executing programs. |
| 320 A neural network hybrid model for an optical braille recognitor |
pp. 3201-3206 |
|
|
by : Giovanna Morgavi Mauro Morando
Abstract: - Braille is the most widely used system for written communication using tactile means. Automation of the Braille reading process will effectively enable the transcription and duplication of existing documents as well as their preservation. By converting a Braille document into electronic form, all the benefits of developments for the electronic form of printed documents can be enjoyed: easier transmission, efficient storage and manipulation etc.. The proposed procedures use approaches commonly applied in standard OCR systems. It uses a commercially available flatbed scanner to acquire a grayscale image of a Braille document from which the characters are recognized. This prototype system has been developed by using methods from image processing and analysis to AI. These methods achieve high speed and accuracy by using hybrid neural network approaches. The system dynamically adapts to factors such as the quality of the input pattern, its intrinsic similarities and differences between patterns. |
| 321 Developing A Predictive Model For Groundwater Quality Based On Bayesian Reasoning |
pp. 3211-3216 |
|
|
by : Khalil I. Shihab, Fawaz A. Masoud and Haider Ramadhan
Abstract: - Bayesian reasoning method is becoming an increasingly important area for research and application in the entire field of Artificial Intelligence. This paper explores the use of a predictive model for groundwater quality based on Bayesian reasoning. Moreover, this work addresses the assessment of groundwater quality in Sultanate of Oman, especially in the areas of Wadi Suq and Wadi Aljizi, based on Bayesian reasoning methods. Its primary aim is to develop a predictive groundwater quality model and computer system prototype to assess and predict the impact of pollutants on the water column. The methods presented are widely applicable and are able to overcome many of the problems encountered with the other methods. |
| 323 Wavelet-Based Digital Image Watermarking using Level Adaptive Threshold |
pp. 3231-3234 |
|
|
by : P. Thitimajshima, Y. Rangsanseri, M. Ladthawanidphan, P. Rakpratanporn
Abstract: - In this paper, we propose a method of digital watermarking for still image using multi-stage discrete wavelet transform which does not require the original image for watermark detection. The watermark, generated by pseudo-random sequence with a normalized distribution of zero mean and unit variance, is added to all high-frequency coefficients which are above a level adaptive threshold. By comparing the correlation between the wavelet coefficients of a possibly corrupted watermarked image and the watermark with a threshold, the embedded watermark can be detected. Experimental results show that the embedded watermark is robust against various signal processing and compression attacks. |
| 324 A New Adaptive Algorithm for Bit Allocation in DMT Modems |
pp. 3241-3243 |
|
|
by : Daniel Franklin and Ian Burnett
Abstract: - This paper presents a new algorithm for distributing bits to the subchannels of a DMT modem. The so-called Inverse Robin Hood algorithm converges rapidly and can effectively optimise the overall bit error rate in a non-Gaussian channel. |
| 326 Using Color Histograms to Recognize People in Real Time Visual Surveillance |
pp. 3261-3264 |
|
|
by : Daniel Wojtaszek, Robert Laganiere
Abstract: - This paper presents a surveillance system that detects and recognizes people in indoor scenes. To distinguish between different people, color histograms on a perceptually uniform color space are used. The goal here is to associate a sequence showing a person leaving a room with the previously recorded sequence showing that same person entering. |
| 327 The Control System of the Device Simulator MINIMOS-NT |
pp. 3271-3276 |
|
|
by : Robert Klima, Tibor Grasser, Siegfried Selberherr
Abstract: - State-of-the-art TCAD software like the multi-dimensional device and circuit simulator MINIMOS-NT require a huge amount of information in addition to the device input data to control several different complex modules and tasks. This information is normally hierarchically structured containing arbitrary cross-relations and dependencies that include, e.g., material properties, circuit descriptions, models and model parameters, or simulation parameters. Therefore, the control system for TCAD software must be able to handle these data and to allow simulation control in an efficient and comfortable manner. To obtain a maximum of flexibility and controllability a new specialized object-oriented dynamic database is used. |
| 328 Microactuators systems of torsion silicon cantilever |
pp. 3281-3284 |
|
|
by : Jacek Golebiowski
Abstract: Magnetic microactuation systems of torsion silicon cantilever is described. Devices are constructed in a batch fabrication process which combines electroplating with conventional photolithography , materials, and equipment. The magnetic transducers are applied to generation of motion in micromechanical structures such as cantilever, beam, membrane. Microelectromechanical systems (MEMS) with magnetic transducers are fabricated in order to combine the conventional silicon based micromachining techniques with the techniques of the magnetic thin film (nickel-iron). The mechanical and magnetic analysis of the silicon cantilever in the magnetic field was also described. Theoretical expressions for the displacement and torque are developed and compared to experimental results. |
| 330 Integrated E-Learning |
pp. 3301-3305 |
|
|
by : Oksana Myronovych, Nazar Younis
Abstract: - In this paper we analyze the evolution of the modern educational delivery system. The main emerging educational paradigms are identified. These paradigms are considered as main features of a new educational delivery model. This model is treated as an e-business model as a justification for its development and implementation using the underlying technologies. The main idea of the complemented and compound e-learning is to create a “culture of Internet education” at school, college and enterprise levels, providing continuous access to the on-line education. |
| 332 The use of clustering methods to combine equalization with decoding |
pp. 3321-3326 |
|
|
by : Adina Burian, Arto Kantsila, Jarmo Takala
Abstract: - In this paper, the task of channel equalization is considered as a classification problem. A classification method using adaptive clustering is presented. The possibility of replacing the channel equalization and decoding solely by classification, using clustering techniques, is investigated. The receiver performs jointly the decoding and the equalization of the encoded received signal. The transmitted data forms at the receiver-end, clusters around the points that corresponds to the noise free communication channel. The used technique consists of unraveling these clusters and exploiting their underlying relation, in order to obtain a direct application of a Viterbi-type combined procedure. Conventional equalization techniques, which employ a linear transversal equalizer and a decision feedback equalizer, are used as benchmarks. |
| 339 Resolution limits of electrocardiography: Evidences of a model study |
pp. 3391-3395 |
|
|
by : Krisztina Szakolczai, Kristof Haraszti, Gyorgy Kozmann
Abstract: - According to previous modeling studies - propagation of depolarizing wavefronts - consists of subintervals, each characterized by an epoch of smooth progression of activation. At the onset and end of these intervals, abrupt changes occur in the 3D pattern of activation waves (e.g. at the time of collision of an activation wave with other waves, or unexcitable myocardium), which manifest themselves in the surface ECG as “jumps” (high frequency notches and slurs). The amplitude and timing of jumps provides diagnostic information on bioelectrical tissue properties of the heart. According to the numerical modeling results the resolution of electrocardiography is in the range of 5-6 mm in the case of wave-initiation point dislocation, while intramural spherical unexcitable tissue segments with a radius of 5-6 mm are detectable. |
| 340 Effect of Ataprive Nonlinearity in Speech Coding |
pp. 3401-3406 |
|
|
by : Jari Turunen, Pekka Loula, Juha T. Tanttu
Abstract: - This paper is focused in the effect of adaptive weighted non-linearity in speech coding. The results showed that the effect of adaptive non-linearity improves the spectral distance measure when compared to the system without non-linearity. |
| 341 Design of a New Chip Architecture for a Home Gateway |
pp. 3411-3416 |
|
|
by : Kwangsoon Choi, Kwangmo Jung, Myunghyun Yoon
Abstract: - As Internet is becoming popular to everyone recently, demands for higher-quality services such as VOD and home networking have been increasing. Especially, home networking system can interconnect and control home appliances which use different protocols via Internet. This means that a common protocol to communicate with each other and a new system architecture to implement the common protocol are needed. In this paper, we propose a common protocol and a novel chip architecture with a memory management scheme for a home gateway system. |
| 342 Speech recognition technology for dysarthric speech |
pp. 3421-3426 |
|
|
by : Peter E Roberts
Abstract: - The initial results of investigations into the use of current commercial automatic speech recognition (ASR) software by people with speech disability (dysarthria) is presented, together with a brief summary of the history of the development of ASR and its applications for the disabled. Results confirm the viability of dysarthric use, identify areas of further investigation for improved recognition performance and for development of a clinical tool for speech measurement. |
| 343 Towards Map Building and Space Coverage Planning in Robotics |
pp. 3431-3436 |
|
|
by : Jiri Pavlicek, Roman Mazl, Petr Stepan, Libor Preucil
Abstract: - This paper describes new method for building occupancy grid from single camera, automatic calibration of this method and approach how to use this internal representation for cleaning unknown environment. The algorithm is designed for indoor application. This assumption is used for calculating inverse perspective transformation of the gathered image. The paper brings an overview of path-planning methods targeted on complete coverage of an operating space. Two algorithms for planning coverage based on occupancy grid are outlined and their fundamental properties are compared. Firstly the existing approach to solution of the coverage problem based on cellular decomposition of the working space is discussed and further the method applies border expansion is introduced. Attached experiments verify and compare the newly designed and the existing methods as well as illustrate the final performance of the methods in real environments. |
| 344 Supproting Collaborative Audio in the Internet |
pp. 3441-3447 |
|
|
by : Milena Radenkovic, Chris Greenhalgh
Abstract: - Distributed Partial Mixing (DPM) is an approach to creating a distributed audio service that supports optimisation of bandwidth utilization across multiple related audio streams (e.g. from concurrently active audio sources) while maintaining fairness to TCP traffic in best effort networks. This paper extends our DPM prototype to include the application control of selector and database components of DPM as a necessary complement to purely network-driven DPM presented till now. In particular we introduce awareness driven and Interest Management driven DPM as one way of that smoothing out the effects of the network adaptation on the end user audio quality in Collaborative Environments. |
| 345 A Cellular Fuzzy Associative Memory for Image Fuzzification in Robot Vision Systems |
pp. 3451-3454 |
|
|
by : Leonarda Carnimeo, Antonio Giaquinto
Abstract: - In this paper a Cellular Fuzzy Associative Memory containing fuzzy rules for bidimensional image fuzzification in robot vision systems is developed. This cellular processor constitutes a subsystem of a CNN-based architecture which can store both bidimensional patterns and the rules to process them. After establishing the fuzzy rules characterizing the Fuzzy Associative Memory, a CNN behaving as an associative memory is properly synthesized to store them. A numerical example is reported to show how the synthesized Cellular FAM can process bidimensional patterns for industrial object matching analyses. |
| 346 A Privacy-Friendly Electronic Payment Model: Balancing Copyright and Privacy Protection Requirements |
pp. 3461-3465 |
|
|
by : V. C. Zorkadis, D. A. Karras, and E. S. Siougle
Abstract: - The electronic distribution of digital works leads to the collection and processing of huge amounts of personal data, and copyright management information. However, international treaties, EU Directives and National Laws contain provisions which protect these data, i.e. there are statutory obligations related to the protection of privacy, copyright management and electronic transactions. The enforcement of the legal framework requires the employment of appropriate security and organizational measures. To address these problems, we propose a privacy-friendly electronic payment model, which aims at balancing contradicting copyright and privacy protection-related requirements. With our model, electronic consumers can anonymously order and pay digital works, while the media distributors involved in the transaction and the copyright owners are protected against intellectual property rights violations. The entities involved in the transactions, range from creator and copyright holder to media distributor, the monitoring and the privacy service provider and the bank gateway. |
| 350 Multi-user Detection for Wideband CDMA based on the Conjugate Gradient Method |
pp. 3501-3505 |
|
|
by : Kiyomi Igarashi, Yoshikazu Arai, Tomoya Sakai, Atsushi Fukasawa, Yumi Takizawa
Abstract: - Multiple users communicate with individually assigned codes on a single radio carrier frequency in a DS-CDMA system. Mutual interference among codes limits the capacity of the system. Conventional detectors of receivers are for single-user, which are designed neglecting the mutual interference. Multi-user detection is expected to enhance the capacity. In this paper, a new scheme is proposed corresponding to solve a huge matrix equation under the multiple access interference (MAI) environments. This scheme is related to the mathematics; the Conjugate Gradient Method (CGM). The modeling of multi-user detection is shown, then the algorithm is given based on the mathematics; the Conjugate Gradient Method (CGM). The characteristics of the proposed scheme are evaluated with the comparison of the conventional single user detector. The proposed scheme is found to have reduced complexity comparing conventional technologies based on the serial and the parallel configurations. |
| 351 A model for product risk estimation through corporate memory and techniques integration |
pp. 3511-3515 |
|
|
by : Vrassidas Leopoulos, Konstantinos Kirytopoulos, Vaggelis Bellos
Abstract: - Aim of this paper is to enhance product risk estimation by providing an integrated model. The similarity of “technical” and “managerial” risk management methods is identified and there is an effort to compromise two of the most known and widely accepted ones, the FMECA (Failure Modes, Effects and Criticality Analysis) and the PMI (Project Management Institute) risk processes. Moreover, the paper suggests that one of the most important factors for an efficient risk management process implementation is the maintaining of a corporate memory. The information existing in the corporate memory feeds a risk estimation algorithm, which provides a common, for “technical” and “managerial” risk management methods, index of the product risk. |
| 352 Bluetooth implementation frameworks |
pp. 3521-3529 |
|
|
by : Harmatne Medve Anna
Abstract: - This paper will present an SDL implementation of Bluetooth mobile system’s protocol stack. This implementation is a part of research and development of a validation protocol systems implementation. The final goal is to develop a method for protocol implementation frameworks using design patterns of Protocol Systems in SDL This work presents common problems and solutions for specification, design and implementation of communication protocols and gives a prototype implementation framework Bluetooth’s usage models. |
| 355 Application of an improved diploid genetic algorithm for optimizing performance through dynamic load balancing |
pp. 3551-3556 |
|
|
by : A. Sima Uyar, A. Emre Harmanci
Abstract: - The dynamic load balancing problem which can be defined as the effective redistribution of workload among the system processing units during execution time is dynamic in nature where the load and the processing power of the system may change in time as units of work enter and leave the system and processing units area added to or removed from the processing pool. To adress this problem, genetic algorithms are used in literature in different ways. In this study, promising results of applying an improved diploid genetic algorithm for load balancing a simulation of a network of processing units are reported. |
| 357 Simulation e-business applications using PNML |
pp. 3571-3576 |
|
|
by : Ines Brosso, Graca Bressan,Wilson Ruggiero
Abstract: - This paper proposes the use of XML Petri Nets (PNML), a kind of high-level Petri Nets, for simulations Web-based systems, e-Activities like e-Business, Business-to-Business (B2B), e-Commerce and mobile applications wireless in the Web or ad-hoc networks. The combination of Petri Nets and XML (Extensible Markup Language) files are emerging as an important approach for the specification of complex processing business tasks in distributed heterogeneous systems, allowing the follow-up of electronic business documents and data interchange process. In a Case Study, the proposed methodology is applied in the development of an e-business application, the customer surfs in the catalogue of products and adds desired items to its basket of purchases, the system verifies the authorization of the credit card and confirms the purchase immediately and soon sends an email to follow contends an identification code so that the customer follows its purchase. |
| 358 Models of mobile payments |
pp. 3581-3586 |
|
|
by : Marjan Gusev, Ljupco Antovski, Goce Armenski
Abstract: - Mobile phones are already approaching penetration rates of close to 80 per cent in some parts of the world. Mobile payments, or “m-payments”, are expected to become an important part of retail payments. M-payments are defined as payments that are carried out via mobile phone. M-commerce as a wide area could be divided into mobile E-commerce and M-trade area. Different models of mobile payments are proposed considering the physical disposition. Financial service provider is essential mediator among customers, merchants and banks. The iMS specification proposed in this paper enables mobile payments with one click of a button. Different levels of security have to be implemented for small, medium and large transaction of funds. |
| 359 Analysis of vibration responses of defective rolling bearings using Blind Source Separation |
pp. 3591-3596 |
|
|
by : C. T. Yiakopoulos and I. A. Antoniadis
Abstract: - Separating vibration signals generated by defective rolling bearings is of major importance for rotating machinery health monitoring, since rolling bearings of the same type are the quite frequently encountered in the industrial environment, especially in large machines, critical in the production process. For this reason, Blind Source Separation (BSS) is considered for application on vibration signals, which may include characteristic simulated defective bearing responses under different types of bearing faults. It is demonstrated that the BSS method, properly implemented, is quite able to decompose the measured signal into a number of independent components, each one corresponding to the vibration induced by an individual bearing. |
| 361 Non linear Wavelet packet denoising of impulsive vibration signals |
pp. 3611-3616 |
|
|
by : Nikolaos G. Nikolaou, Ioannis A. Antoniadis
Abstract: - In this paper an application is presented of the wavelet packet method for denoising of impulsive vibration signals. Vibration response of machines often includes signals with periodic excitation of resonances. The aim is to extract information regarding the physical mechanism which generates the impulsive characteristics of the signals. The signals are transformed using the wavelet packet method, and the resulting coefficients are nonlinearly modified. The reconstructed time waveform of the signal using the modified coefficients may reveal the impulsive characteristics, resulting in a safer identification of the source of the impacts. The approach and its parameters are evaluated on industrial signals resulting from defective bearings. |
| 363 A comparison of three neural networks for building local grid maps |
pp. 3631-3636 |
|
|
by : Jose Cruz, Urbano Nunes Jose Metrolho and Eurico Lopes
Abstract:- This paper addresses the local grid map-building problem. Sensor readings are interpreted using a feedforward neural network, and Bayesian rule is used to update the occupancy probabilities of the grid cells. A comparison of three neural network configurations is made in the map building of indoor environments, using the sonar sensors of the Nomad 200 simulator. The architecture with best results in simulations was implemented in a real mobile robot. |
| 364 Optimizing Conservative Parallel DE Simulation of WANs |
pp. 3641-3645 |
|
|
by : Benjamin Forgeau Ulrich Killat
Abstract: - A common problem encountered in the simulation of WANs is the huge amount of data that the simulator handles. Especially with a parallel discrete-events simulation scheme, this affects dramatically the performance because most of the data overloads the list of pending events. We propose a simple algorithm which increases the performance of parallel simulation by separating events according to their destination, and we verify the performance speedup on a simple parallel simulation model. |
| 365 Integrated video object segmentation and shape coding |
pp. 3651-3656 |
|
|
by : Janez Zaletelj
Abstract: - To enable content-based access and manipulation of video content, modern multimedia communications require per-object based access to video data. Two additional building blocks of video transmission systems needs to be defined, which enable access to video objects at the receiver end. These are video segmentation, which decomposes video frame into a set of layers, where each layer represents segmentation mask of one semantic video object, and shape coding system, which enables transmission of the segmentation information to the receiver. This paper presents a novel, integrated approach to the fore mentioned problem, where the parameters of object segmentation module and shape coding module are dynamically adjusted depending on bandwidth constraints. Background differencing with morphological filtering is used for extracting shape information, and B-splines are used for shape approximation. A predictive scheme for encoding of spline control points is presented and compared to MPEG-4 CAE shape coding. |
| 366 Analytic Tests for a Robust Numerical Algorithm for Predicting Semiconductor Freeze-out and Exhaustion |
pp. 3661-3663 |
|
|
by : Ron Pieper and Sherif Michael
Abstract: - In this report a complete analytic description for the temperature dependence of the majority carrier in a single impurity doped equilibrium semiconductor is proposed. This model should provide an attractive alternative to commonly used rules predicting the temperature boundaries for the exhaustion regime. This model is used to provide a confidence test for a less restrictive robust numerical model recently proposed in the literature. Both the analytic description and results for the numerical algorithm are compared under a variety of assumed conditions including band-gap narrowing. Limitations of the analytic model are also clarified. |
| 368 A Simplex-Cosine Method for Solving Hard Linear Problems |
pp. 3681-3686 |
|
|
by : Federico Trigos, Juan Frausto-Solis and Rafael Rivera-Lopez
Abstract: - In this paper is presented a method that permits to determine a best initial point of the simplex method (SM) proposed by Dantzig for the resolution of linear problems. This method is based on the analysis of the angles between the gradient of the objective function and the gradient of the each constraint in the problem. This method, called simplex-cosine method (SCM), selects one vertex of the feasible region where the simplex procedure starts, using the minor angles between gradients so that the number of iterations of the simplex procedure is decreasing, and the time for finding the optimum is better than the traditional SM. In this work, several instances of the linear problems are evaluated and compared with the traditional SM and the results suggest that SCM has better behavior that the SM. |
| 376 An adaptive QoS management based on priority control for home networking |
pp. 3761-3764 |
|
|
by : Seung-Ok Lim,Kwang-Mo Jung
Abstract: - As network technology is growing very rapidly, many people have had PCs and used Internet service with high-speed modems at home. So providing high quality network services in home and connecting home devices together in a network become a main issue in home networking. There are mainly three kinds of data traffic types in home network; control network data, Internet data, and high-speed multimedia stream data. A key technology in home networking is how to guarantee the quality of three kinds of services between WAN and Home networks, and also within Home networks. In this paper, we propose an efficient and adaptive quality of service (QoS) management using priority control. |
| 377 Natural Language Interface for Web-based Databases |
pp. 3771-3776 |
|
|
by : J. Antonio Zarate M., Rodolfo A. Pazos R., Alexander Gelbukh, Joaquin Perez O.
Abstract: - Advances in the work on interfaces that facilitate access to databases through Internet are presented. Increasing needs of the users that access computer resources, the technological advance in this field, and the limitations of the graphic interfaces and forms motivate the development of new solutions in human-machine interfaces. In recent years, natural language processing has received a new impulse and achieved sufficient maturity to become a real solution in human-machine interfaces. A general architecture of a system of natural language interface to Web-based databases is described, as well as the current advance of the project. A detailed review of the history and the state of the art of the problem is given. |
| 380 Selection of typical documents in a document flow |
pp. 3801-3806 |
|
|
by : Pavel Makagonov , Mikhail Alexandrov , Alexander Gelbukh
Abstract: - Given a large set of natural language documents, their numerical characteristics and the corresponding metrics allow selecting documents relevant for a given topic, dividing the set of relevant documents into groups that reflect sub-topics of the given topic, and selecting the most representative document in each group. A system based on such methods is used in the city administration of Moscow, Russia. |
| 381 Robot Mapping with a Topological Map of Local Space Representations |
pp. 3811-3816 |
|
|
by : Margaret Jefferies, Wai-Kiang Yeap, Jesse Baker
Abstract: - In this paper we show how a cognitive mapping theory [1,2] can be used to implement a navigational map for a robot. At the core of this theory is the notion that a representation is computed for each local space the robot visits. These representations are connected in the way they are experienced to form a topological network of local space descriptions. We show how the local space representation is computed using a laser scanner which gives a limited 180 degree view of the robot’s environment. We describe how the topological map grows as the robot moves through different local spaces. One of the most studied problems in robot mapping is that of tracking the robot’s location when sensory and odometric errors are accumulating in the robot’s map. We argue that the structure of the robot’s map should simplify the process of localising the robot and improving the map as the robot becomes more familiar with its environment. Thus we show how our representation facilitates both updating the map and tracking the robot’s location. |
| 383 Adaptive channel reservation scheme for satellite-based mobile communication systems |
pp. 3831-3835 |
|
|
by : Yee-Loo Foo, Kenzo Takahashi, Sze-Wei Lee
Abstract: - Due to the high mobility of low earth orbit (LEO) satellites, there is a significant number of handover attempts in a LEO satellite mobile communication system, causing a high handover failure rate. This paper proposes an Adaptive Channel Reservation Scheme (ACRS) which gives an overall optimal success rate for both handover and new call initialisation. ACRS is able to adapt to the local traffic condition and makes the best decision on the timing of sending out a channel reservation request. A simulation model has been built and tested on uneven traffic density condition. Promising results have been obtained and they agree with theoretical expectation. |
| 385 Complex Object Recognition Using a Biologically Plausible Neural Model |
pp. 3851-3856 |
|
|
by : Raul Muresan
Abstract: - The complex object recognition tasks are still one big problem in neurocomputing today. This paper presents a method of detecting and recognizing complex objects, in cluttered environment, in a purely feed-forward way, being able to account for ultra-rapid visual categorization. We used a retinotopic architecture of simple spiking neurons with different types of receptive fields, organized in a hierarchical fashion similar to the mammal visual path. Fast shunting inhibition had been implemented using a rank-order coding similar to that described by S. Thorpe. The main advantage of the neural model proposed is that it accepts a very small number of training examples (4-7) and is able to generalize very well. The model has been used to detect faces and automobiles in complex intensity images. |
| 386 Complementary Extremum Principles for Isoperimetric Optimisation Problems |
pp. 3861-3867 |
|
|
by : J.P. Curtis
Abstract: - An isoperimetric problem of the calculus of variations is reviewed. An integral functional, dependent on a function u (x), its derivative u' (x) and on the independent variable x, is minimised. The minimisation is subject to the isoperimetric constraint that the length of the integration interval remains constant. The well-known Euler equation of the fundamental problem of the calculus of variations is recovered with an additional relationship connecting the values of the function u and its derivative u„S at the ends of the interval. A new complementary extremum principle is derived, that offers an algorithm for determining lower bounds on the minimum value of the original functional. A special treatment of the degenerate case where there is linear dependence on u is presented. Examples of each case are given and the scope for future work is discussed. |
| 387 Simulation-Based Approach to Evaluation of Management Strategies in a Distributed Web Search System |
pp. 3871-3876 |
|
|
by : Rinat Khoussainov, Ahmed Patel
Abstract: - In our previous works, we advocated the use of distributed search architectures where multiple independently owned and managed topic-specific search engines act as one search system. This approach has significant advantages over the currently predominant centralised Web search model, including low market entry cost for individual search providers and the potential to stimulate the provision of high-quality services through competition. The potential for competition between engines, however, requires new approaches to effective engine management. Many complex is-sues arise such as deciding what topic an engine should specialise in as well as service pricing strategies. Performance evaluation is an important stage in the development of management algorithms. It is also a challenging problem due to the complex and interactive nature of distributed search systems. We propose a simulation-based approach to this problem that provides for fast low-cost and verifiable performance measurements. In this paper, we describe our simulation environment and present experimental results for a heuristic topic management algorithm obtained using the developed simulator. |
| 388 Building Robust Military Networks Using Advanced Software Tools |
pp. 3881-3885 |
|
|
by : D. Vassis, A. Tsakrikadakis, K. Panagiotopoulos, G. Kormentzas, D. Vergados, F. Lazarakis
Abstract: - The paper discusses a design framework for building robust military networks. The proposed framework includes advanced software tools and methods for defining optimal alternative routing paths, for examining network’s viability in cases of crisis and for performing network simulation. The application of the presented framework towards the design of a military network for the purposes of a national project concludes the paper. |
| 389 A GIS-Based Methodology for Designing Wireless Links |
pp. 3891-3895 |
|
|
by : F. Lazarakis, G. Kormentzas, D. Vergados, K. Panagiotopoulos, D. Vassis, A. Tsakrikadakis
Abstract: - Wireless technology is frequently used in military networks for establishing communication in areas characterized by non-uniform terrain morphology (i.e., mountains, islands, etc.) where mobile military units have to be moved and for creating also a security back-up network easily to be used for emergency situations (i.e., an enemy’s sabotage). Towards the exploitation of the wireless technology, the paper presents a methodology for specifying the wireless links, which constitute the backbone of a wireless military network. Although the presented methodology concerns the design of a military-purposed wireless network, it can be easily generalized in order to be applied in other types of wireless networks (e.g. commercial, corporative, pilot networks, etc.). An application of the proposed methodology is also discussed in the paper |
| 390 Using Agents to Reduce Maintenance Cost Complex Software Systems |
pp. 3901-3906 |
|
|
by : Costas Vassiliadis,Mikhail Dobrynin
Abstract: - PaTra (PAtch TRAcker) is an agent capable of keeping track of patches that have been applied (old) and have to be applied (new) to a complex software system. Patra uses Oracle's Metalink, a support web site that can provide a list of patches for a specific product and platform. The agent queries daily, to get a list of patches for platforms and products that are specified in its configuration file. PaTra finds new patches, stores their information into its database and sends electronic notifications to a system administrator. Upon receiving the notifications, the system administrator uses PaTra's interface to examine the new patches and make a final decision whether they have to be applied to the system or not. Although in its current implementation PaTra can be used with Oracle's ERP system, it can be modified for other complex software systems with a similar scheme of interaction between a software producer and a software consumer. This is a time consuming process especially when there are several instances of the same system and even different versions that have to be maintained. Software systems based on intelligent agent technology can effectively be used to automate some of the maintenance procedures. This work proposes to delegate the task of keeping track of patches to an intelligent software agent. This automation saves time and cuts the cost of maintaining the system in general. |
| 392 Running mobile agent code over simulated inter-networks: an extra gear towards distributed system evaluation |
pp. 3921-3926 |
|
|
by : Antonio Liotta, Carmelo Ragusa, George Pavlou
Abstract: - Mobile Agent (MA) systems are complex software entities whose behavior, performance and effectiveness cannot always be anticipated by the designer. Their evaluation often presents various aspects that require a careful, methodological approach as well as the adoption of suitable tools, needed to identify critical overheads that may impact the overall system performance, stability, validity and scalability. In this paper, we propose a novel approach to evaluating complex mobile agent systems based on a hybrid framework which allows the execution of prototype agent code over simulated internet-works. In this way it is possible to realize arbitrarily complex MA systems and evaluate them over arbitrarily complex inter-networks, relying on full support to physical, link, network and transport layers for fixed and mobile networks. We illustrate the potential of our approach through an example agent system which we have prototyped and assessed over large-scale IP networks. |
| 394 An Innovative Model for Quality Optimization |
pp. 3941-3946 |
|
|
by : Vrasidas Leopoulos, Catherine Poustourli
Abstract: - This paper presents a theoretical analysis and a case study based on Taguchi’s model in quality engineering. The model promotes a holistic view of quality. Quality related cost and money losses, are not considered just for the manufacturer at the time of production, but for the consumer and to society as a whole. The model determines the response characteristics (measurable and quality characteristics), separates factors which affect the product/process response and classify these factors. The loss function is measured by the deviation from the ideal value. Techniques like Orthogonal Arrays have been developed to reduce the elements of (product) variation around the (product) mean in Total Loss Function. The model gives an efficient way of designing experiments for industrial problems and provides a tool for optimizing manufacturing processes. Case studies are included from a furniture industry and a wood-particleboard manufacturing. |
| 405 Sensor System with Accelerometer for Tilt Measurement |
pp. 4051-4054 |
|
|
by : Miroslav Husak
Abstract: - Presented paper describes design and realization of a sensor system for tilt measurement in x and y axes directions in the range of 680?. For construction non-standard connection with accelerometer is used. Measurement is controlled by a microprocessor, measured data are sent to a PC via RS232. The PC is used for data storage, further calculations and display of measured tilt. Convertor of logic levels is used for communication of the sensor system with the PC via RS232 bus. X-ANALYZE in LabView development environment is utilized for service program. Realized sensor system has measurement precision of 0.5? for each axis in the range of 680?, not measurable hysteresis, immediate response to tilt change. |