Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275836
P. Arató, B. Csák
We present a methodology of how application specific hardware is defined using slightly limited C-language source code. The method is based on corresponding state-machines to each C-language statement, operation, and variable access and chain them by handshaking signals. Optimization steps make high-level parallelism possible without special instructions. Allocation and scheduling make a final compromise between speeds and cost possible. A signal processing example shows the features.
{"title":"Programming language based definition of application oriented hardware","authors":"P. Arató, B. Csák","doi":"10.1109/ISP.2003.1275836","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275836","url":null,"abstract":"We present a methodology of how application specific hardware is defined using slightly limited C-language source code. The method is based on corresponding state-machines to each C-language statement, operation, and variable access and chain them by handshaking signals. Optimization steps make high-level parallelism possible without special instructions. Allocation and scheduling make a final compromise between speeds and cost possible. A signal processing example shows the features.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129344642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275819
L. Beheim, A. Zitouni, F. Belloir
The structural pattern recognition passes by an extraction stage of a certain number of characteristic features of the form. In the majority of the cases, it is not necessary to use the whole of the primitives extracted to obtain good performances of the recognition system. One uses the criteria of feature selection like Fisher or the criteria based on covariance matrix to determine the optimal primitives which characterize the best the form. We will show that these criteria give the most discriminating primitives but not necessarily the most optimal for a given classifier.
{"title":"Problem of optimal pertinent parameter selection in buried conductive tag recognition","authors":"L. Beheim, A. Zitouni, F. Belloir","doi":"10.1109/ISP.2003.1275819","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275819","url":null,"abstract":"The structural pattern recognition passes by an extraction stage of a certain number of characteristic features of the form. In the majority of the cases, it is not necessary to use the whole of the primitives extracted to obtain good performances of the recognition system. One uses the criteria of feature selection like Fisher or the criteria based on covariance matrix to determine the optimal primitives which characterize the best the form. We will show that these criteria give the most discriminating primitives but not necessarily the most optimal for a given classifier.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115548305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275844
J. Klink
We discuss a problem of data transfer in large measurement, diagnostics and control systems (MDCS) over telecommunication networks. We give a proposal of using integrated services digital network for such kind of transfers and present the selected set of services of this network, which allow doing it in the best/cost-effective way.
{"title":"Remote data transfer in distributed measurement, diagnostics and control systems","authors":"J. Klink","doi":"10.1109/ISP.2003.1275844","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275844","url":null,"abstract":"We discuss a problem of data transfer in large measurement, diagnostics and control systems (MDCS) over telecommunication networks. We give a proposal of using integrated services digital network for such kind of transfers and present the selected set of services of this network, which allow doing it in the best/cost-effective way.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114061842","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275831
P. Berényi, G. Horváth
We discuss the online identification of nonlocal static hysteresis functions, which are encountered in mechanical friction, magnetic materials, and piezoelectric actuators and causes problems by the design of controllers. We introduce a compensation method for friction in presliding regime based on the simplified Leuven friction model and on technology borrowed from neural networks. We present a solution how to identify the hysteresis caused by the friction and how to use this identified model for the compensation of the friction effects. Results from both simulations and experiments are shown.
{"title":"Neural networks for nonlocal hysteresis function identification and compensation","authors":"P. Berényi, G. Horváth","doi":"10.1109/ISP.2003.1275831","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275831","url":null,"abstract":"We discuss the online identification of nonlocal static hysteresis functions, which are encountered in mechanical friction, magnetic materials, and piezoelectric actuators and causes problems by the design of controllers. We introduce a compensation method for friction in presliding regime based on the simplified Leuven friction model and on technology borrowed from neural networks. We present a solution how to identify the hysteresis caused by the friction and how to use this identified model for the compensation of the friction effects. Results from both simulations and experiments are shown.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131340117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275853
B. Csukás, G. Bánkuti
The idea of the direct computer mapping (DCM) of process models, and the recently developed generic, bilayered net (GBN) model have been introduced. The method gives a common framework for the simulation of the hybrid (discrete & continuous, quantitative & qualitative) conservation and information processes. The GBN/DCM framework makes possible a new interpretation of the information processes, as well as the development of model based parallel SW/HW architectures. The application of the simulation method for intelligent signal processing is illustrated by the example of the model based diagnostics and control of a simulated moving bed unit. This method is based on the use of the simplest process model of the weakly isomorphic network structure, which learns from the detailed simulation trials, continuously.
{"title":"Hybrid, quantitative and qualitative process simulation by the generic bi-layered net model","authors":"B. Csukás, G. Bánkuti","doi":"10.1109/ISP.2003.1275853","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275853","url":null,"abstract":"The idea of the direct computer mapping (DCM) of process models, and the recently developed generic, bilayered net (GBN) model have been introduced. The method gives a common framework for the simulation of the hybrid (discrete & continuous, quantitative & qualitative) conservation and information processes. The GBN/DCM framework makes possible a new interpretation of the information processes, as well as the development of model based parallel SW/HW architectures. The application of the simulation method for intelligent signal processing is illustrated by the example of the model based diagnostics and control of a simulated moving bed unit. This method is based on the use of the simplest process model of the weakly isomorphic network structure, which learns from the detailed simulation trials, continuously.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132589132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275809
O. Takács, A. Várkonyi-Kóczy
With the help of the SVD-based (singular value decomposition) complexity reduction method, not only the redundancy of fuzzy rule-bases are eliminated, but also further, nonexact reduction are made, considering the allowable error. Namely, in case of higher allowable error, the result is a less complex fuzzy inference system, with a smaller rule-base. This property of the SVD-based reduction method makes possible the usage of fuzzy systems in time-critical applications and makes possible the combining of fuzzy systems with anytime techniques to cope with the changing circumstances during the operation of the system. However, while the SVD-based reduction can be applied to PSGS fuzzy systems, in case of rule-bases, constructed from expert knowledge, the input fuzzy sets are not always in Ruspini-partition. This paper extends the SVD-based reduction to "near PSGS" fuzzy systems, where the input fuzzy sets are not in Ruspini-partition.
{"title":"SVD-based complexity reduction of \"near PSGS\" fuzzy systems","authors":"O. Takács, A. Várkonyi-Kóczy","doi":"10.1109/ISP.2003.1275809","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275809","url":null,"abstract":"With the help of the SVD-based (singular value decomposition) complexity reduction method, not only the redundancy of fuzzy rule-bases are eliminated, but also further, nonexact reduction are made, considering the allowable error. Namely, in case of higher allowable error, the result is a less complex fuzzy inference system, with a smaller rule-base. This property of the SVD-based reduction method makes possible the usage of fuzzy systems in time-critical applications and makes possible the combining of fuzzy systems with anytime techniques to cope with the changing circumstances during the operation of the system. However, while the SVD-based reduction can be applied to PSGS fuzzy systems, in case of rule-bases, constructed from expert knowledge, the input fuzzy sets are not always in Ruspini-partition. This paper extends the SVD-based reduction to \"near PSGS\" fuzzy systems, where the input fuzzy sets are not in Ruspini-partition.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131656534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275833
J. Watada
Our objective is to employ Dempster-Shafer theory into analyzing the hierarchical and logical structure with logical operations AND & OR such as a fault tree. One of main differences of the Dempster-Shafer theory from Baysian one is that it enables us to take the lack of knowledge or information into account in its analyses. The considered set of hypothesis must be mutual exclusive and exhaustive in the Dempster-Shafer theory. We define state patterns in order to satisfy the mutual exclusiveness and exhaustiveness. The method using state patterns requires us much computational time to solve the evidential reasoning. In order to overcome this difficulty, we also propose the improved method in terms of logical relations of the events in the fault tree.
{"title":"Logical computation of evidential reasoning in a hierarchical structure","authors":"J. Watada","doi":"10.1109/ISP.2003.1275833","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275833","url":null,"abstract":"Our objective is to employ Dempster-Shafer theory into analyzing the hierarchical and logical structure with logical operations AND & OR such as a fault tree. One of main differences of the Dempster-Shafer theory from Baysian one is that it enables us to take the lack of knowledge or information into account in its analyses. The considered set of hypothesis must be mutual exclusive and exhaustive in the Dempster-Shafer theory. We define state patterns in order to satisfy the mutual exclusiveness and exhaustiveness. The method using state patterns requires us much computational time to solve the evidential reasoning. In order to overcome this difficulty, we also propose the improved method in terms of logical relations of the events in the fault tree.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125217147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275848
R. Poluzzi, L. Arnone, A. Savi, M. Brescianini
In order to enhance the percentage of recognised words for ASR (automatic speech recognition) systems, an adaptive neuro-fuzzy filtering technique in the time domain is proposed. Architecture of neural networks is shown having the purpose of exploiting spatial information, with the advantage of little computational load with respect to other classical beamforming techniques. Several results are described for complex acoustic scenarios with VOX ASR system by DIBE (University of Genova).
{"title":"Neuro-fuzzy filtering techniques for automatic speech recognition enhancement","authors":"R. Poluzzi, L. Arnone, A. Savi, M. Brescianini","doi":"10.1109/ISP.2003.1275848","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275848","url":null,"abstract":"In order to enhance the percentage of recognised words for ASR (automatic speech recognition) systems, an adaptive neuro-fuzzy filtering technique in the time domain is proposed. Architecture of neural networks is shown having the purpose of exploiting spatial information, with the advantage of little computational load with respect to other classical beamforming techniques. Several results are described for complex acoustic scenarios with VOX ASR system by DIBE (University of Genova).","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"45 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114130635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275834
X. Wang, D. Brown, T. Haynes, T.M.J. Hui
A weighted optimisation method based on the AdaBoost algorithm is proposed and used in neural network incremental construction. Compared to the traditional gradient-based method, it has the advantage of being easy to implement and are applied where the cost function is not smooth. The experimental results are included.
{"title":"A new method in incremental neural network construction by using boosting","authors":"X. Wang, D. Brown, T. Haynes, T.M.J. Hui","doi":"10.1109/ISP.2003.1275834","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275834","url":null,"abstract":"A weighted optimisation method based on the AdaBoost algorithm is proposed and used in neural network incremental construction. Compared to the traditional gradient-based method, it has the advantage of being easy to implement and are applied where the cost function is not smooth. The experimental results are included.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132408295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-09-04DOI: 10.1109/ISP.2003.1275842
L. Balogh, I. Kollár
The angle of arrival estimation problem based on interferometer principle has great accuracy in the angle of arrival estimation. To increase the accuracy there are two main ways. In the first one we add more antennas to the configuration, hence we get more measurement data, but it means that the physical configuration has to be changed. In the second method we design numerical algorithms, which gives better estimations. The prize is the complexity and the increased computation time. A method from the second set is shown and analysed. The algorithm is based on the cross correlation technique.
{"title":"Angle of arrival estimation based on interferometer principle","authors":"L. Balogh, I. Kollár","doi":"10.1109/ISP.2003.1275842","DOIUrl":"https://doi.org/10.1109/ISP.2003.1275842","url":null,"abstract":"The angle of arrival estimation problem based on interferometer principle has great accuracy in the angle of arrival estimation. To increase the accuracy there are two main ways. In the first one we add more antennas to the configuration, hence we get more measurement data, but it means that the physical configuration has to be changed. In the second method we design numerical algorithms, which gives better estimations. The prize is the complexity and the increased computation time. A method from the second set is shown and analysed. The algorithm is based on the cross correlation technique.","PeriodicalId":285893,"journal":{"name":"IEEE International Symposium on Intelligent Signal Processing, 2003","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125913024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：