Pub Date : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296479
H. Seddik, A. Rahmouni, M. Sayadi
Modern speaker recognition applications require high accuracy at low complexity and easy calculation. In this paper, we propose a new method of text independent speaker recognition based on the use of the mean of the Mel frequency cepstral coefficients (MFCC) as a speaker model. These MFCC are extracted from the speaker phonemes in the pre-segmented speech sentences. A multi-layer neural network trained with the back propagation algorithm is proposed to classify these discriminative models. A study is carried out in order to view these models efficiency. Several experiments are made and show that the proposed method gives a high speaker recognition rate. Furthermore, throw these experiments; a technique is proposed to improve this recognition rate by an appropriate phonemes database selection.
{"title":"Text independent speaker recognition using the Mel frequency cepstral coefficients and a neural network classifier","authors":"H. Seddik, A. Rahmouni, M. Sayadi","doi":"10.1109/ISCCSP.2004.1296479","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296479","url":null,"abstract":"Modern speaker recognition applications require high accuracy at low complexity and easy calculation. In this paper, we propose a new method of text independent speaker recognition based on the use of the mean of the Mel frequency cepstral coefficients (MFCC) as a speaker model. These MFCC are extracted from the speaker phonemes in the pre-segmented speech sentences. A multi-layer neural network trained with the back propagation algorithm is proposed to classify these discriminative models. A study is carried out in order to view these models efficiency. Several experiments are made and show that the proposed method gives a high speaker recognition rate. Furthermore, throw these experiments; a technique is proposed to improve this recognition rate by an appropriate phonemes database selection.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133327450","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296247
S. Benedetto
Summary form only given. This paper presents in a tutorial from a survey of the recent, powerful error correcting coding techniques know as "turbo-like" codes. Based on the concatenation of two convolutional encoders and an interleaver, and endowed with a relatively simple iterative decoding algorithm, these codes have revolutionized the coding field, showing that performance very close to the Shannon theoretical capacity limits with limited hardware complexity. The presentation focuses on the analysis, design, and applications of this important class of codes.
{"title":"Advanced error correcting coding techniques and iterative decoding","authors":"S. Benedetto","doi":"10.1109/ISCCSP.2004.1296247","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296247","url":null,"abstract":"Summary form only given. This paper presents in a tutorial from a survey of the recent, powerful error correcting coding techniques know as \"turbo-like\" codes. Based on the concatenation of two convolutional encoders and an interleaver, and endowed with a relatively simple iterative decoding algorithm, these codes have revolutionized the coding field, showing that performance very close to the Shannon theoretical capacity limits with limited hardware complexity. The presentation focuses on the analysis, design, and applications of this important class of codes.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133730016","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296480
B. brahim, K. Ouni, N. Ellouze
In this paper we present a study of the temporal and spectral human auditory masking phenomena for speech signal analysis. For the purpose of modeling these masking phenomena, we used a gammachirp filterbank [T. Irino and M. Unoki, Nov. 1999], [T. Irino, 1999] to model the spectral masking and a temporal window to model the temporal masking. A global model combining these two models was built for a spectro-temporal representation. We performed a comparison of two types of spectro-temporal representations called gammagrams. The first one is based only on a gammachirp filterbank and the second is based on the global model. In addition, we performed some series of tests on different speech signals for establishing examples of masking effect curves.
{"title":"A time frequency representations of speech signals based on a modeling of the auditory system: the gammagrams","authors":"B. brahim, K. Ouni, N. Ellouze","doi":"10.1109/ISCCSP.2004.1296480","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296480","url":null,"abstract":"In this paper we present a study of the temporal and spectral human auditory masking phenomena for speech signal analysis. For the purpose of modeling these masking phenomena, we used a gammachirp filterbank [T. Irino and M. Unoki, Nov. 1999], [T. Irino, 1999] to model the spectral masking and a temporal window to model the temporal masking. A global model combining these two models was built for a spectro-temporal representation. We performed a comparison of two types of spectro-temporal representations called gammagrams. The first one is based only on a gammachirp filterbank and the second is based on the global model. In addition, we performed some series of tests on different speech signals for establishing examples of masking effect curves.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115607416","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296341
C. Watagodakumbura, A. Jennings, N. Shenoy
Differentiated services (DiffServ) architecture is based on aggregation of traffic as opposed to per flow traffic. When DiffServ is used in a real time traffic environment, it is appealing to study absolute effects of traffic aggregation on quality of service parameters. That is, effects are obtained when a burstiness controller such as a token bucket is not used at the entrance of the network. One-way delay and instantaneous packet delay variation (IPDV) are the most commonly used quality of service parameters for real time traffic. Also, fractal or self-similar nature of the Internet traffic has been identified in recent years. In this paper, we study how one-way delay and IPDV are affected by aggregation of self-similar traffic. We make quantitative comparisons of IPDV and one-way delay between aggregated self-similar traffic and exponential inter-arrival traffic using simulations. The only traffic control used at the entrance of the network is an aggregated real time traffic utilization threshold. Further, we bring out the notion of "level of active sources" to explain the statistical multiplexing gains observed in delay variation.
{"title":"Absolute effects of aggregation of self-similar traffic on quality of service parameters","authors":"C. Watagodakumbura, A. Jennings, N. Shenoy","doi":"10.1109/ISCCSP.2004.1296341","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296341","url":null,"abstract":"Differentiated services (DiffServ) architecture is based on aggregation of traffic as opposed to per flow traffic. When DiffServ is used in a real time traffic environment, it is appealing to study absolute effects of traffic aggregation on quality of service parameters. That is, effects are obtained when a burstiness controller such as a token bucket is not used at the entrance of the network. One-way delay and instantaneous packet delay variation (IPDV) are the most commonly used quality of service parameters for real time traffic. Also, fractal or self-similar nature of the Internet traffic has been identified in recent years. In this paper, we study how one-way delay and IPDV are affected by aggregation of self-similar traffic. We make quantitative comparisons of IPDV and one-way delay between aggregated self-similar traffic and exponential inter-arrival traffic using simulations. The only traffic control used at the entrance of the network is an aggregated real time traffic utilization threshold. Further, we bring out the notion of \"level of active sources\" to explain the statistical multiplexing gains observed in delay variation.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114578050","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296338
S. Poyhonen, P. Jover, H. Hyotyniemi
Vibration monitoring is studied for fault diagnostics of an induction motor. Several features of vibration signals are compared as indicators of broken rotor bar of a 35 kW induction motor. Regular fast Fourier transform (FFT) based power spectrum density (PSD) estimation is compared to signal processing with higher order spectra (HOS), cepstrum analysis and signal description with autoregressive (AR) modelling. The fault detection routine and feature comparison is carried out with support vector machine (SVM) based classification. The best method for feature extraction seems to be the application of AR coefficients. The result is found out with real measurement data from several motor conditions and load situations.
{"title":"Signal processing of vibrations for condition monitoring of an induction motor","authors":"S. Poyhonen, P. Jover, H. Hyotyniemi","doi":"10.1109/ISCCSP.2004.1296338","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296338","url":null,"abstract":"Vibration monitoring is studied for fault diagnostics of an induction motor. Several features of vibration signals are compared as indicators of broken rotor bar of a 35 kW induction motor. Regular fast Fourier transform (FFT) based power spectrum density (PSD) estimation is compared to signal processing with higher order spectra (HOS), cepstrum analysis and signal description with autoregressive (AR) modelling. The fault detection routine and feature comparison is carried out with support vector machine (SVM) based classification. The best method for feature extraction seems to be the application of AR coefficients. The result is found out with real measurement data from several motor conditions and load situations.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125146481","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296333
C. Seddik, F. Fnaiech
This paper is concerned by the use of neural networks and fuzzy logic for controlling a non-linear process namely an induction machine. In the first case study, the design procedure uses a neural model trained with the inverse model of the process. Thus, the overall controlled system is formed using this inverse model. In the second case study, a fuzzy logic controller is implemented. In both cases, the controller is cascaded with the process ensuring the robustness and the stability of the controlled system regarding parameters uncertainties and disturbances. This work analyses the advantages and the drawbacks of each controller in terms of tracking and regulation. It is shown that the fuzzy logic controller is slightly better with respect to the neural network controller in the transient while they have quite similar behaviour in the steady-state regime.
{"title":"Neural networks and fuzzy nonlinear controllers applied to an induction machine","authors":"C. Seddik, F. Fnaiech","doi":"10.1109/ISCCSP.2004.1296333","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296333","url":null,"abstract":"This paper is concerned by the use of neural networks and fuzzy logic for controlling a non-linear process namely an induction machine. In the first case study, the design procedure uses a neural model trained with the inverse model of the process. Thus, the overall controlled system is formed using this inverse model. In the second case study, a fuzzy logic controller is implemented. In both cases, the controller is cascaded with the process ensuring the robustness and the stability of the controlled system regarding parameters uncertainties and disturbances. This work analyses the advantages and the drawbacks of each controller in terms of tracking and regulation. It is shown that the fuzzy logic controller is slightly better with respect to the neural network controller in the transient while they have quite similar behaviour in the steady-state regime.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129489028","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296212
O. Akay, E. Erozden
In the recent past, fractional cross-correlation and autocorrelation operations associated with the fractional Fourier transform (FrFT) have been formulated. A detection statistic based on fractional autocorrelation has been employed for detection of linear frequency modulated (LFM) signals. In this paper, we extend the utility of the detection statistic based on fractional autocorrelation for detection of pulse compression radar waveforms. In particular, we consider the step LFM signal and the waveforms of Frank code and P4 code.
{"title":"Use of fractional autocorrelation in efficient detection of pulse compression radar signals","authors":"O. Akay, E. Erozden","doi":"10.1109/ISCCSP.2004.1296212","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296212","url":null,"abstract":"In the recent past, fractional cross-correlation and autocorrelation operations associated with the fractional Fourier transform (FrFT) have been formulated. A detection statistic based on fractional autocorrelation has been employed for detection of linear frequency modulated (LFM) signals. In this paper, we extend the utility of the detection statistic based on fractional autocorrelation for detection of pulse compression radar waveforms. In particular, we consider the step LFM signal and the waveforms of Frank code and P4 code.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121356310","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296263
A. Cetin, H. Ozaktus, H.M. Ozaktus
The fractional Fourier transform has found many applications in signal and image processing and optics. An iterative algorithm for signal recovery from partial fractional Fourier transform information is presented. The signal recovery algorithm is constructed by using the method of projections onto convex sets and convergence of the algorithm is assured.
{"title":"Signal recovery from partial fractional Fourier transform information","authors":"A. Cetin, H. Ozaktus, H.M. Ozaktus","doi":"10.1109/ISCCSP.2004.1296263","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296263","url":null,"abstract":"The fractional Fourier transform has found many applications in signal and image processing and optics. An iterative algorithm for signal recovery from partial fractional Fourier transform information is presented. The signal recovery algorithm is constructed by using the method of projections onto convex sets and convergence of the algorithm is assured.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121242776","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296259
Min Dong, L. Tong, B. Sadler
We consider the problem of reconstructing a one-dimensional Gauss Markov field measured by a large-scale sensor network. Two data retrieval strategies are considered: the scheduling that collects data from equally spaced sensors locations and random access. Assuming the sensors in the field form a Poisson field with density /spl rho/, we examine the reconstruction performance of the signal field based on the data retrieved under the two strategies. Our comparison shows that, the performance under the optimal scheduling is sensitive to the outage probability P/sub out/ of sensors in a given region. If P/sub out/ is large than the threshold, the performance of scheduling suffers from missing data samples, and simple random access outperforms optimal scheduling.
{"title":"Optimal reconstruction of Gauss Markov field in large sensor networks","authors":"Min Dong, L. Tong, B. Sadler","doi":"10.1109/ISCCSP.2004.1296259","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296259","url":null,"abstract":"We consider the problem of reconstructing a one-dimensional Gauss Markov field measured by a large-scale sensor network. Two data retrieval strategies are considered: the scheduling that collects data from equally spaced sensors locations and random access. Assuming the sensors in the field form a Poisson field with density /spl rho/, we examine the reconstruction performance of the signal field based on the data retrieved under the two strategies. Our comparison shows that, the performance under the optimal scheduling is sensitive to the outage probability P/sub out/ of sensors in a given region. If P/sub out/ is large than the threshold, the performance of scheduling suffers from missing data samples, and simple random access outperforms optimal scheduling.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125673876","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 : 2004-09-27DOI: 10.1109/ISCCSP.2004.1296321
S. Matta, D.K. Kumar, Xinghuo Yu, M. Burry
This paper presents a new approach for image to sound mapping. The proposed method utilizes the music parameters such as pitch and rhythm to support translation of images into sounds. Many people have tried image-to-sound mapping or data-to-sound mapping and failed to prove the useful results and many people haven't followed the principles of psychoacoustics in implementing image to sound conversion methods. The important bottleneck in these kinds of experiments is that humans can't remember the normal sounds as compared to music. A method is developed to overcome this bottleneck by utilizing musical parameters. Most of the available tools have been tested on the participants and it has been discovered that the technology available to convert data streams into sounds was not sufficient and needed an improvement.
{"title":"An approach for image sonification","authors":"S. Matta, D.K. Kumar, Xinghuo Yu, M. Burry","doi":"10.1109/ISCCSP.2004.1296321","DOIUrl":"https://doi.org/10.1109/ISCCSP.2004.1296321","url":null,"abstract":"This paper presents a new approach for image to sound mapping. The proposed method utilizes the music parameters such as pitch and rhythm to support translation of images into sounds. Many people have tried image-to-sound mapping or data-to-sound mapping and failed to prove the useful results and many people haven't followed the principles of psychoacoustics in implementing image to sound conversion methods. The important bottleneck in these kinds of experiments is that humans can't remember the normal sounds as compared to music. A method is developed to overcome this bottleneck by utilizing musical parameters. Most of the available tools have been tested on the participants and it has been discovered that the technology available to convert data streams into sounds was not sufficient and needed an improvement.","PeriodicalId":146713,"journal":{"name":"First International Symposium on Control, Communications and Signal Processing, 2004.","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129126232","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：