Pub Date : 2003-12-24DOI: 10.1109/ISSPA.2001.949827
S. Marple
Despite the enhanced time-frequency analysis (TFA) detailing capability of quadratic TFAs like the Wigner and Cohen representations, their performance with signals of large dynamic range (DNR in excess of 40 dB) is quite poor due to the inability to totally suppress the cross-term artifacts which typically are much stronger than the weakest signal components that they obscure. This paper presents one of two modifications of linear TFA to provide the enhanced detailing behavior of quadratic TFAs without introducing cross terms, making it possible to see the time-frequency detail of extremely weak signal components. The technique described is based on subspace-enhanced linear predictive extrapolation of the data within each analysis window to create a longer data sequence for conventional short-time Fourier transform (STFT) TFA. The other technique, based on formation of a special two-dimensional transformed data matrix analyzed by high-definition two-dimensional spectral analysis methods such as 2-D AR or 2-D minimum variance, is presented in a separate time-frequency textbook under the development editorship of B. Boashash (see Time-Frequency Signal Analysis and Processing, Prentice Hall, 2002).
{"title":"Large dynamic range time-frequency signal analysis with application to helicopter Doppler radar data","authors":"S. Marple","doi":"10.1109/ISSPA.2001.949827","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.949827","url":null,"abstract":"Despite the enhanced time-frequency analysis (TFA) detailing capability of quadratic TFAs like the Wigner and Cohen representations, their performance with signals of large dynamic range (DNR in excess of 40 dB) is quite poor due to the inability to totally suppress the cross-term artifacts which typically are much stronger than the weakest signal components that they obscure. This paper presents one of two modifications of linear TFA to provide the enhanced detailing behavior of quadratic TFAs without introducing cross terms, making it possible to see the time-frequency detail of extremely weak signal components. The technique described is based on subspace-enhanced linear predictive extrapolation of the data within each analysis window to create a longer data sequence for conventional short-time Fourier transform (STFT) TFA. The other technique, based on formation of a special two-dimensional transformed data matrix analyzed by high-definition two-dimensional spectral analysis methods such as 2-D AR or 2-D minimum variance, is presented in a separate time-frequency textbook under the development editorship of B. Boashash (see Time-Frequency Signal Analysis and Processing, Prentice Hall, 2002).","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126610576","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 : 2002-06-01DOI: 10.1109/ISSPA.2001.949841
M. Ibnkahla
The paper presents a statistical analysis of neural network modeling and identification of nonlinear systems with memory. The nonlinear system model is comprised of a discrete-time linear filter H followed by a zero-memory nonlinear function g(.). The system is corrupted by input and output independent Gaussian noise. The neural network is used to identify and model the unknown linear filter H and the unknown nonlinearity g(.). The network architecture is composed of a linear adaptive filter and a two-layer nonlinear neural network (with an arbitrary number of neurons). The network is trained using the backpropagation algorithm. The paper studies the MSE surface and the stationary points of the adaptive system. Recursions are derived for the mean transient behavior of the adaptive filter coefficients and the neural network weights for slow learning. It is shown that the adaptive filter converges to a scaled version of the unknown filter H, and that the nonlinear neural network converges to an approximation of the unknown nonlinearity. Computer simulations show good agreement between theory and experimental results.
{"title":"Statistical analysis of neural network modeling and identification of nonlinear systems with memory","authors":"M. Ibnkahla","doi":"10.1109/ISSPA.2001.949841","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.949841","url":null,"abstract":"The paper presents a statistical analysis of neural network modeling and identification of nonlinear systems with memory. The nonlinear system model is comprised of a discrete-time linear filter H followed by a zero-memory nonlinear function g(.). The system is corrupted by input and output independent Gaussian noise. The neural network is used to identify and model the unknown linear filter H and the unknown nonlinearity g(.). The network architecture is composed of a linear adaptive filter and a two-layer nonlinear neural network (with an arbitrary number of neurons). The network is trained using the backpropagation algorithm. The paper studies the MSE surface and the stationary points of the adaptive system. Recursions are derived for the mean transient behavior of the adaptive filter coefficients and the neural network weights for slow learning. It is shown that the adaptive filter converges to a scaled version of the unknown filter H, and that the nonlinear neural network converges to an approximation of the unknown nonlinearity. Computer simulations show good agreement between theory and experimental results.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123991725","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950197
Claudia Schremmer
The wavelet transform has become the most interesting new algorithm for still image compression. Yet there are many parameters within a wavelet analysis and synthesis which govern the quality of a decoded image. In this paper, we discuss different decomposition strategies of a two-dimensional signal and their implications for the decoded image: a pool of gray-scale images has been wave let-transformed with different settings of the wavelet filter bank, quantization threshold and decomposition method. Contrary to the new standard JPEG-2000, where nonstandard decomposition is implemented, our investigation proposes standard decomposition for low-bitrate coding.
{"title":"Decomposition strategies for wavelet-based image coding","authors":"Claudia Schremmer","doi":"10.1109/ISSPA.2001.950197","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950197","url":null,"abstract":"The wavelet transform has become the most interesting new algorithm for still image compression. Yet there are many parameters within a wavelet analysis and synthesis which govern the quality of a decoded image. In this paper, we discuss different decomposition strategies of a two-dimensional signal and their implications for the decoded image: a pool of gray-scale images has been wave let-transformed with different settings of the wavelet filter bank, quantization threshold and decomposition method. Contrary to the new standard JPEG-2000, where nonstandard decomposition is implemented, our investigation proposes standard decomposition for low-bitrate coding.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114985305","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950180
A. Hussain, M. Sukairi, A. Mohamed, R. Mohamed
Many works involving detection and classification of power quality (PQ) events report the use of artificial neural network (ANN) to perform the classification task. No doubt, many have found ANN successfully performs the required task, but the approach requires a long training process and is too rigid if expansion or modification is desired. This paper proposes an alternative approach for the detection of PQ disturbances, which is simple, expandable and does not require training. The proposed system is built and tested using field-measured voltage waveforms, which are made of five types of PQ disturbances, namely, impulsive transient, oscillatory transient, single notch, repetitive notch and voltage sag. It perfectly detects and categorizes all test frames as either "clean" or "not clean", in which the frame labeled as "not clean" consists of some form of PQ disturbances. Results show that the frame identification consisting of impulsive and oscillatory transient disturbances achieved an overall accuracy rate of nearly 95%.
{"title":"Automatic detection of power quality disturbances and identification of transient signals","authors":"A. Hussain, M. Sukairi, A. Mohamed, R. Mohamed","doi":"10.1109/ISSPA.2001.950180","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950180","url":null,"abstract":"Many works involving detection and classification of power quality (PQ) events report the use of artificial neural network (ANN) to perform the classification task. No doubt, many have found ANN successfully performs the required task, but the approach requires a long training process and is too rigid if expansion or modification is desired. This paper proposes an alternative approach for the detection of PQ disturbances, which is simple, expandable and does not require training. The proposed system is built and tested using field-measured voltage waveforms, which are made of five types of PQ disturbances, namely, impulsive transient, oscillatory transient, single notch, repetitive notch and voltage sag. It perfectly detects and categorizes all test frames as either \"clean\" or \"not clean\", in which the frame labeled as \"not clean\" consists of some form of PQ disturbances. Results show that the frame identification consisting of impulsive and oscillatory transient disturbances achieved an overall accuracy rate of nearly 95%.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121131197","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.949786
N. Grbic, Jan Mark de Haan, I. Claesson, S. Nordholm
Subband adaptive filters have been proposed to avoid the drawbacks of slow convergence and high computational complexity associated with time domain adaptive filters for acoustic echo cancellation. Subband processing introduces transmission delays caused by the filter bank and signal degradations due to aliasing effects. One efficient way to reduce the aliasing effects is to allow a higher sample rate than critically needed in the subbands and thus reduce subband signal degradation. We suggest a design method, for a uniform DFT filter bank with any over sampling factor, where the total filter bank group delay may be specified, and where the aliasing and magnitude/phase distortions are minimized.
{"title":"Design of oversampled uniform DFT filter banks with reduced inband aliasing and delay constraints","authors":"N. Grbic, Jan Mark de Haan, I. Claesson, S. Nordholm","doi":"10.1109/ISSPA.2001.949786","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.949786","url":null,"abstract":"Subband adaptive filters have been proposed to avoid the drawbacks of slow convergence and high computational complexity associated with time domain adaptive filters for acoustic echo cancellation. Subband processing introduces transmission delays caused by the filter bank and signal degradations due to aliasing effects. One efficient way to reduce the aliasing effects is to allow a higher sample rate than critically needed in the subbands and thus reduce subband signal degradation. We suggest a design method, for a uniform DFT filter bank with any over sampling factor, where the total filter bank group delay may be specified, and where the aliasing and magnitude/phase distortions are minimized.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"22 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114059004","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.949789
Y. Abramovich, N. Spencer
Nonidentifiability with respect to detection is derived, and is found to be the condition that the covariance matrix of a mixture of a particular number of independent Gaussian sources plus noise is identical to that of another mixture of a different number of such sources. This condition is formulated in terms of the rank-deficiency of some matrix that is related to the antenna geometry's co-array. We propose and discuss a method to determine the detection identifiability or otherwise of any given nonuniform linear antenna array.
{"title":"Detection nonidentifiability for independent Gaussian sources in nonuniform linear antenna arrays","authors":"Y. Abramovich, N. Spencer","doi":"10.1109/ISSPA.2001.949789","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.949789","url":null,"abstract":"Nonidentifiability with respect to detection is derived, and is found to be the condition that the covariance matrix of a mixture of a particular number of independent Gaussian sources plus noise is identical to that of another mixture of a different number of such sources. This condition is formulated in terms of the rank-deficiency of some matrix that is related to the antenna geometry's co-array. We propose and discuss a method to determine the detection identifiability or otherwise of any given nonuniform linear antenna array.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116603773","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950244
Nariman Majdi-Nasab, M. Analoui
There are several methods for analyzing and estimating parameters for mixture models. These approaches seek to optimize various aspects of mixture model estimation, such as accuracy and computation cost. We present a new approach for estimating parameters of a Gaussian mixture model by genetic algorithms (GA). GA are adaptive search techniques designed to find near-optimal solutions of large-scale optimization problems with multiple local maxima. It is shown that using GA can find mixture model parameters accurately and efficiently for noisy and noiseless data sets.
{"title":"Mixture conditional estimation using genetic algorithms","authors":"Nariman Majdi-Nasab, M. Analoui","doi":"10.1109/ISSPA.2001.950244","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950244","url":null,"abstract":"There are several methods for analyzing and estimating parameters for mixture models. These approaches seek to optimize various aspects of mixture model estimation, such as accuracy and computation cost. We present a new approach for estimating parameters of a Gaussian mixture model by genetic algorithms (GA). GA are adaptive search techniques designed to find near-optimal solutions of large-scale optimization problems with multiple local maxima. It is shown that using GA can find mixture model parameters accurately and efficiently for noisy and noiseless data sets.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128530166","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950189
Liang Zhao, M. Amin, A. Lindsey
Frequency modulated signals in the frequency band 1.217-1.238 GHz and 1.565-1.586 GHz present a source of interference to the GPS, which should be properly mitigated. We derive the signal-to-interference-and-noise ratio (SINR) of the GPS receiver implementing subspace projection techniques for suppression of FM jammers. We consider the general case in which the jammer may have equal or different cycles than the coarse acquisition (C/A) code of the GPS signals. It is shown that the weak correlations between the FM interference and the Gold codes allow effective interference cancellation without significant loss of the desired signal.
{"title":"Mitigation of periodic interferers in GPS receivers using subspace projection techniques","authors":"Liang Zhao, M. Amin, A. Lindsey","doi":"10.1109/ISSPA.2001.950189","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950189","url":null,"abstract":"Frequency modulated signals in the frequency band 1.217-1.238 GHz and 1.565-1.586 GHz present a source of interference to the GPS, which should be properly mitigated. We derive the signal-to-interference-and-noise ratio (SINR) of the GPS receiver implementing subspace projection techniques for suppression of FM jammers. We consider the general case in which the jammer may have equal or different cycles than the coarse acquisition (C/A) code of the GPS signals. It is shown that the weak correlations between the FM interference and the Gold codes allow effective interference cancellation without significant loss of the desired signal.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129272431","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950208
A. Sha'ameri
The windowed Wigner-Ville distribution (WWVD) and smooth windowed Wigner-Ville distribution (SWWVD) are used to analyze time-varying signals. Two classes of signals used are the dual-component linear FM and frequency shift-keying (FSK) signals. The time-lag characteristics of the bilinear product of both signals are analyzed and used to setup the parameters of the WWVD and SWWVD. In general, a more accurate time-frequency representation is obtained using the SWWVD compared to the WWVD. The accuracy of the resulting time-frequency representation is quantified based on the mainlobe width (MLW) and signal-to-interference (SNI) ratio.
{"title":"Parameters definition for the windowed and smooth windowed Wigner-Ville distribution of time-varying signals","authors":"A. Sha'ameri","doi":"10.1109/ISSPA.2001.950208","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950208","url":null,"abstract":"The windowed Wigner-Ville distribution (WWVD) and smooth windowed Wigner-Ville distribution (SWWVD) are used to analyze time-varying signals. Two classes of signals used are the dual-component linear FM and frequency shift-keying (FSK) signals. The time-lag characteristics of the bilinear product of both signals are analyzed and used to setup the parameters of the WWVD and SWWVD. In general, a more accurate time-frequency representation is obtained using the SWWVD compared to the WWVD. The accuracy of the resulting time-frequency representation is quantified based on the mainlobe width (MLW) and signal-to-interference (SNI) ratio.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129291592","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 : 1900-01-01DOI: 10.1109/ISSPA.2001.950171
Y. Ye, P. Cosman
Shape unifying is a very efficient preprocessing technique used in lossy SPM-JBIG2 systems. It permits isolated errors between the current bitmap and its reference to improve refinement coding efficiency. Compared to lossless coding, it can improve compression by about 32% while causing very little visual information loss. When bigger error clusters are permitted in shape unifying, further compression gain can be achieved but at the price of more noticeable visual information loss and even character substitution errors. We propose a feature monitored shape unifying procedure that can significantly lower the risk of substitution errors when permitting bigger errors. Experiments show that, compared to the unmonitored shape unifying, the feature monitored version can suppress more than 2/3 of all substitution errors while achieving additional compression improvements of 30-40%.
{"title":"Feature-monitored shape unifying for lossy SPM-JBIG2","authors":"Y. Ye, P. Cosman","doi":"10.1109/ISSPA.2001.950171","DOIUrl":"https://doi.org/10.1109/ISSPA.2001.950171","url":null,"abstract":"Shape unifying is a very efficient preprocessing technique used in lossy SPM-JBIG2 systems. It permits isolated errors between the current bitmap and its reference to improve refinement coding efficiency. Compared to lossless coding, it can improve compression by about 32% while causing very little visual information loss. When bigger error clusters are permitted in shape unifying, further compression gain can be achieved but at the price of more noticeable visual information loss and even character substitution errors. We propose a feature monitored shape unifying procedure that can significantly lower the risk of substitution errors when permitting bigger errors. Experiments show that, compared to the unmonitored shape unifying, the feature monitored version can suppress more than 2/3 of all substitution errors while achieving additional compression improvements of 30-40%.","PeriodicalId":236050,"journal":{"name":"Proceedings of the Sixth International Symposium on Signal Processing and its Applications (Cat.No.01EX467)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124658267","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：