Pub Date : 1994-04-19DOI: 10.1109/ICASSP.1994.389673
H. Meana, L. N. D. Rivera, M. Nakano-Miyatake, F. Casco-Sanchez, J. Sánchez-García
This paper proposes a time varying normalized LMS (TVS-NLMS) algorithm for adapting an echo canceler structure. Proposed algorithm reduces distortion during double-talk, without increasing the computational cost nor decreasing the convergence rate of the normalized LMS algorithm significantly. Simulation results confirm the desirable features of the proposed schema.<>
{"title":"A time varying step size normalized LMS echo canceler algorithm","authors":"H. Meana, L. N. D. Rivera, M. Nakano-Miyatake, F. Casco-Sanchez, J. Sánchez-García","doi":"10.1109/ICASSP.1994.389673","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389673","url":null,"abstract":"This paper proposes a time varying normalized LMS (TVS-NLMS) algorithm for adapting an echo canceler structure. Proposed algorithm reduces distortion during double-talk, without increasing the computational cost nor decreasing the convergence rate of the normalized LMS algorithm significantly. Simulation results confirm the desirable features of the proposed schema.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126674433","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389838
Jonggil Lee
In a Doppler weather radar, high resolution windspeed profile measurements are needed to provide reliable detection of a hazardous weather condition. For this purpose, the pulse pair method is generally considered to be the most efficient estimator. However, this estimator has some bias errors due to nonsymmetric spectra and may yield meaningless results in the case of a multimodal return spectrum. The bias errors were analyzed and an improved method was suggested to reduce these errors. For the case of a multimodal or seriously skewed spectrum, the modes of the spectrum may provide more reliable information than the statistical mean. Therefore, the idea of a relatively simple mode estimator is also developed.<>
{"title":"Robust estimation of mean Doppler frequency for the measurement of average wind velocity in a weather radar","authors":"Jonggil Lee","doi":"10.1109/ICASSP.1994.389838","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389838","url":null,"abstract":"In a Doppler weather radar, high resolution windspeed profile measurements are needed to provide reliable detection of a hazardous weather condition. For this purpose, the pulse pair method is generally considered to be the most efficient estimator. However, this estimator has some bias errors due to nonsymmetric spectra and may yield meaningless results in the case of a multimodal return spectrum. The bias errors were analyzed and an improved method was suggested to reduce these errors. For the case of a multimodal or seriously skewed spectrum, the modes of the spectrum may provide more reliable information than the statistical mean. Therefore, the idea of a relatively simple mode estimator is also developed.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126675151","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389218
Jean-Marc Boite, H. Bourlard, B. D'hoore, S. Accaino, Johan Vantieghem
Compares speaker independent isolated word recognition performance obtained with standard phonemic hidden Markov models (HMMs) and hybrid approaches using a multilayer perceptron (MLP) to estimate the HMM emission probabilities. This latter approach has previously been shown particularly effective on a large vocabulary, speaker independent, continuous speech recognition task (i.e., ARPA Resource Management) by using simple context-independent phoneme models and single pronunciation word models. As a consequence, the main goal of the paper is to compare the performance which can be achieved by the different approaches for both task dependent and independent training.<>
{"title":"Task independent and dependent training: performance comparison of HMM and hybrid HMM/MLP approaches","authors":"Jean-Marc Boite, H. Bourlard, B. D'hoore, S. Accaino, Johan Vantieghem","doi":"10.1109/ICASSP.1994.389218","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389218","url":null,"abstract":"Compares speaker independent isolated word recognition performance obtained with standard phonemic hidden Markov models (HMMs) and hybrid approaches using a multilayer perceptron (MLP) to estimate the HMM emission probabilities. This latter approach has previously been shown particularly effective on a large vocabulary, speaker independent, continuous speech recognition task (i.e., ARPA Resource Management) by using simple context-independent phoneme models and single pronunciation word models. As a consequence, the main goal of the paper is to compare the performance which can be achieved by the different approaches for both task dependent and independent training.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123187111","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389886
D. Tufts, A. A. Shah
In this paper we present a nested sequence of hypothesis tests for determining the number of signals. The procedure is based on the interpretation of the sum of singular values, of the backward/forward-backward matrix as energy in a particular subspace. The threshold selected is based on a user-specified significance level. Using matrix perturbation ideas, we provide an approximation to the probability of overestimation.<>
{"title":"Rank determination in time-series analysis","authors":"D. Tufts, A. A. Shah","doi":"10.1109/ICASSP.1994.389886","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389886","url":null,"abstract":"In this paper we present a nested sequence of hypothesis tests for determining the number of signals. The procedure is based on the interpretation of the sum of singular values, of the backward/forward-backward matrix as energy in a particular subspace. The threshold selected is based on a user-specified significance level. Using matrix perturbation ideas, we provide an approximation to the probability of overestimation.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123514781","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389863
M. Usman, A. Hero
Computation of the Cramer-Rao bound involves inversion of the Fisher information matrix (FIM). The inversion can become computationally intractable when the number of unknown parameters is large. Hero et. al. (see IEEE Nuclear Science Symposium and Medical Imaging Conference, Orlando, 1983) has presented a recursive, monotonically convergent and computationally efficient algorithm to invert sub-matrices of the FIM corresponding to a small region of interest in image reconstruction. The convergence rate of this algorithm depends on a splitting matrix which can be interpreted as a complete-data FIM. We investigate the acceleration of the algorithm using several different choices of the complete-data FIM. We also present a conjugate gradient based algorithm which achieves a much faster convergence rate at the expense of monotone convergence. We apply the methods developed in this paper to emission tomography.<>
{"title":"Recursive CR bounds: algebraic and statistical acceleration","authors":"M. Usman, A. Hero","doi":"10.1109/ICASSP.1994.389863","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389863","url":null,"abstract":"Computation of the Cramer-Rao bound involves inversion of the Fisher information matrix (FIM). The inversion can become computationally intractable when the number of unknown parameters is large. Hero et. al. (see IEEE Nuclear Science Symposium and Medical Imaging Conference, Orlando, 1983) has presented a recursive, monotonically convergent and computationally efficient algorithm to invert sub-matrices of the FIM corresponding to a small region of interest in image reconstruction. The convergence rate of this algorithm depends on a splitting matrix which can be interpreted as a complete-data FIM. We investigate the acceleration of the algorithm using several different choices of the complete-data FIM. We also present a conjugate gradient based algorithm which achieves a much faster convergence rate at the expense of monotone convergence. We apply the methods developed in this paper to emission tomography.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116322244","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389496
J. Nieweglowski, Timo Moisala, P. Haavisto
A new method for temporal interpolation of video sequences is proposed. For each pair of original frames a new frame is created by averaging frames obtained by motion compensated warping of original frames. The parameters of the geometrical transformation applied to original frames are determined by the motion in the sequence. The advantage of the proposed method is that it can compensate for various motion types such as translations, rotations, and scaling. The primary application of the presented algorithm is the frame interpolation of video sequences coded at very low bitrates where the warping scheme ran also he used for motion compensated prediction. In that case the decoder does not have to perform motion estimation but can use motion vectors from encoder for frame interpolation.<>
{"title":"Motion compensated video sequence interpolation using digital image warping","authors":"J. Nieweglowski, Timo Moisala, P. Haavisto","doi":"10.1109/ICASSP.1994.389496","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389496","url":null,"abstract":"A new method for temporal interpolation of video sequences is proposed. For each pair of original frames a new frame is created by averaging frames obtained by motion compensated warping of original frames. The parameters of the geometrical transformation applied to original frames are determined by the motion in the sequence. The advantage of the proposed method is that it can compensate for various motion types such as translations, rotations, and scaling. The primary application of the presented algorithm is the frame interpolation of video sequences coded at very low bitrates where the warping scheme ran also he used for motion compensated prediction. In that case the decoder does not have to perform motion estimation but can use motion vectors from encoder for frame interpolation.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"140 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116506044","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389683
M. Niranjan, I. Cox, S. L. Hingorani
We report on an approach to recursively track parameters of a cascade formant model. The work follows from that of Rigoll (1986) who showed how an extended Kalman filter (EKF) may be used for recursive estimation of formants. The success of this approach depends on our ability to tune the model noise variances properly. The approach also fails when there is a mismatch between the complexity of the data and that of the model (i.e. wrong number of formants). We show how a multiple model (MM) approach may be used to overcome these problems. We run several models in parallel and use the innovation probabilities of the EKF to recursively evaluate the likelihoods of each of the models. Experimental results demonstrate the feasibility of the approach; accurate switching between models and good tracking of the formants is achieved.<>
{"title":"Recursive tracking of formants in speech signals","authors":"M. Niranjan, I. Cox, S. L. Hingorani","doi":"10.1109/ICASSP.1994.389683","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389683","url":null,"abstract":"We report on an approach to recursively track parameters of a cascade formant model. The work follows from that of Rigoll (1986) who showed how an extended Kalman filter (EKF) may be used for recursive estimation of formants. The success of this approach depends on our ability to tune the model noise variances properly. The approach also fails when there is a mismatch between the complexity of the data and that of the model (i.e. wrong number of formants). We show how a multiple model (MM) approach may be used to overcome these problems. We run several models in parallel and use the innovation probabilities of the EKF to recursively evaluate the likelihoods of each of the models. Experimental results demonstrate the feasibility of the approach; accurate switching between models and good tracking of the formants is achieved.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116516864","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389912
L. Zadeh
There has been a rapid growth in the number and variety of applications of fuzzy logic. The successes of fuzzy logic have also generated a skeptical reaction. Most of the criticisms directed at fuzzy logic are rooted in a misunderstanding of what it is and/or a lack of familiarity with it. In many cases, what is not recognized is that the term fuzzy logic (FL) is actually used in two different senses. In a narrow sense, fuzzy logic (FLn) is a logical system which is an extension of multivalued logic and is intended to serve as a logic of approximate reasoning. But in a wider sense, fuzzy logic (FLw) is more or less synonymous with the theory of fuzzy sets (FST). Today the term fuzzy logic is used predominantly in its wider sense. It is in this sense that any field X can be fuzzified-and hence generalized by replacing the concept of a crisp set in X by a fuzzy set. What is gained through fuzzification is greater generality, higher expressive power, an enhanced ability to model real-world phenomena and a methodology for exploiting the tolerance for imprecision. Most of the applications of fuzzy logic relate to control in the context of industrial systems and consumer products. What is discernible, however, is (a) the trend toward the use of fuzzy logic in task-oriented-rather than set-point-oriented-control; and (b) the incorporation of fuzzy logic and neural network techniques in the conception and design of complex systems in which control and expert system techniques are used in combination.<>
{"title":"Fuzzy logic: issues, contentions and perspectives","authors":"L. Zadeh","doi":"10.1109/ICASSP.1994.389912","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389912","url":null,"abstract":"There has been a rapid growth in the number and variety of applications of fuzzy logic. The successes of fuzzy logic have also generated a skeptical reaction. Most of the criticisms directed at fuzzy logic are rooted in a misunderstanding of what it is and/or a lack of familiarity with it. In many cases, what is not recognized is that the term fuzzy logic (FL) is actually used in two different senses. In a narrow sense, fuzzy logic (FLn) is a logical system which is an extension of multivalued logic and is intended to serve as a logic of approximate reasoning. But in a wider sense, fuzzy logic (FLw) is more or less synonymous with the theory of fuzzy sets (FST). Today the term fuzzy logic is used predominantly in its wider sense. It is in this sense that any field X can be fuzzified-and hence generalized by replacing the concept of a crisp set in X by a fuzzy set. What is gained through fuzzification is greater generality, higher expressive power, an enhanced ability to model real-world phenomena and a methodology for exploiting the tolerance for imprecision. Most of the applications of fuzzy logic relate to control in the context of industrial systems and consumer products. What is discernible, however, is (a) the trend toward the use of fuzzy logic in task-oriented-rather than set-point-oriented-control; and (b) the incorporation of fuzzy logic and neural network techniques in the conception and design of complex systems in which control and expert system techniques are used in combination.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"31 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114007263","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389901
A. Zoubir
We review bootstrap methods for testing statistical hypotheses. We begin with single hypothesis testing and give an example of its use. We then consider multiple testing based on Holm's step-down procedure and discuss how the bootstrap can incorporate relevant correlation and distributional characteristics. These techniques are applied to tests for zeros of transfer functions of linear systems at different frequencies, conventionally solved using "large sample" approximations. Our results highlight the applicability of the bootstrap in a situation where asymptotic results are not valid.<>
{"title":"Multiple bootstrap tests and their application","authors":"A. Zoubir","doi":"10.1109/ICASSP.1994.389901","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389901","url":null,"abstract":"We review bootstrap methods for testing statistical hypotheses. We begin with single hypothesis testing and give an example of its use. We then consider multiple testing based on Holm's step-down procedure and discuss how the bootstrap can incorporate relevant correlation and distributional characteristics. These techniques are applied to tests for zeros of transfer functions of linear systems at different frequencies, conventionally solved using \"large sample\" approximations. Our results highlight the applicability of the bootstrap in a situation where asymptotic results are not valid.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114849230","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 : 1994-04-19DOI: 10.1109/ICASSP.1994.389605
S. Bernadas, C. Thompson
An 8-channel decimation filter, cascadable to 1024 channels, for use with a tri-level delta-sigma modulator (DSM) is described. The filter incorporates a least mean squares (LMS) adaptive correction algorithm to digitally correct the tri-level DSM input data. Additionally, digital offset and gain calibration features and output limiting circuitry reduce the requirements on the analog input to the DSM. The decimation ratio is 128 to 1 and the output data is 22 bits with an integrated floor of -116 dB. The filter coefficient design procedure was unique requiring both impulse response and frequency response optimization. LMS correction can improve the Signal-Noise Ratio by 40 dB.<>
{"title":"A multi-channel decimator for a tri-level delta-sigma modulator","authors":"S. Bernadas, C. Thompson","doi":"10.1109/ICASSP.1994.389605","DOIUrl":"https://doi.org/10.1109/ICASSP.1994.389605","url":null,"abstract":"An 8-channel decimation filter, cascadable to 1024 channels, for use with a tri-level delta-sigma modulator (DSM) is described. The filter incorporates a least mean squares (LMS) adaptive correction algorithm to digitally correct the tri-level DSM input data. Additionally, digital offset and gain calibration features and output limiting circuitry reduce the requirements on the analog input to the DSM. The decimation ratio is 128 to 1 and the output data is 22 bits with an integrated floor of -116 dB. The filter coefficient design procedure was unique requiring both impulse response and frequency response optimization. LMS correction can improve the Signal-Noise Ratio by 40 dB.<<ETX>>","PeriodicalId":290798,"journal":{"name":"Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124451323","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：