Pub Date : 1994-06-26DOI: 10.1109/SSAP.1994.572528
R. Noumeir, G. Mailloux, R. Lemieux
A bayesian image reconstruction algorithm is proposed for emission tomography. It incorporates the Poisson nature of the noise in the projection data and characterizes the image to be reconstructed by an homogeneous Gauss-Markov process that can be represented by an autoregressive model. The modelling error is assumed to be a zero mean whitenoise process. The expectation maximization method is applied to find the maximum a posteriori (MAP) estimator. Comparisons between the maximum likelihood (ML) algorithm and the MAP algorithm are carried out with a numerical phantom. The porposed algorithm succeeds in overcoming the noise artefact inherent to ML and gives results superior to the best results reached by ML.
{"title":"Bayesian Image Reconstruction: An Application to Emission","authors":"R. Noumeir, G. Mailloux, R. Lemieux","doi":"10.1109/SSAP.1994.572528","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572528","url":null,"abstract":"A bayesian image reconstruction algorithm is proposed for emission tomography. It incorporates the Poisson nature of the noise in the projection data and characterizes the image to be reconstructed by an homogeneous Gauss-Markov process that can be represented by an autoregressive model. The modelling error is assumed to be a zero mean whitenoise process. The expectation maximization method is applied to find the maximum a posteriori (MAP) estimator. Comparisons between the maximum likelihood (ML) algorithm and the MAP algorithm are carried out with a numerical phantom. The porposed algorithm succeeds in overcoming the noise artefact inherent to ML and gives results superior to the best results reached by ML.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"9 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122600003","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572429
C. Ying, L. Potter, R. Moses
We present an algorithm for model order determination and simultaneous maximum likelihood parameter estimation for complex exponential signal modeling. The algorithm exploits initial nonparametric (i.e., FFT) frequency location estimates and Cram&-Rru, Bound (CRB) resolution limits to significantly reduce the search space for the correct model order and parameter estimates. The algorithm initially overestimates the model order. After iterative minimization to obtain maximum likelihood (ML) parameter estimates for that order, a post-processing step eliminates the extraneous sinusoidal modes using CFU3 resolution limits and statistical detection tests. Because the algorithm searches on only a limited set of model orders and parameter regions, it is computationally tractable even for large data lengths and model orders. In this paper we analyze the performance of the algorithm and compare with other existing approaches.
{"title":"On Model Order Determination For Complex Exponential Signals: Performance Of An FFT-initialized ML Algorithm","authors":"C. Ying, L. Potter, R. Moses","doi":"10.1109/SSAP.1994.572429","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572429","url":null,"abstract":"We present an algorithm for model order determination and simultaneous maximum likelihood parameter estimation for complex exponential signal modeling. The algorithm exploits initial nonparametric (i.e., FFT) frequency location estimates and Cram&-Rru, Bound (CRB) resolution limits to significantly reduce the search space for the correct model order and parameter estimates. The algorithm initially overestimates the model order. After iterative minimization to obtain maximum likelihood (ML) parameter estimates for that order, a post-processing step eliminates the extraneous sinusoidal modes using CFU3 resolution limits and statistical detection tests. Because the algorithm searches on only a limited set of model orders and parameter regions, it is computationally tractable even for large data lengths and model orders. In this paper we analyze the performance of the algorithm and compare with other existing approaches.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126144811","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572422
I. Gorodnitsky, B. Rao
{"title":"Truncated Total Least Squares Regularization Underdetermined Problems","authors":"I. Gorodnitsky, B. Rao","doi":"10.1109/SSAP.1994.572422","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572422","url":null,"abstract":"","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130424170","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572464
S. Affes, S. Gazor, Y. Grenier
In this paper, we generalize a former work recently presented in the narrowband case of robust adaptive beamforming via target tracking to the wideband domain. The original algorithm is applied to each frequency component of the signal in an Analysis/Synthesis scheme. The source tracking and localization are simply performed in one frequency selected with the minimum location misadjustment. A more complex combination of location estimates can be computed in a specific set of frequencies, with a relatively better performance. Simulation results confirm in both cases the efficiency of the generalized algorithm regarding source localization and noise reduction.
{"title":"Wideband Robust Adaptive Beamforming Via Target Tracking","authors":"S. Affes, S. Gazor, Y. Grenier","doi":"10.1109/SSAP.1994.572464","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572464","url":null,"abstract":"In this paper, we generalize a former work recently presented in the narrowband case of robust adaptive beamforming via target tracking to the wideband domain. The original algorithm is applied to each frequency component of the signal in an Analysis/Synthesis scheme. The source tracking and localization are simply performed in one frequency selected with the minimum location misadjustment. A more complex combination of location estimates can be computed in a specific set of frequencies, with a relatively better performance. Simulation results confirm in both cases the efficiency of the generalized algorithm regarding source localization and noise reduction.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133537391","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572452
T. Bronez
Direction finding (DF) in the high-frequency (HF) band is challenging since the signal and noise environment can at best be modeled only nominally, yet the high resolution of model-based methods is typically needed. In our analytical and experimental investigation of HF/DF, we have developed a new bearing estimation method, MICL, that incorporates an identifiability constraint into the standard ML method. We have also developed a companion source enumeration method, EIL, based on estimated incremental likelihoods. We describe MICL/EIL and apply it to real HF field data, demonstrating its utility for significant, HF/DF improvements.
{"title":"MICL/EIL- An Effective Approach for Simultaneous Source Enumeration and ML Direction Finding","authors":"T. Bronez","doi":"10.1109/SSAP.1994.572452","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572452","url":null,"abstract":"Direction finding (DF) in the high-frequency (HF) band is challenging since the signal and noise environment can at best be modeled only nominally, yet the high resolution of model-based methods is typically needed. In our analytical and experimental investigation of HF/DF, we have developed a new bearing estimation method, MICL, that incorporates an identifiability constraint into the standard ML method. We have also developed a companion source enumeration method, EIL, based on estimated incremental likelihoods. We describe MICL/EIL and apply it to real HF field data, demonstrating its utility for significant, HF/DF improvements.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131164268","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572432
W. B. Bishop, P. Djurić, D. E. Johnston
Information provided by the accurate model selection of exponential time series is indispensable in many areas of science and engineering. This paper presents a method for the simultaneous detection and estimation of signals composed of sums of damped exponentials in additive noise. The method is entirely Bayesian in that the utility of a marginalized posterior probability density allows for the formulation of a maximum a posteriori (MAP) model selection criterion. Numerical integrations are accomplished through the application of a computationally efficient algorithm known as Adaptive Importance Sampling (AIS). This procedure, which requires no knowledge regarding the functional form of the integrands and enforces parameter constraints with relative ease, presents itself as a welcome alternative to constrained multidimensional optimization. Monte-Carlo simulations on two component synthesized data indicate a n e table improvement in selection performance of the MAP over both, the AIC and MDL.
{"title":"Bayesian Model Selection of Exponential Time Series Through Adaptive Importance Sampling","authors":"W. B. Bishop, P. Djurić, D. E. Johnston","doi":"10.1109/SSAP.1994.572432","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572432","url":null,"abstract":"Information provided by the accurate model selection of exponential time series is indispensable in many areas of science and engineering. This paper presents a method for the simultaneous detection and estimation of signals composed of sums of damped exponentials in additive noise. The method is entirely Bayesian in that the utility of a marginalized posterior probability density allows for the formulation of a maximum a posteriori (MAP) model selection criterion. Numerical integrations are accomplished through the application of a computationally efficient algorithm known as Adaptive Importance Sampling (AIS). This procedure, which requires no knowledge regarding the functional form of the integrands and enforces parameter constraints with relative ease, presents itself as a welcome alternative to constrained multidimensional optimization. Monte-Carlo simulations on two component synthesized data indicate a n e table improvement in selection performance of the MAP over both, the AIC and MDL.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130943839","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572511
D. Carhoun
Reduced-rank subspace projection methods are used indirectly in frequency and angle-of arrival estimation algorithms such as MUSIC and its relatives, but they are not commonly used directly in least-squares detection applications. We have been exploring their use for the processing of underwater acoustic receiver array data for detection and matched-field localization. We will describe several techniques that have been developed and applied to signals recorded from different types of arrays.
{"title":"Signal Subspace Projection Methods of Adaptive Sensor Array Processing","authors":"D. Carhoun","doi":"10.1109/SSAP.1994.572511","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572511","url":null,"abstract":"Reduced-rank subspace projection methods are used indirectly in frequency and angle-of arrival estimation algorithms such as MUSIC and its relatives, but they are not commonly used directly in least-squares detection applications. We have been exploring their use for the processing of underwater acoustic receiver array data for detection and matched-field localization. We will describe several techniques that have been developed and applied to signals recorded from different types of arrays.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123359130","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572436
I. Kirsteins
An analysis and interpretation of the reduced-rank generalized likelihood-ratio test (RR-GLRT) detector is presented in this paper. First, simple and accurate approximations to the RR-GLRT test statistics are derived. The approximations are verified by computer simulation and are shown to be accurate over a wide range of interference and signal levels These approximation are then used to show that the RR-GLRT is related to the UMP invariant test and to calculate the moments and deflection of the RR-GLRT statistic.
{"title":"Analysis and Interpretation of the Reduced-Rank Generalized Likelihood-Ratio Test","authors":"I. Kirsteins","doi":"10.1109/SSAP.1994.572436","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572436","url":null,"abstract":"An analysis and interpretation of the reduced-rank generalized likelihood-ratio test (RR-GLRT) detector is presented in this paper. First, simple and accurate approximations to the RR-GLRT test statistics are derived. The approximations are verified by computer simulation and are shown to be accurate over a wide range of interference and signal levels These approximation are then used to show that the RR-GLRT is related to the UMP invariant test and to calculate the moments and deflection of the RR-GLRT statistic.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114836786","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572420
J.M. Frances, B. Friedlander
We consider a class of nonstationary multi component signals, where each component has a random amplitude and a deterministic phase. The amplitude is a stationary Gaussian process plus a time varying mean. The phase and the amplitude mean are characterized by linear parametric models, while the covariance of the amplitude function is parameterized in some general manner. This model encompasses signals which are commonly used in communications, radar, sonar, and other engineering systems. We derive the Cramer Rao Bound for the estimates of the amplitude and phase parameters.
{"title":"Parametric Estimation Of Multi-component Signals With Random Amplitude And Deterministic Phase","authors":"J.M. Frances, B. Friedlander","doi":"10.1109/SSAP.1994.572420","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572420","url":null,"abstract":"We consider a class of nonstationary multi component signals, where each component has a random amplitude and a deterministic phase. The amplitude is a stationary Gaussian process plus a time varying mean. The phase and the amplitude mean are characterized by linear parametric models, while the covariance of the amplitude function is parameterized in some general manner. This model encompasses signals which are commonly used in communications, radar, sonar, and other engineering systems. We derive the Cramer Rao Bound for the estimates of the amplitude and phase parameters.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122752705","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 : 1994-06-26DOI: 10.1109/SSAP.1994.572471
R. DeLap, A. Hero
In this paper we develop an adaptive beamsummer for direction of arrival (DOA) estimation of slow fading Raleigh signals, using a new design approach, termed "Adaptive Detection/Estimation for specific Tasks" (ADEPT). For DOA estimation, the ADEPT method yields a weight adaptation criterion which is optimized for those weights that minimize the Cramer-Rao (CR) lower bound on achievable mean-square-error of any unbiased DOA estimator constructed on the beamsummer outputs. Simulation results are provided which show that the ADEPT DOA beamsummer yields DOA estimates whose meansquared error (MSE) approaches that of the more complex maximum likelihood implementation.
{"title":"Adaptive Beamforming for Slow Fading Rayleigh Signals","authors":"R. DeLap, A. Hero","doi":"10.1109/SSAP.1994.572471","DOIUrl":"https://doi.org/10.1109/SSAP.1994.572471","url":null,"abstract":"In this paper we develop an adaptive beamsummer for direction of arrival (DOA) estimation of slow fading Raleigh signals, using a new design approach, termed \"Adaptive Detection/Estimation for specific Tasks\" (ADEPT). For DOA estimation, the ADEPT method yields a weight adaptation criterion which is optimized for those weights that minimize the Cramer-Rao (CR) lower bound on achievable mean-square-error of any unbiased DOA estimator constructed on the beamsummer outputs. Simulation results are provided which show that the ADEPT DOA beamsummer yields DOA estimates whose meansquared error (MSE) approaches that of the more complex maximum likelihood implementation.","PeriodicalId":151571,"journal":{"name":"IEEE Seventh SP Workshop on Statistical Signal and Array Processing","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123064792","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：