No abstract

No abstract

No abstract

No abstract

The problem of robust beamformer design in the presence of moving sources is considered. A new technique based on a generalization of the constrained minimum variance beamformer is proposed. The method explicitly takes into account c hanges in the scenario due to the movement of the desired and interfering sources, requiring only estimation of the desired DOA. Computer simulations show that the resulting performance constitutes a compromise between interference and noise rejection, computational complexity, and sensitivity to source movement.

Similarly to other Technical Committees, SSAP ran workshops, recommended paper awards, and reviewed papers for ICASSP. T o facilitate the paper review process and provide focus for award nominations the scope of SSAP was divided into several subareas, called SP EDICS" categories. These categories were: spectral analysis; higher order statistical analysis; cyclostationary signal analysis; statistical multi-channel ltering; statistical modeling; parameter estimation; detection; performance analysis; system identi cation; computational algorithms; and applications. These categories are covered in this article and continue to berepresented in the aggregated EDICS of the SPTM, SPCOM and SAM Technical Committees. As the reader will see from this article, SSAP impacts a very wide range of applications. Among the applications mentioned in the sequel are: radar signal processing; sonar signal processing; geophysics and climate; radar and optical remote sensing; electrocardiography ECG; electroencepholography EEG; magnetoencepholography MEG; nuclear magnetic resonance NMR imaging; radio-isotope imaging PET and SPECT; chemical sensing of the environment; physical oceanography; fractal internet tra c modeling; astronomy; biology; econometrics; speech; and music analysis or synthesis. Over the past several years the application of signal processing to communications has become a prevelant theme in SSAP. The prexistence of many relevant core SSAP areas made communications a v ery ripe applications area. In particular, research in cyclostationarity, higher order statistics, and system identi cation was a springboard to the development o f n o vel methods for channel equalization in digital communications. Likewise work in detection and estimation led naturally to iterative multiuser detection, source separation, and high performance modulation classi cation algorithms. As another example, deployment of phased antenna arrays and the associated signal processing has spearheaded much recent activity in spatial diversity reception for wireless communications. The sections by Giannakis and Tong highlight some of these communications applications of SSAP. Our article begins with a group of two sections on recent developments in detection estimation algorithms written by Alfred Hero and Petar Djuric, respectively. The section by Hero focuses on two areas of signi cant activity: constant false alarm rate CFAR detection and iterative maximum likelihood parameter estimation using the expectation-maximization EM algorithm. The section by Djuric describes the emerging area of Bayesian signal processing including estimation, detection, tracking and monte carlo markov chain MCMC sampling, a technique which was largely impractical before the current generation of high speed computers. The article continues with a section on time delay estimation written by Hagit Messer and Jason Goldberg and a section on multi-window spectral estimation by D a vid Thomson. From a historical

No abstract

No abstract