Bispectral methods of signal processing. Applications in radar, telecommunications and digital image restoration (Q2877479)

The objective of this unique volume is to provide the theory and case studies of digital signal processing based on bispectrum, which is one of the most sensitive tools for non-Gaussianity estimation. The key difference of bispectrum from energy spectrum is phase information preservation which makes it a valuable technique in various fields including automatic target recognition, image processing, biomedical signals processing. There are four chapters in this volume.NEWLINENEWLINENEWLINEThe first two chapters provide the theory and basic properties of bispectrum and accuracy of its estimation. These chapters can be employed to design new adaptive nonlinear filters. The objective of chapter 3 is to demonstrate the efficiency of bispectrum-based signal processing techniques for restoration of images corrupted by jitter and additive noise. Finally, the authors include their recent results concerning automatic targets recognition and third-order tests statistics employment for wireless communication systems. It includes case studies for naval, ground surveillance, aerial target recognition and classification.