Fundamentals of Statistical Signal Processing: Estimation and Detection Theory: Estimation and Detection Theory

Author: Steven M. Kay
Publisher:
Publish Date: 2003-08-01
Features: This book is a classic and authoritative work on statistical signal processing. It is divided into two volumes, each explaining the fundamental theories of estimation and detection in statistical signal processing. This book can serve as a graduate textbook or teaching reference for statistical signal processing courses in electronic and information engineering majors, and can also be used as a reference for teaching, research, and engineering personnel engaged in signal processing. This book is a classic and authoritative work on statistical signal processing. It is divided into two volumes, each explaining the fundamental theories of estimation and detection in statistical signal processing. Volume 1 provides a detailed introduction to classical estimation theory and Bayesian estimation, summarizes various estimation methods, considers Wiener filtering and Kalman filtering, and introduces estimation methods for complex data and parameters. This volume presents numerous application examples, including high-resolution spectral analysis, system identification, adaptive noise cancellation, tracking, and localization; it also includes a large number of exercises to deepen readers' understanding of the basic concepts. Volume 2 comprehensively introduces the optimal detection algorithms implemented on computers and focuses on real-world signal processing applications, including modern speech and communication technologies as well as traditional sonar/radar systems. This volume begins with the fundamental theories of detection, reviews Gaussian, chi-squared (χ2), F, Rayleigh, and Rice probability density functions; explains quadratic forms of Gaussian random variables, as well as asymptotic Gaussian probability density and Monte Carlo performance evaluation; introduces the theoretical foundation of detection based on simple hypothesis testing; Neyman-Pearson theorem, handling of irrelevant data, Bayesian risk, multiple hypothesis testing, as well as detection of deterministic and random signals; and finally, provides a detailed analysis of composite hypothesis testing suitable for unknown signal and unknown noise parameters. This book can serve as a graduate textbook or teaching reference for statistical signal processing courses in electronic and information engineering majors, and can also be used as a reference for teaching, research, and engineering personnel engaged in signal processing.