Information theory helps to clarify and suggest new approaches to the design and analysis of statistical signal processing algorithms. This book explores the link between information theory and the design of signal processing algorithms used in radar, sonar, imaging, and pattern recognition, and others. The intent of this book to expand its applicability to new problems of interest. The principal information-theoretic measures currently in use: Kullback-Leibler divergence, mutual information, and Fisher information are studied with numerous examples given to illustrate the concepts. Special features are: comprehensive discussion of exponential probability density functions, information geometry insights, robust detection and spectral estimation, the p* formula and saddlepoint approximations, feature/subset/order selection for statistical modeling, and extended scoring method for maximum likelihood estimation.