Radar Detection (Radar, Sonar and Navigation)
This book presents a comprehensive tutorial exposition of radar detection using the methods and techniques of mathematical statistics. The material presented is as current and useful to today's engineers as when the book was first published by Prentice-Hall in 1968 and then republished by Artech House in 1980. The book is divided into six parts. Part I is introductory and describes the nature of the radar detection problem. Part II reviews the mathematical tools necessary for a study of detection theory. Part III contains tutorial expositions in a radar context of the classical signal-to-noise and a posteriori theories, both of which have played important roles in the evolution of modern radar. The unifying theme of the book is provided by statistical decision theory, introduced in the last chapter of Part III, which provides the framework for the chapters that follow. The first three chapters of Part IV contain a unified tutorial exposition of single and multiple hit detection theory. The last two chapters are respectively devoted to the use of the radar equation and a discussion of cumulative detection probability. The latter includes a procedure for minimizing the power-aperture product of a search radar. The performance of near-optimum multiple hit detection strategies are considered in Part V. These include binary and pulse train detection strategies. The first chapter in Part VI applies sequential detection theory to the radar detection problem. It includes the Marcus and Swerling test strategy and a two-step approximation to sequential detection. The second chapter contains the development of Bayes decision rules and Bayes receivers for optimizing the detection of multiple targets with unknown parameters, such as range, velocity, angle, etc.
Why Read This Book
You should read this book if you want a rigorous, tutorial-style grounding in radar detection framed entirely in statistical decision theory; you will learn how classical and modern detection rules (Neyman–Pearson, Bayesian, GLRT, matched filtering) are derived and applied to real radar problems. The book's clear exposition of probability tools and performance measures makes it immediately useful for engineers who must analyze detector performance, trade off false alarms vs. detection, or implement principled radar decision algorithms.
Who Will Benefit
Practicing radar/communications engineers, graduate students, and researchers with some signal-processing background who need to design or analyze statistical detectors for radar, sonar, or communications systems.
Level: Advanced — Prerequisites: Undergraduate calculus and linear algebra, probability and random processes (basic stochastic processes and Gaussian models), and familiarity with signals and linear systems; familiarity with MATLAB or similar for numerical experiments is helpful.
Key Takeaways
- Apply Neyman–Pearson and Bayesian decision rules to formulate optimal radar detection tests under simple and composite hypotheses
- Derive and implement matched filters, likelihood-ratio tests, generalized likelihood-ratio tests (GLRT), and related detectors for common radar signal models
- Analyze detector performance quantitatively using SNR measures, ROC curves, probability of detection and false alarm, and asymptotic approximations
- Design and understand constant false alarm rate (CFAR) schemes and strategies for fluctuating targets and nonstationary backgrounds
- Model and handle stochastic signal and noise processes in detection problems, including Gaussian and certain non-Gaussian scenarios
Topics Covered
- Part I: Introduction — The radar detection problem and practical motivations
- Part II: Mathematical Preliminaries — Probability, random variables, pdfs, expectations, characteristic functions
- Part II (cont.): Random processes and second-order statistics relevant to radar
- Part III: Classical Detection Theory — SNR-based approaches and matched filter derivations
- Part III (cont.): Neyman–Pearson lemma and likelihood-ratio tests for simple hypotheses
- Part IV: A Posteriori and Bayesian Detection — Prior information, decision costs, and Bayes rules
- Part V: Composite Hypotheses and Practical Detectors — GLRT, locally most powerful tests, and parameter estimation
- Part V (cont.): CFAR methods and treatments of fluctuating targets and inhomogeneous backgrounds
- Part VI: Performance Analysis — ROC curves, error probabilities, asymptotic results, and numerical examples
- Supplementary Topics: Sequential detection, multiple hypothesis testing, and worked radar examples
- Appendices: Mathematical derivations, tables, and reference material
Languages, Platforms & Tools
How It Compares
Covers similar ground to S. M. Kay's "Fundamentals of Statistical Signal Processing, Volume 2: Detection Theory" but is more focused on radar-specific applications and historical perspectives; Van Trees' "Detection, Estimation, and Modulation Theory" is broader and more exhaustive, while DiFranco provides a compact, tutorial presentation tailored to radar engineers.
