Student Paper 2.26
A method using influence function for evaluating robustness of CFAR detectors

Huadong Meng - Dept. of Electronic Engr., Tsinghua Univ., P.R.China , Xiqin Wang - Dept. of Electronic Engr., Tsinghua Univ., P.R.China , Hao Zhang - Dept. of Electronic Engr., Tsinghua Univ., P.R.China , Yingning Peng - Dept. of Electronic Engr., Tsinghua Univ., P.R.China

Tue, 27 April 2004, 3:20 PM - 4:10 PM

Abstract

A method for evaluating the robustness of constant false alarm rate (CFAR) detectors is presented in this paper, which is based on the powerful methodology of influence function (IF) developed in the literatures on robust statistic. The robustness of different kinds of CFAR detectors can be evaluated and compared by calculating the first derivative of the false alarm probability (FAP) and detection probability (DP) at an underlying distribution, which are named IF-FAP and IF-DP. The comparison of those two IFs among some kinds of CFAR detectors are illustrated. Then it is concluded that the robustness of detector can be asymptotically represented by the IF of clutter power estimator. Finally, according to a robust measure drawn from the IF of clutter power estimator, the "most robust" detectors in three ordered-statistic-based groups are presented.

Bios

Huadong Meng received the B.E and M.E from Tsinghua University, P.R.China, in 1999 and 2001. He is currently the Ph.D. Candidate of the High-Speed Real-Time Signal Processing Division, Department of Electronic Engineering, Tsinghua University, doing research in Statistical Singal Processing and Radar. He is a student member of the IEEE.

Prof. Xiqin Wang - Dept. of Electronic Engr., Tsinghua Univ., P.R.China

Xiqin Wang received his B.E. and Ph.D. degrees in electronic engineering from Tsinghua University, Beijing, China, in 1991 and 1996, respectively. From 1991 to 2000, he worked on radar and image signal processing and systems in the Department of Electronic Engineering in Tsinghua University. From 2000 to 2003, as a visiting scholar and assistant research engineer he worked on algorithms and implementations of road traffic collision warning systems in PATH/ITS, School of Engineering, UC Berkeley. He is currently a professor and the vice chair of the Department of Electronic Engineering in Tsinghua University. His recent research interests include intelligent transportation systems, information systems and engineering.

Prof. Yingning Peng - Dept. of Electronic Engr., Tsinghua Univ., P.R.China

Yingning Peng received the B.S. and M.S. degrees in electical engineering from Tsinghua University, Beijing, China, in 1962 and 1965, respectively. Since 1965, he has worked at the Department of Electronic Engineering, Tsinghua University. He is currently a professor and the director of High-Speed Signal Processing and Network Research Institute. He has worked with high-speed and real-time signal processing for many years. His recent research interests include spectral estimation, adaptive filter, adpative array signal processing, and parallel signal processing. He is a senior member of IEEE.