Joint Sensor Failure Detection and Corrupted Covariance Matrix Recovery in Bistatic MIMO Radar With Impaired Arrays

Abstract

A bistatic multiple-input multiple-output (MIMO) radar, which is usually constituted by a large number of antenna sensors, may encounter one or more failures in the transmit or receive array after long periods of continuous operation. In this paper, we present a novel technique that integrates the detection of faulty sensors and the recovery of the corrupted covariance matrix into a systematic process for a bistatic MIMO radar with impaired arrays. A simple scheme based on the cross-correlation method, which exploits different behaviors between the faulty and normal sensors, is developed for the detection of the faulty transmit and receive sensors without requiring external probes. To mitigate the negative effects brought by sensor failures, the matrix completion method is fully integrated with the structured processing of the four-fold Hankel matrix to more reliably recover the successive corrupted data in both the covariance matrix rows and columns that result from the failure of the sensors. The proposed technique can alleviate or even avoid the angle estimation performance loss of the array covariance matrix-based algorithms in the face of a wide range of faulty sensor numbers. Extensive numerical simulations verify the effectiveness of the proposed technique.