Adaptive Detection Strategies With Electronic Protection Capabilities Against Chaff Echoes

In this article, we focus on an important problem for many radar systems, namely, the adaptive detection in the presence of chaff. This kind of countermeasure is aimed at distracting radar attention from the sought target also filling a portion of the spatial region of interest with a cloud of dipoles that are capable of hiding target echoes. In this context, we develop a detailed discrete-time signal model of the target and chaff components at the output of the matched filter for an array of sensors and pursue a model-based approach unlike most of the existing solutions that are data driven or based upon compressed data. As a matter of fact, we use this discrete-time signal model to formally state the detection problem at hand as a multiple hypothesis test with one null hypothesis and several alternative hypotheses. These hypotheses represent different situations of practical value. The architecture design is grounded on the hierarchical Bayesian framework and sparse estimation combined with the so-called penalized log-likelihood ratio test. The proposed detectors are assessed in terms of detection/classification capabilities, and the analysis shows that these solutions are a viable means to counteract the chaff attack at least for the considered scenarios.