A robust approach to decentralized CFAR detection in spiky clutter

In this paper we consider the problem of decentralized constant-false-alarm-rate (CFAR) detection in a distributed multiradar system, by means of an approach based on the concept of robustness. State-of-the-art biparametric CFAR algorithms are computationally demanding and lossy, so the question that arises is: Are they really necessary? We show that the answer is no, or rather not always. A new simple monoparametric CFAR detector with postdetection integration is proposed, that, in spite of the single parameter estimation, allows to obtain a false alarm probability at the fusion center robust against changes of the degree of clutter spikiness. The robustness is obtained by a joint action: multisensorial integration, local temporal integration, in conjunction with a specific monoparametric estimator. After an analysis of the robust algorithm, we determine the detection performance of the multisensor network and present a comparison with a decentralized system employing biparametric algorithms.