The optimality in Neyman-Pearson sense in the distributed CFAR detection with multisensor

Abstract

The optimality in Neyman-Pearson (NP) sense in distributed CFAR detection with multisensor is discussed. Most of the existing analysis of optimization of distributed CFAR detection in the NP sense is done under the limitation of binary local decision and no communication among local processors. We find that the real optimization in the NP sense can not be realized under this limitation. If local test statistics (LTS) are used and fused, the real optimal NP test could be implemented by likelihood ratio test (LRT).