The Detection Arithmetic of Soft Decision Based on Non-ideal Channel in Distributed Detection

Abstract

In this paper we expand the arithmetic of the distributed detection in two aspects. In order to improve the detection performance and use communicated resource rationally, the local decision is extended from hard decision to soft decision. According to N non-ideal independent channels, the local decision and reliability information are transmitted to the fusion center. The non-ideal channel is extended from binary symmetry channel to memorial channel. Then we derive the optimal decision form of the local detectors and fusion center by NP(Neyman pearson) rule. At last stimulation of two detectors shows that using soft decision improves the detection performance and worsen channel depress the detection performance surely.