Optimal distributed decision fusion in the sense of the Neyman-Pearson test

Abstract

The theoretical analysis of the relationship between the fusion rule and local decision rules in the distributed detection fusion (DDF) system with multiple sensors is proposed in the sense of the Neyman-Pearson (N-P) test. The determination of the decision rules of two suboptimal systems and a globally optimal system are derived completely. Accordingly, the detection performance and receiver operating characteristics (ROCs) of three systems above are computed numerically for the problem of detecting a known signal embedded in Rayleigh noise. The results obtained show the globally optimal system performs better than the others.