Uniformly most powerful distributed detection and its application in cooperative spectrum sensing

In this paper, a special class of distributed composite binary hypothesis testing problem with monotonic likelihood ratio is investigated. The sensor observations are assumed to be conditionally independent given a fixed but unknown parameter θ where θ ∈ Θ₁ under the H₁ hypothesis and θ = θ₀ under the H₀ hypothesis. The optimal form of sensor decision rule is established under both the Neyman-Pearson and Bayesian criteria. As an illustrative example, the design of an optimal cognitive radio rule for cooperative spectrum sensing is established.