Distributed detection with memory

A binary distributed detection system comprises a bank of local decision makers (LDMs) and a central information processor (the data fusion center, DFC). All LDMs survey a common volume for a binary {H₀, H₁} phenomenon. Each LDM forms a binary decision: it either accepts H₁ ('target-present') or H₀ ('target-absent'). The LDM is fully characterized by its performance probabilities (probability of false alarm and probability of detection). The decisions are transmitted to the DFC through noiseless communication channels. The DFC then optimally combines the local decisions to obtain a global decision ('target-present' or 'target-absent') which minimizes a Bayesian objective function. The main difference between the present study and previous ones is that, along with the local decisions, the DFC in our architecture remembers and uses its most recent decision in synthesizing each new decision. We show that this feature endows our architecture with a detection performance that is generally much better than that of a memoryless DFC system. Moreover, when operating in a stationary environment, our architecture converges to a steady-state decision in finite time with probability one, and its detection performance during convergence and in steady state is strictly determined.