Simple suboptimal design of Bayesian distributed detection systems

The authors analyze and compare two multi-sensor multi-observation detection schemes, and discuss their hardware complexity. The schemes are a Bayesian optimal parallel-sensor centralized architecture and a suboptimal binary distributed-detection system. Both systems have the same performance, as measured in terms of a Bayesian risk. The authors study two specific cases: 1) discrimination between two Gaussian populations which differ in their means; and 2) discrimination between two Poisson populations which differ in their parameters. The authors demonstrate the tradeoff between performance and hardware complexity, and calculate the cost in terms of hardware units of the design simplicity which characterizes the suboptimal system. It is shown that in the Gaussian case, a high signal-to-noise ratio decentralized system with 2N sensor/detectors performs at least as well as the centralized system with N sensors and a single detector.