Communication-efficient decentralized hypothesis testing for sensor networks with minimum fusion center feedback

A new decentralized hypothesis detection framework is developed, where local sensors are memoryless, receive independent observations, and minimum feedback from the fusion center. In addition to the standard criterion of minimizing detection delay under error probability constraints, an additional constraint on the number of communications between local sensors and the fusion center is introduced here. This communication metric is able to reflect both the cost of establishing communication links as well as overall energy consumption over time. The proposed detection scheme minimizes detection delay with constraints on both error probabilities and on the number of communications. Unfortunately, optimality requires derivation of time-delay distributions of reports to the fusion center and solution of nonlinear equations. Under certain approximations, a Poisson arrival model is shown to yield a tractable solution. The efficiency of the proposed algorithm is quantified analytically in comparison to centralized detection and is shown to be in agreement with simulation results.