Estimating Test Statistic Distributions for Multiple Hypothesis Testing in Sensor Networks

Abstract

We recently proposed a novel approach to perform spatial inference using large-scale sensor networks and multiple hypothesis testing [1]. It identifies the regions in which a spatial phenomenon of interest exhibits different behavior from its nominal statistical model. To reduce the intra-sensor-network communication overhead, the raw data is pre-processed at the sensors locally and a summary statistic is send to the cloud or fusion center where the actual spatial inference using multiple hypothesis testing and false discovery control takes place. Local false discovery rates (lfdrs) are estimated to express local believes in the state of the spatial signal. In this work, we extend our approach by proposing two novel lfdr estimators stemming from the Expectation-Maximization method. The estimation bias is considered to explain the differences in performance among the compared lfdr estimators.