An Estimator for Multi-Sensor Data Fusion

In this paper, we examine binary hypothesis testing and parameter estimation problem in a sensor network. We address the problem of detection and also the estimation of the underlying parameter at the fusion center by optimally combining the test statistics sent by different sensors. We make no assumptions on the noise statistics at the sensor nodes except that their first and second order statistics are known. We show that the proposed method is optimal as an estimator as well as a detector.