Joint Detection and Localization in Sensor Networks Based on Local Decisions

A generalized likelihood ratio test (GLRT) based decision fusion method that uses quantized data from local sensors is proposed to jointly detect and localize a target in a wireless sensor field. The signal intensity is assumed to be inversely proportional to a power of the distance from the target. The GLRT, its corresponding maximum likelihood (ML) estimator, and the Cramer-Rao lower bound (CRLB) are derived. Simulation results show that this fusion rule has a significantly improved detection performance, compared with the counting rule (for hard local decisions) or the intuitive fusion rules based on the average of sensor data (for soft local decisions).