BC_EM: A Link Loss Inference Algorithm for Wireless Sensor Network

Compared to wired networks, sensor network pose limited resources challenges for monitoring functions. Therefore, we propose to use only end-to-end application data to infer loss performance of internal network links based on the data aggregation process. We introduce inference techniques based on BC (Bound and Collapse) and EM (Expectation Maximize) principles, which often handle well to infer link loss rate form incomplete data set. Through the simulation, we can safely reach the conclusion that the internal link loss rate can be inferred more accurately than before, and the simulation also shows that the proposed algorithm scales well according to the sensor network size.