A Data-centric Distributed Detection Scheme in Wireless Sensor Networks

Detection of certain events or targets constitutes an important application of wireless sensor networks (WSN) in which spatially distributed nodes communicate to a decision center. Compared with such decision-centric communication strategies extensively proposed in literature, however, data-centric distributed detection scheme is rather a new design methodology. In this paper, we simulate and determine an efficient way to detect distributed information among sensor nodes. A novel combination of realistic communication channel approximations, Dijkstra´s shortest path algorithm, and the optimal ML receiver is introduced to do it efficiently. It is expected and verified through simulation that as the number of nodes in a given area increases, the estimation of the signal at the center node gets better and produces less error