Distributed Change Detection Based on a Consensus Algorithm

摘要: In this paper a novel distributed recursive algorithm is proposed for real time change detection using sensor networks. The based on combination of geometric moving average control charts generating local statistics and global consensus strategy; it does not require any fusion center, so that the final decision made by testing state node in network with respect to given common threshold. mean-square error centralized solution defined weighted sum analyzed case constant asymmetric matrices varying forgetting factors underlying recursions, assuming spatially temporally correlated data. These results are consistently extended random matrices, encompassing gossip schemes, lossy networks intermittent measurements, proving can be an efficient tool practice. simulation illustrate main characteristics algorithm, including matrix design, mean square as function factor, obtained quality expressed deflection estimation instant parameter change.