A low-complexity cooperative localization algorithm based on variational message passing in wireless networks

We consider a wireless network where target nodes aim to locate themselves. A cooperative localization algorithm can provide high accuracy by exchanging location information between nodes. In this paper, a novel distributed cooperative localization algorithm based on factor graph and variational message passing (VMP) is proposed. Compared with other message passing method, VMP algorithm features lower communication overhead. An approximation of likelihood function is made thus all messages on factor graph can be obtained in closed Gaussian forms and therefore reduce the computational complexity and communication cost. Simulation results show that the proposed method performs very close to SPAWN with much lower complexity.