Distributed bayesian target identification algorithm

An eract distributed Bayesian algorithm for multisensor iarget identijcation is presented in this paper. Using this new developed algorithm. the local posterior probabiliiy is jrsi calculated in each sensor, and then seni to fusion center, where all received data arefised with the prior information, to arrive at the global posterior probabiliiy. While using former hierarchical Bayesian algorithm, the local likelihood is first calculated in each sensor node, and then sent to fusion center, where all received daia are fused wiih ihe prior injonnation, io arrive at the global posterior prohabiliy. Compare io former algorithm, this new algorithm needs fewer computation in fusion center and fewer communication between sensor node andfusion center, while needs a little more computation in the sensor node. In situations while the fusion cenier si$fers from computaiion oppression, this new developed disiribuied Bayesian iarget identification algorithm provides a good solution.