To conquer shortcoming of equivalent between accuracy and energy consumption

a dynamically clustering algorithm

called nearest neighborhood collaboration

is proposed for collaborative target tracking in wireless sensor networks.The algorithm forms target tracking cluster to adapt the real time change of moving target for optimal tracking precision.When target moves into the monitoring area

multi-sensor nodes that can sense the target work collaboratively as a tracking cluster for initial position of target based on least square.Kalman prediction mechanism is introduced for calculation of moving target location.Cluster head is selected based on collaboration nearest neighboring criterion.Cluster head chooses the tasking nodes to observe target location in its neighborhood

and fulfills state estimation with central Kalman filter.Simulation results show that

compared with existing algorithm

the proposed algorithm not only achieve superior tracking accuracy but also energy saving due to single hop communication distance.