Abstract:

In order to reconcile the competition conflict in the task allocation of sensor nodes during the multi-target tracking in wireless sensor networks（WSN）,a novel task allocation method is proposed based on the fuzzy clustering and the multiple-elastic module（MEM） self-organizing neural network.In this method,first,the target number is estimated via fuzzy clustering.Next,an integrated performance index is presented to describe the tracking precision and the energy consumption in the multi-target tracking.Then,a self-organizing neural network with non-fully-connected elastic neuron-ring structure is constructed.Moreover,the elastic submodule of neurons are initia-lized by means of the nearest neighbor method,whose receptive field is finally adjusted dynamically according to the WTA（Winner Take All） principle so as to quickly lock the optimal node alliance and ensure an accurate multi-target tracking.Experimental results show that the proposed method effectively reconciles the competition conflict in the task allocation of sensor nodes during the multi-target tracking,avoids further increase in energy consumption and prevents any possible real-time performance degradation due to the competition conflict;and that,in the conditions of random uniform node topology and linear target movement,the proposed method is superior to both the nearest neighbor method and the conventional MEM neural network method,respectively with an energy consumption decrease by 48.2%~55.9% and 37.4%~42.5%,and with a calculation speed increment of 19.0%~27.4%.

Key words: wireless sensor networks, multi-target tracking, task allocation, multiple-elastic module self-organizing map neural network, fuzzy clustering