A competent fault tolerant system using simulated annealing approach for target tracking in wireless sensor networks

The wireless sensor networks are usually energy constrained and to ensure this network is highly consistent, the reliability factor should be given top priority. Because of hardware crash, software bugs, environmental risks etc., there would be errors and faults occur and hence the reliability factor is severely affected. Most of the faults at sensor nodes are either permanent in nature due to energy depletion or transient fault due to momentary disruption like high heat etc. Thus the sensor network should be fault tolerant and hence, a novel approach towards clustering with optimal coverage of tracking multiple objects in WSNs is being experimented. The Proposed method engaged improved K-means clustering algorithm with simulated annealing technique which employs to find optimal use of sensor nodes and recover from fault as quickly as possible so that target detection won't be disrupted. This technique is simulated in Matlab software to analyze consumption of energy and network lifetime by using proposed algorithm. Accordingly this has been checked and demonstrated that this model shows major improvement when comparing the Traditional clustering in rate of energy consumption and network lifetime.