In wireless sensor networks（WSNs）

a faulty sensor nodes may produce and transmit incorrect data.This would consume the limited energy and bandwidth of WSNs.Furthermore

the base station may make inappropriate decision when it receives the incorrect data.To solve these problems

a distributed fault detection algorithm for WSNs is presented.Sensor nodes are identified based on comparisons of data from neighboring nodes and the test state is transmitted to other nodes.Time redundancy is used to reduce faulty alarm rate for transient faults of nodes in WSNs.The algorithm has a good performance for fault nodes detection.Experimental results verify feasibility and effectiveness of the algorithm.