Research on Fault-Tolerant Target Detection algorithm in wireless Sensor Networks

We have designed an efficient method for fault-tolerant target detection in wireless sensor networks. With which the whole network consists of two main kinds of wireless sensors-primary sensors and base ones. Small primary sensors are usually cheap and there is a large number of this kind of sensors distributed in the area. The base sensors have a strong computational capability but may be more expensive and bigger. They can be divided into two kinds of appearances-sleeping and active. The active ones cover the detection area completely with little redundancy. The signal strength a primary sensor detects depends on other primary sensors in their fixed radio range. Primary sensors locally communicate and compute with others at the first step and filter out junior sensors based on signal strength. By use the similar algorithm, they can also filter out senior sensors at the second step which communicate with the base sensors directly. The base sensor computes the final result by data aggregation. We reduce the importance of each single sensor node in order to lower its destructiveness once it is failed. In other words, whether a real target exists or not must be jointly determined by neighboring sensors at the same time. We consider delegating one senior sensor to communicate with the base node for each target and compute the position of the target locally in order to decrease the communication overhead.