Study And Implementation Of Passive Invasion Detection System Based On WSN

There is a theoretical and practical significance for passive invasion detection technology using in fire rescue, energy conservation and military protection. On the one hand, the object doesn’t need take any wireless transceivers, on the other hand, it solves the problem that the detection technologies used in video and image are restricted to the light and other environment. But traditional invasion detection method based on the signal variation is easily influenced by the complex environment and the detection error increases dramatically with the change of the environment. So it is a big challenge to achieve both higher detection rate and lower error detection rate.This paper proposes two independent invasion detection algorithms:passive invasion detection algorithm based on differential signals, it utilizes the similarity of characteristic matrix which is composed of mean and variance of signal between the subsequent two moments to indicate the change of signal and achieves invasion detection robustly and successfully. The method solves the problem of the increasing detection error caused by the time-variation of referential signal and offsets the noise caused by the variation of test environment. Improved VIBE invasion detection algorithm, firstly, it utilizes the real value of signal to make template which exactly reflects the distribution of signal. Secondly, it proposes the invasion decision criteria of weighted processing to single link’s detection result to improve the whole detection performance of the detection algorithm. Lastly, it takes the updating principle of first-in first-out to accurately update the signal template in real time.This paper realizes a detection system with the two proposed algorithms, the system is composed of wireless scan nodes, wireless control node and PC, the communication protocol among the nodes is improved token ring protocol, In consideration of the network robustness, we set a timer to prevent the network paralysis in case of single node breakdown. This paper also proposes a method of vertical scanning the channels to find the minimum signal variance channel as the work channel of wireless sensor nodes, it reduces the same frequency interference to a large extent.Finally, this paper tests the realized invasion detection system in three scenes with different complex level, the results show that the two invasion detection algorithms put forward in this paper have a higher detection rate and lower error detection rate than traditional invasion detection algorithm, and they are less likely to be affected by the number of wireless scan nodes. The detection algorithms proposed in this paper can still get optimistic detection performance with less number of wireless scan nodes.