Self-localization System of Wireless Sensor Based on Orthogonal Basis Neural Network Algorithm
Xinjuan Sun, Zhiyu Gao and Chengcai Zhang
In order to improve the locating accuracy of wireless sensor self-localization system, it is necessary to study the self-localization method of wireless sensor. When using the current method to locate nodes in wireless sensor networks, there are many problems, such as low signal-to-noise ratio, low location efficiency, low accuracy of node identification and low accuracy of location results. Based on orthogonal basis neural network algorithm, a self-localization method for wireless sensor networks is proposed. According to compressed sensing theory, the noise signal in wireless sensor networks is removed. The initial position information of nodes in wireless sensor networks is obtained by RSSI, and the initial position information is input into orthogonal basis neural networks as input value. In the network, by adjusting the weights to correct the error function, the localization results of the wireless sensor network nodes are output to complete the localization of the wireless sensor itself. Experimental results show that the proposed method has high signal-to-noise ratio, high positioning efficiency, high accuracy of node recognition and high positioning accuracy.
Keywords: orthogonal basis neural network; wireless sensor network; location system; node identification