A Compressed Sensing Method for Wireless Sensor Networks with Evolution Model Based on KH-SVM

Wireless sensor networks are able to provide crucial and real time information in many scenarios of crisis response and management. Sensor node localization is one of research hotspots in the applications of wireless sensor networks (WSNs) field. A localization algorithm based on improved Support Vector Machine (SVM) for WSNs is proposed in this paper. SVM classification accuracy is the key to the localization accuracy. The selection of parameters is the important factor that influences the performance of SVM. Therefore, this paper proposes a parameter optimization algorithm based on Krill-herd algorithm (KH-SVM). The experimental results show that KH-SVM algorithm has better searching optimization ability compared with other optimization algorithms. In order to improve the fault tolerance against both random attacks and deliberate attacks for wireless sensor networks, this paper proposes a evolution model.We present an efficient seismic data sensing scheme in wireless sensor networks based on the promising compressed sensing technology to mitigate wireless communication load, data processing and caching complexity on nodes.