RSS Based Compressed Sensing Localization Algorithm For Malicious Node Attack

Among the traditional Wireless Sensor Networks(WSN)positioning algorithms,the received signal strength(RSS)based method is simple and has received significant attention from both academia and industry over the last few years.WSN positioning system with compressed sensing(CS)theory can reduce the number of RSS measurements and has better positioning performance.However,RSS measurements are vulnerable to physical attacks for practical application.Therefore in this thesis,we study the secure positioning algorithm for malicious node attacks in CS-based WSN positioning system.The main works include:(1)The previous works for CS-based localization algorithms from domestic and abroad are introduced.And some related knowledge about CS theory are described in detail.Then,a CS-based positioning model are formulated.Its security are also proposed for malicious attack description.At last,the linear attack model is set up to simulate the real RSS attacks.All of these have provided essential theoretical support for secure positioning mechanism design.(2)A least median based secure positioning approach is proposed.It uses the least median of residual error to detect the attacked nodes and has effectively improved the positioning performance.For another,by the theoretical analysis and experimental simulation,it solves the problem of the parameter design of the algorithm which can ensure the best anti-attack performance.(3)A secure mechanism based on residual analysis was proposed.Firstly,the CS reconstruction algorithm is used to estimate the target position,and the residual vector corresponding to the target position is calculated.Then,using the residual error criterion,we can find all the suspected attacked nodes.According to the maximum residual criterion,the attacked measurement is determined.When the sum of the residuals tends to be stable,it shows that all the attacked nodes have been removed.At last,the final positions are estimated by using the attacked free measurements.Simulation results show that the proposed algorithm can perform well under malicious node attacked environements.