Multi-relay Selection Schemes For Physical Layer Security Cooperative Networks

Cooperative communication can improve the physical layer security performance of the wireless networks and further exploit the spatial diversity.With the increase of the number of relay nodes,information transmission faces the problem of large energy loss.In order to enhance the system security,the relay selection problem has received extensive attention.It is necessary to design the suitable relay selection algorithms for different network models.This paper studies the problem of multi-relay selection in compressed sensing networks using the amplify-and-forward mechanism.Firstly,to find the local optimal relay nodes,a distributed multi-relay selection algorithm based on coalition formation game(CFG)is proposed to improve the secrecy capacity of distributed network with low complexity.The impact of the parameters,such as source power and relay power,on the network performance is studied,and the validity and adaptability of the algorithm are verified.Futhermore,compared with the traditional Pareto order,the security performance of proposed CFG algorithm is better.However,to ensure the accuracy of the result,the accurate channel state information is required.To further improve the security performance,it is necessary to design a global optimal relay selection method.Thus,a self-organizing algorithm(L-MRS)based on stochastic learning is proposed.According to the newly defined probability updating rules,L-MRS algorithm finds the optimal relays in a self-learning and self-optimization way.The destination node chooses the best state and the relay nodes decide whether to join the cooperation according to the environment feedback.Simulation results show that L-MRS algorithm can select the optimal relay combination correctly,and the security performance is better than the coalition formation game algorithm.