Performance evaluation of cooperative mobile communication security using reinforcement learning.

In wireless networks, there are many channel impairments including shadowing, path loss, and fading. Due to the shadowing, path loss, and fading, the direct transmission between the sender and the receiver may not attain acceptable signal quality. In that case, the cooperative communication has shown attractive solutions because it can use an alternate route to the destination. Cooperative communication increases the stability of the channel by cooperative spatial diversity. Quite recently, the advancement in the cellular network has resulted in new security requirements and challenges. Cooperative communications can increase spectral efficiency, power efficiency, and reliability; however, it is often vulnerable to eavesdroppers. Hence, inspired by the need for safe wireless mobile networks, the objective of the research is to increase the secrecy capacity of the wireless networks when there are moving cooperative communication devices. The technique proposed to enhance the wireless network security is organized by reinforcement learning which learns the transmit parameters according to the interaction of the transmitter, receiver, relay-node, and the eavesdropper devices. The simulation results have shown that the proposed technique has enhanced the secrecy level of the legitimate receiver.