Optimal power allocation and secrecy capacity of the full-duplex amplify-and-forward wire-tap relay channel under residual self-interference.

摘要

Due to the broadcast nature of wireless channels, security and privacy are of utmost concern for future wireless technologies. However, securely transferring confidential information over a wireless network in the presence of adversaries still remains a challenging task. As one of the most important aspects of wireless communication security, Physical Layer (PHY) security has started gaining research attention in the past few years. In wireless PHY security, the breakthrough idea is to exploit the characteristics of wireless channels such as fading or noise to transmit a message from a source to an intended destination while trying to keep this message confidential from passive eavesdroppers. Unlike cryptographic methods, no computational constraints are placed on the eavesdroppers. Benefiting from information-theoretic studies in cooperative relaying communications, relaying strategies have also recently received considerable attention in the context of PHY security over wireless networks. Specifically, in wireless PHY security, relay nodes can be used as trusted nodes to support a secured transmission from a source to a destination in the presence of one or more eavesdroppers. This thesis studies a wireless relay network in which a source node wants to communicate securely to a destination node in the presence of an eavesdropper under the aid of an amplify-and-forward (AF) relay operating in full-duplex (FD) mode for further security enhancement. The focus is on the optimal power allocation (PA) schemes to maximize the secrecy rate in different wireless environments.;The first part of the thesis considers the problem of optimizing the PA at the source node and the relay node to achieve the secrecy capacity for slowly varying fading channels. Under this consideration, the optimal PA problem is shown to be quasi-concave. As such, the globally optimal power allocation solution exists, and it is unique. A simple bisection method for root finding can then be used to obtain the optimal PA scheme. To further provide an insight on the solutions, the method of dominant balance is applied to analyze the secrecy capacity and PA schemes in different high power regions. It is then demonstrated that full PA at the relay is only needed when the power at the relay is sufficiently small compared to the power at the source. Comparison with half-duplex (HD) relaying also shows that FD relaying can achieve a significantly higher secrecy capacity.;In the second part of the thesis, the study is extended to ergodic fading channels where the channel gains are assumed to be available at the receivers but not the transmitters. Due to the presence of fading, analysis on secrecy rate are normally very challenging because of the lack of an insightful method to calculate secrecy rate in closed-form. To this end, a novel method to calculate the expectation of an exponentially distributed random variable is first proposed. By exploiting this calculation, the ergodic secrecy rate of the system can be then established in closed-form. The optimal PA scheme and the corresponding secrecy capacity are then studied. Numerical results also reveal the superiority of FD over HD relaying in ergodic fading.