Cooperative Relay Networks with Target Security Constraints: Reliability Analysis and Power Optimization
MetadataShow full item record
In this thesis, we investigated the problem of eavesdropping control and power optimization in amplify-and-forward cooperative relay networks. When a confidential message is transmitted from the source to the destination via the help of the relay, the system may suffer from a potential security breach or information leakage due to the possibility that the relay may be able to decode the relayed signal. Comparing to the traditional methods such as public key agreement and authentication, in this work, we addressed the security problem by considering power allocation in order to reduce information leakage. By keeping the signal-to-noise ratio (SNR) at the relay who may try to decode the relayed messages below a certain threshold, we developed an optimized power allocation scheme to prevent such security vulnerability, while maintain or improve the system overall reliability performance. First, we introduce a performance metric called Secure Reliability Rate that combines with the target security constraint and the reliability requirement for the AF cooperative relay networks. Then we developed an asymptotically tight approximation for the Secure Reliability Rate which is valid nearly at any given total power budget. An interesting observation is that increasing the total power budget will not necessarily improve the whole system performance in terms of the Secure Reliability Rate, which is different from some common understanding. Based on the approximated expression, we used the numerical search techniques to obtain the optimum power allocation for the AF cooperative communication system with any targeted security constraints. It turns out that the equal power allocation scheme is far from optimal to satisfy both of the target security constraint and reliability requirement. We showed that the optimum power allocation scheme achieves and maintains the best possible system performance as the total power budget increases. Moreover, we observed that the optimum power allocation scheme also depends on the channel link quality. Extensive simulation and numerical results are provides to validate the theoretical analysis.