Research On Cross-layer Resource Allocarion In Secure Cognitive Radio Networks For Delay QoS Assurance

The rapid development of wireless communication technology has brought great convenience to the modern society, but spectrum resource scarcity is becoming more and more serious now. Cognitive Radio (CR) can effectively improve spectrum efficiency and is now raising a worldwide research upsurge. In addition, owing to the open nature of wireless transmission media, the security of information transmission in CR networks is also an important problem to be solved. Using the physical layer security technology, the reliability of secure transmission in CR networks can be greatly improved. However, the research of physical layer information security with the information-theoretic perspective has some limitations, since it may ignore the delay quality of service (QoS) requirements of real-time multimedia applications.To solve problems about security and delay QoS, we research on the effective cross-layer resource assignment policies in CR networks thoroughly. With inducing the convex optimization theory and secure effective capacity concept, we formulate and solve the cross-layer resource assignment optimization problem. We also verify the results by the corresponding simulations. The main work and innovations are as follows:(1) Delay-QoS-driven cross-layer analysis structure in secure CR networks is briefly sketched. Firstly, we introduce the concept of effective capacity which is in favor of studying the statistical QoS performance of wireless transmission. Secondly, we discuss the relationship between secure effective capacity and delay QoS performance in a secrecy system. Thirdly, we propose a cross-layer resource optimization method for the secure CR networks.(2) Based on the secure effective capacity model, the cross-layer resource assignment scheme for delay QoS assurance is proposed for the single-user secure CR networks, which works in underly spectrum sharing mode. First of all, assuming that the channel state information (CSI) of all channels is completely known, we formulate two power optimization problems, one is to maximize the secure effective capacity, the other is to minimize the interference temperature. Then, the corresponding optimal solutions can be found by using convex optimization method. The simulation results demonstrate that the above power control policies have strong superiority. Finally, assuming that the CSI of the primary channel is uncertain, we investigate the robust solution for power control with a view to maximize the secure effective capacity. The simulation results show that the robust design is more practical.(3) This thesis further studies the more complex multi-user secure CR networks, where the time-division multiple-access (TDMA) is adopted as the multiple access scheme. With the perspectives of fairness, delay and security, we formulate an optimization problem by minimizing the β-fair cost functions. Then, a cross-layer power and time optimization strategy is proposed. Relying on the convex optimization theory and stochastic optimization method, the above optimization problem can be solved. Simulation results show that the cross-layer resource assignment policy can balance the tradeoff between energy efficiency and fairness effectively. Moreover, this policy is superior to other strategies obviously. In addition, assuming that the CSI of the primary channels is not known perfectly, the thesis addresses a robust design for resource optimization which makes the theoretical research closer to practical application.