Research On The Key Technology For Resource Management In Cognitive Radio Networks

In recent years, with the rapid development of information technology and wireless applications, the future wireless communications face the problem of spectrum scarcity. Although the fixed spectrum allocation could avoid the interference between wireless networks, its spectrum efficiency is very low due to the uneven and suddenness of the distribution of users and services. Cognitive radio technology improves the spectrum utilization in dimensions of time, frequency and space without changing the original bandwidth through dynamic spectrum allocation. Cognitive wireless networks provide cognitive users QoS support without affecting the normal communication of the primary users by resource management techniques. This thesis focuses on three aspects of power allocation and user scheduling, the random access technique with QoS assurance and MAC protocol with link maintenance design. The main content and innovation of the thesis are illustrated in three aspects and summarized below.1. To maximize the network capacity and cognitive users’ satisfaction, this thesis proposes an iterative algorithm for power allocation and user scheduling in overlay spectrum sharing cognitive radio networks. First, we analyze the impact of spectrum sensing technology to the resource management, and the inter-network interference caused by side-lobe leakage. Secondly, multi-objective convex optimization model is established to maximize the cognitive capacity and the cognitive users’satisfaction under the total power constraint of the AP and the interference constraint of the primary users. Then, interactive algorithm of power allocation and user scheduling is obtained by Lagrangian dual optimization method. Finally, computer simulation is used to verify the network performance of the proposed algorithm. 2. To provide cognitive users with continuous requirement a higher access probability, this thesis builds quasi-synchronous multi-channel access priority scheme in self-organizing cognitive wireless networks. First, since different cognitive users with different bandwidth requirement need different sensing time, time frame structure with asynchronous contention is designed. Meanwhile, the cognitive users could obtain a high access probability by priority conversion during the contention for common control channel. Then, in saturated cognitive wireless networks, a first-order finite-state discrete markov chain is modeled to describe the time slots’state, and then we analyzed the system throughput and access delay performance. Finally, in unsaturated cognitive wireless networks, a three state discrete markov chain is modeled to describe the state of the cognitive users, and then we analyze the system throughput, the access latency and continuous access probability of the modified proposed scheme.3. To enhance the cognitive user’s communication continuity, this thesis designs a MAC protocol with link maintenance provision in self-organizing cognitive wireless networks. First, the father spectrum list is designed and we obtain its life-cycle. The father spectrum list is divided into different sub-channels with different functions where redundant sub-channels and backup sub-channels are used to make up the package lost caused by spectrum mobility. Secondly, time frame structure with adaptive sensing phase is designed to lengthen the transmission time. Then, three access modes are designed according to the cognitive users’states, and the communicate continuity is guaranteed by multiple access opportunity. Finally, system performance of the system throughput, access latency, communicate continuity and inter-network interference and the overhead of the sensing time are analyzed by both theory and computer simulation.