Capacity And Energy-Efficiency Performance Research In Cognitive Radio Networks

With the development of wireless communication technology, spectrum resources are becoming scarcer in recent years.’ How to make more effective use of radio spectrum is an urgent technology problem for spectrum scarcity mitigation. The survey showed that the utilization of authorized spectrum was remarkably low. Out of this consideration, the researchers put forward the concept of cognitive radio (CR). The basic idea of CR is to realize communications function among unauthorized users by exploiting "spectrum holes" while not affecting the use of spectrum by authorized users. As a result of CR, spectrum utilization and system capacity are increased largely.Nowadays, in order to realize the sustainable development, many researchers have focused on Green communication. However, there is little work about the green communication in cognitive radio networks (CRNs), but compared to traditional networks, CRNs need extra energy to sense the spectrum, therefore, it is very important to investigate the green communication in CRNs.This dissertation focuses on the capacity and energy-efficiency performance in CRNs. This dissertation investigates the joint optimization of spectrum sensing time and data transmission power problem in single-carrier CRNs. Though the objective is to maximize the CR users’ average throughput, it’s hard to analyze and solve the original problem. Thus, this dissertation proposes an approximated model and a modified golden search algorithm which can obtain the optimal performance for the approximated model. Simulation results show that the solution to the approximated model is close to the original one, moreover, the performance obtained by the proposed algorithm is better than that obtained by traditional algorithm.This dissertation also studies the energy-saving resource allocation in OFDM-based CRNs, and proposes a two-step resource allocation algorithm, i.e., executes the subcarrier allocation with fixed power allocation and then basing on the subcarrier allocation result, proposes an optimal power allocation algorithm. Simulation results show, the algorithm proposed in this dissertation can obtain optimal performance, and compared to non-cognitive radio scenarios, energy-saving algorithm in this dissertation can gain high interference reduction while sacrificing a little power consumption.This dissertation further investigates the energy-efficiency problem in OFDM-based CRNs. The problem is to maximize the energy-efficiency measured using the "throughput per Joule" metric subject to the total transmit power and interference constraints. It is then transformed into an equivalent convex problem using parametric programming. Furthermore, an optimal iterative algorithm based on convex optimization theory and parametric programming is proposed. The numerical results show that the proposed optimal algorithm can achieve higher energy-efficiency than that obtained by solving the original problem directly because of its non-convexity. Energy-efficiency maximization can also achieve a good tradeoff between capacity and energy in CRNs.Capacity bottleneck for future communication system lies in spectrum-efficiency and energy-efficiency, therefore it is essential to investigate the capacity and energy-efficiency performance in CRNs. This dissertation conduct the research with the support from973issue "Cognitive radio network behavior analysis and network performance research", moreover related works have been published in IEEE journals and international conferences.