Research On Spectrum Sensing And Resource Allocation In Cognitive Relay Networks

The remarkable growth of wireless communication technology over the last decade demonstrates a vast and increasing demand for radio spectrum. Within the current spectrum regulatory framework, most frequency bandwidths are exclusively allocated for licensed services, unlicensed users are not allowed to utilize these bandwidths although they have been shown to be under-utilized in vast temporal and geographic dimensions with traditional static spectrum allocation strategy. Therefore, the cognitive radio technology has been proposed for solving the contradiction between the spectrum scarcity and the spectrum under-utilization. On the other hand, cooperative relaying technology is one of the key technologies in improving the spectrum utilization because it has been shown effective to improve the capacity, promote the transmission reliability and increase the coverage of the wireless communication networks. The combination of cognitive radio and cooperative relaying technology can effectively improve the performance of the wireless network system and obtain higher spectrum utilization, then, cognitive relay networks emerge and become a research hot spot in the field of wireless communication. As a prerequisite for cognitive radio, spectrum sensing should exploit the opportunity of unoccupied frequencies for cognitive relay networks, then by optimally distributing limited resources, the best quality of service for cognitive relay networks could be achieved while minimizing interference to primary users.Based on the intensive study on the existing literatures of cognitive relay networks,the present thesis delves into the issues of cooperative spectrum sensing and joint resource allocation within cognitive relay networks. Taking the cooperative relaying technology as the main means, the thesis solves two fundamental issues in cognitive relay networks:1) devise the cooperative spectrum sensing algorithms with high reliability and accuracy to provide reliable spectrum access opportunities as much as possible for the cognitive relay networks; 2) design the efficient joint resource allocation algorithms to improve the utilization efficiency of the spectrum hole and increase the system capacity of the cognitive relay network. The main research works and outcomes of this thesis can be summarized into the following aspects:First, by considering the effect of the Rayleigh fading, this thesis proposes two novel cooperative spectrum sensing algorithms based on virtual MIMO with out-of-band control channel. The first algorithm utilizes transmission diversity to combat with the adverse effect of channel fading. By regarding multiple secondary users as a virtual antenna array and adopting an efficient space-time code, this algorithm can increase the diversity of detection and improve the performance of cooperative spectrum sensing. Meanwhile,by considering the fading of reporting channels and the reliability of secondary users,the second algorithm targets at optimizing the weights of all secondary users in decision fusion. It adopts randomized space-time coding to improve the sensing reliability under channel fading, i.e., reduce the probability of reporting errors given an arbitrary number of secondary users, and incorporates an optimal weighted fusion rule to improve the detection reliability. Following this algorithm, we finally derive the optimal detection threshold for energy detection numerically. The simulation results show that the The proposed algorithms improve the accuracy and reliability of cooperative spectrum sensing,so, the cognitive relay networks can get more reliable spectrum access opportunities.Next, while out-of-band control channels can avoid the interference to the licensed users, it requires extra interference-free spectrum resources and incurs additional complexity for channel resource management. To remedy this, this thesis proposes a novel cooperative spectrum sensing algorithm based on selective-relay with in-band control channel, which does not rely on out-of-band control channel. This algorithm adopts a selective-relay signal transmission mechanism based on initial detection results to reduce the interference from secondary users to the primary user and also an optimal weighted fusion rule with the detection reliability test to take the reliability of secondary users into consideration. Following this algorithm, we again derive the optimal detection threshold of energy detection and also study its detection performance by jointly considering the detection phase and the initial reporting phase during cooperative spectrum sensing.Numeric simulation shows that the proposed algorithm ensures the cooperative spectrum sensing performance, effectively save channel resources, improve the utilization rate of licensed spectrum and reduce the complexity of channel management.Then, by taking into account the positive impact of OFDM and cooperative relay technology in cognitive radio network, to better utilize the spectrum holes detected by cooperative spectrum sensing, the present thesis then proposes several joint resource allocation algorithms for OFDM-based cognitive radio networks with multiple relays in the amplify-and-forward(AF) fashion. The first algorithm can achieve an asymptotically optimality on joint subcarrier pairing, relay selection and power allocation based on the technique of dual decomposition. In order to reduce the computational complexity of joint resource allocation, we then propose a fast algorithm for those applications with limited hardware resources and without performance requirements. Finally, to balance the conflict between performance and computational complexity, the third algorithm adopts a two-step method to achieve efficient allocation of resources. It requires a lower and constant computational complexity and achieves a similar performance as the optimal joint resource allocation.Finally, besides AF-based cognitive relaying, the thesis also proposes two novel joint resource allocation algorithms for OFDM-based cognitive radio networks with multiple relays in the decode-and-forward(DF) fashion. Considering the detrimental impact of the complexity of the algorithm in practical application, this thesis proposes a fast joint resource allocation scheme. The algorithm can minimize the computational complexity in solving the problem for joint power allocation relay selection and subcarrier assignment with suboptimal performance. To remedy this, the second algorithm is then proposed to achieve a reasonable tradeoff between the computational complexity and the performance of joint resource allocation. It overcomes the adverse effect of power allocation mechanism and subcarrier allocation order in the fast algorithm and achieves a similar performance as the optimal joint resource allocation.