Research On Dynamical Clustering Cooperative Spectrum Sensing Technology In Cognitive Radio Networks

With the increasing progress of science technology and society, wireless communication technology has made rapid development and been widely used, making the radio spectrum resource that is necessary for wireless communication systems a kind of scarce and valuable resource. Up to now, most countries in the world adopt the fixed allocation method in managing the radio spectrum resource. Most of the spectrum resource is authorized to communications, television, broadcasting, remote control and other services and applications. While only a small portion of radio spectrum is used for other network applications without authorization. Along with the fast development of wireless network technology and our increasing requirement of information interaction, more and more wireless applications require using the radio spectrum resource, resulting in the shortage of unlicensed spectrum resource. Meanwhile, according to some researches, the utilization of licensed spectrum is very low and many spectrum bands are usually idle. Therefore, the contradiction between the limited available spectrum resource and the inefficient utilization of spectrum resource has become a critical issue to be addressed in wireless communications.In order to maximize the utilization efficiency and ease the tension condition of radio spectrum, cognitive radio (CR) technology is proposed and widely studied. Spectrum sensing technology is one of the key and important technologies of cognitive radio. To overcome the hidden terminal problem caused by local spectrum sensing, cooperative spectrum sensing (CSS) technology which can improve the system performance significantly through joint detection becomes a hot point of spectrum sensing research. Considering the actual environment of cognitive radio networks (CRN), this thesis makes some innovation and improvement of cooperative spectrum sensing and proposes several dynamical clustering schemes to gain higher detection performance and save bandwidth. The main research work of the thesis is as follows. Firstly, a robust space-time block coding (STBC) based dynamical clustering CSS scheme is proposed. Considering the inter-user channels between cognitive users are non-ideal, we propose a more robust dynamical clustering method. After analyzing the quality conditions of inter-user channels, this method groups the two users with better inter-user channels. This scheme can overcome the drawbacks of traditional clustering methods based on users’location or user selection, and is more robust. This thesis presents the concrete grouping process. Simulation results show that the proposed scheme can further improve the system performance in realistic communications.Secondly, considering that the environment and location of users are different in realistic situations, the users with high signal to noise ratio (SNR) can obtain more reliable detection results, whereas the local spectrum sensing results of the users with low SNR are not quite reliable. Thus this thesis proposes a weighted CSS scheme based on STBC. After the robust clustering process, some weight factors are assigned to cognitive users according to their detecting channel SNRs. This scheme could enhance the contribution of reliable user in detection and improve the system performance via reducing the bit error rate (BER) of reporting channels.Finally, considering the control bandwidth is usually limited, to reduce the number of bits transmitted, a dynamical clustering CSS scheme which could ensure the detection performance is proposed in this thesis. The scheme selects users with better reporting channels as cluster heads and groups the users according to their inter-user channels with the heads. The detecting results of cluster members are sent to heads and then forwarded to base station (BS) by the heads. To reduce the number of bits, double thresholds method is applied in each cluster to only allow the users with reliable detection performance to send results. Simulation results show that the proposed scheme can guarantee the system performance and reduce system overhead.