Research On Collaboration Spectrum Sensing Algorithm For Cognitive Radio

With communication devices such as smart phones and tablet PC popular around the world, people demand for more communication system of wireless spectrum resources. There is an increasingly prominent contradiction between a huge wireless communication business needs and the limited spectrum resources. Cognitive radio(CR) technology is an intelligent wireless communication system with a function of dynamic spectrum access, which can improve utilization of communication spectrum. In recent years, the research mainly focus on collaborative sensing algorithm, but the traditional collaborative spectrum sensing algorithm treat every secondary user's sensing result equally. In fact,different secondary user's sensing result has different reliability, so this thesis makes some research on this problem.Firstly, we introduce the basic study situation of CR and bring in spectrum sensing technology. Then the single node and multi-node algorithm in collaborative spectrum sensing are introduced and the advantages and disadvantages of single node are analyzed. Due to the shortcoming of single node, this thesis selects multi-node collaborative spectrum sensing algorithm as the research focus. The traditional multi-node collaborative spectrum sensing algorithm includes AND, OR and K rank, which are analyzed and simulation results demonstrate the characteristics of every algorithm.Secondly, we propose a collaborative spectrum sensing algorithm based on double threshold and D-S evidence theory. Considering the unreliability of secondary users'sensing result, we combine double threshold and D-S evidence theory and get a final rule fusion by using OR algorithm. Then it is contracted with algorithm based on single threshold and double threshold respectively. The simulation results show this algorithm can be effectively applied to the sheltered terminals, shadow fading and noise in the communication systems with uncertainty.Finally, the traditional D-S evidence is improved. When the conflict is larger, the improved algorithm can be used in fusion decision and gain a more reliable spectrum sensing results. In the CR communication system, not only sensing channel envirornment quality greatly affects the performance of collaborative spectrum sensing, but also the performance of notice channel is also a challenge. To solve this problem, the grouped collaborative spectrum sensing algorithm based on double threshold and D-S evidence theory is researched. All secondary users are divided into groups to detect the existence of primary user. The simulation results show grouped collaborative spectrum sensing algorithm can get a better spectrum sensing ability. In addition, this algorithm can reduce transmission error probability in notification channel to a certain extent by increasing the number of secondary users in a group and reduce the notification channel fading effect on the performance of the collaboration spectrum sensing.