Research Of Trusted Cooperative Spectrum Sensing Technologies Based On Reputation

With the rapid development of wireless communication technology, licensed spectrum resources are increasingly scarce and its utilization rate is very low. The redistribution of spectrum resources can be very limited. Cognitive radio(CR) technology dynamically finds idle spectrum, which effectively improves the utilization of the available spectrum resources. Cooperative spectrum sensing(CSS) technologies make multiple users collaborate with each other, and thus improve the performance of single-user spectrum sensing. However, multi-user collaboration is vulnerable to the spectrum sensing data falsification(SSDF) attack so that the entire performance of CSS will be seriously affected.Based on the cooperative spectrum sensing technology analysis, reputation mechanism is introduced in this paper to study trusted technologies which can defense SSDF attack effectively. The main work of this paper is as follows.Firstly, this paper analyzes cooperative spectrum sensing technologies and data fusion algorithms, especially discussing malicious users’behavior and SSDF attack in detail. Based on this, some reputation-based CSS schemes are summarized.Secondly, this paper proposes double thresholds of reputation-based hard decision fusion(RHDF) algorithm.In RHDF algorithm,the fusion center merely receives sensing reports from trusted secondary users by introducing the priority-half fusion method based on double thresholds. Thus, RHDF can improve the efficiency of fusion decision and avoid the impact of malicious users.The simulation results show that RHDF algorithm can more effectively defense SSDF attacks and ensure better cooperation spectrum sensing performance.Finally, based on hierarchical clustering ideas, this paper proposes a multi-hop cooperative sensing networks model and designs a reputation-based cooperative spectrum sensing simulator(RecssSim) by Matlab GUI tools in multi-hop networks.This simulator has a good expansibility and can analyze the security of some CSS schemes embedded by reputation mechanism in the case of three SSDF attacks modes(always-free, always-false and always-busy).