Research On Game Algorithms Of Cooperative Spectrum Sensing Based On MCS

With the rapid development of wireless communication technology,the rapid increase of mobile devices has made the already limited spectrum resources increasingly scarce,and the fixed allocation mode makes the utilization rate of spectrum resources very low.Therefore,cognitive radio technology has emerged.Cognitive radio can sense the surrounding environment and dynamically share the idle spectrum of authorized users with unlicensed users,thereby improving the utilization of spectrum.In the cognitive radio system,the unauthorized user,that is,the secondary user,needs to detect whether the primary user's spectrum is occupied,so the spectrum sensing technology is very important.Cooperative spectrum sensing involves multiple users,and participating in sensing tasks consumes users' resources.Therefore,a suitable mechanism is needed to encourage users.It is necessary to study how to combine the incentive mechanism in crowd-sensing with spectrum sensing.This thesis mainly studies the cooperative spectrum sensing game algorithm based on MCS.By introducing incentive mechanism in cooperative spectrum sensing and considering remaining power and transmission distance of users comprehensively,the effectiveness of secondary users and the platform are optimized by optimizing the sensing time of users and the price of task based on game theory.In addition,a differential privacy algorithm based on the exponential mechanism is introduced to protect the privacy of users' bids,thereby encouraging more users to participate in sensing tasks and improving the performance of the system.The main research contents and innovations are as follows:(1)From the perspective of secondary users,a cooperative spectrum sensing game algorithm with optimal secondary user utility is proposed.The utilities of secondary users are defined by considering the transmission distance,remaining power,detection probability of secondary users and the platform budget comprehensively.The secondary user optimizes the sensing time through the game and calculates its own utility to determine whether to participate in the sensing task.Secondary users who are willing to participate will report the detection probability to the platform.The platform selects a certain number of secondary users to participate in the sensing task according to the detection probability in descending order,and calculates the final cooperative detection probability through voting fusion,and finally the platform will issue corresponding rewards to users who report.The simulation results show that this method can improve the average utility of secondary users,reduce the energy consumption of secondary users with less remaining energy,and extend the standby time.(2)From the perspective of secondary users and the platform,a crowd-sensing spectrum sensing algorithm based on differential game is designed.The utility of the platform is defined as the reward paid by the third party minus the reward paid to the secondary users,and the utility of the secondary user is defined as the reward paid by the platform minus the cost of participating in the spectrum sensing task.A non-cooperative differential game model is designed with the goal of maximizing their respective utility.By solving the feedback Nash equilibrium,the optimal strategy of the platform and the user is proved,that is,the optimal price determined by the platform and the optimal detection time determined by each secondary user.The simulation results show that compared with the fixed strategy,the platform utility of the proposed strategy can be increased by60% on average,and the user utility can be increased by 55% on average.(3)Combining the incentive mechanism of privacy protection with spectrum sensing,a reverse auction algorithm of spectrum sensing with privacy protection is proposed.The algorithm introduces the concept of differential privacy and applies the exponential mechanism to the reverse auction between the platform and the user.The platform will consider the user's reputation when selecting the winning users,and uses a linear scoring function based on the exponential mechanism to select the best user set.The simulation results show that the incentive mechanism can protect user privacy and optimize the detection probability.