Research Of Collusive Csdf Attack And Its Defense Algorithm In Mobile Crowd Sensing

The rapid development of mobile terminal equipment and sensor technology has given birth to an emerging sensing paradigm of Io T—Mobile Crowd Sensing(MCS).MCS uses mobile terminal equipment held by mobile users as the basic sensing node to collect and upload sensing data,thereby achieving some large-scale and complex sensing tasks.However,the system environment of the MCS is more open,which makes it face a series of new challenges.Malicious MCS workers conduct the crowd sensing data falsification(CSDF)attack and report false sensing data to the mobile crowd sensing service provider(SP)can lead to a wrong decision in the data aggregation.In this thesis,the problems of collusive CSDF attack in mobile crowd sensing are analyzed in depth.Aiming to the characteristics of MCS workers reporting sensing data in "0-1" mode in the data collection,a collusive CSDF attack defense mechanism based on binary-minmaxs clustering analysis—BMCA is proposed.The research focus and innovations of this paper are as follow :(1)Noting the duality of workers' sensing behavior,we propose a defense scheme called BMCA from the design idea of binary-minmaxs clustering analysis to suppress collusive CSDF attack.In the BMCA scheme,the logic AND operation corresponding to the type of “1” and “0” historical sensing behavior is used to measure the similarity between any two workers.Based on this,we find the feature that collusive CSDF attackers usually hold high trust value and a low variance in their similarity vector,a new binary-minmaxs clustering algorithm is designed to detect collusive CSDF attackers.The BMCA scheme can perfect trust evaluation to prevent the trust value growth of collusive CSDF attackers,and thus enhance the accuracy of trust evaluation.(2)To verify the performance of the BMCA defense scheme against collusive CSDF attack,the success rate of an attack,the changes of trust values and trust errors of collusive attacker under BMCA and Bayesian trust mechanism are simulated and compared.The simulation results show that with the increase of the number of MCS activities,the trust value of collusive attackers in BMCA scheme continuous to decline,and soon falls below the trust threshold,resulting in a smaller trust error.Therefore,the collusive attacks are difficult to succeed.(3)Verifying the proposed BMCA defense mechanism in the Io T idle spectrum sensing scenario.First,introducing the BMCA defense mechanism to improve the trust weighting of traditional hard decision criteria such as "AND","OR",and "Majority",ahard decision data fusion algorithm is developed to suppress collusive CSDF attack.Meanwhile,the specific implementation process of the algorithm is designed.Then,based on the false alarm probability and detection probability,the ROC operating characteristic curves for improved "AND","OR",and "Majority" hard decision criteria of BMCA defense mechanism as well as the attack success rate of the collusive CSDF attackers in the hard decision data fusion algorithm are simulated and analyzed.Finally,simulation results show that the BMCA defense mechanism enables the hard decision data fusion algorithm to have a higher defense performance against collusive CSDF attack.Therefore,the BMCA defense mechanism has a better performance in suppressing collusive attacks in the Io T idle spectrum sensing scenario.