Participants Selecting Method Based On Mobile Crowd Sensing

Mobile crowd sensing is a hot research topic in the fields of mobile computing and mobile Internet.Its basic idea is to collect the surrounding sensing data with the help of the mobile nodes,and apply them to a variety of innovative applications.The quality of sensing data is the key to mobile crowd sensing system.However,Participants Selecting Method and incentive constraints are two important factors affecting the quality of sensing data.The goal of participant selection strategy is to choose a limited number of optimal mobile nodes for collecting the most covered comprehensive sensing data.Starting from this goal,this thesis introduces the theory of multi-objective decision based on weighted entropy,and put forward a participant selected method through employing multi-objective decision theory based on weighted entropy.It makes comprehensive evaluation to sensing ability,the transition probability and incentive requirements of nodes for selecting the highest quality sensing nodes to finish the sensing tasks.Firstly,a mobile crowd sensing system model is built.It is mainly composed of the individual or entity,data server center and mobile nodes which releases the perceptive tasks.Secondly,a mobility model of mobile nodes is established according to the Continuous Markov chain.Finally,MODSM algorithm is proposed through the introduction of multi-objective decision theory based on weighted entropy.In order to verify the performance of the algorithm,python script language is used for experiment simulation,and the designed method is proved to be effective.