Task Scheduling Algorithm For Mobile Crowdsensing Based On P2P Framework

In recent years,with the rapid development of wireless communication network technology,the demand of wireless applications for various sensory data is constantly increasing,people gradually have some conceptions of Mobile Crowdsensing(MCS),which makes use of people's mobile devices for data sensing.This new sensing pattern is easier to maintain and more flexible than the traditional way of deploying sensor devices.For the emerging sensing pattern of MCS,its "server-client" structure is not the best architectural patterns,because this will make the central server to deal with a large number of sensory data,including data upload,processing,storage and distribution,etc.Therefore,the exist architecture may lead to the server's serious data congestion and high operation cost.In the work of this subject,we introduced a P2P-based MCS system,which allows the sensory data to be stored directly in users' own devices,and after the device preprocessing,it can be shared among users in a peer-to-peer way.Therefore,by utilizing the computing and storage resources of a large number of users' devices,the burden of the server can be effectively reduced.However,in reality,a user's data sharing ability will be limited by the physical link layer,device battery,or weather,for these reasons,it is reasonable to set a limited sharing ability for users,so we put forward a new P2P-based MCS model with a limited sharing number,i.e.,each user can only shared his seneory data with a limited number of other users who needed.In this number-limited sharing MCS system,we main focus on the optimal task scheduling problem and game equilibrium scheduling problem of the new model.Specifically,in order to incentive some sensory users to share the data with other people who needed,this paper have proposed a data market,combined with hybrid pricing mechanism,to let users to sell their sensory data.Then,with the knowledge of evolutionary game theory,we analyzed the evolution of data market and behavior of users in detail.At the same time,we described and deduced the equilibrium of the market and the whole system,furthermore,we analyzed the stability of the system after reaching the equilibrium state.Finally,through the simulation results,we have found that the mixed price incentive mechanism can adjust flexibly,so that the game equilibrium model can achieve an ideal state with the appropriate incentive mechanism.What's more,this article also found that a reasonable incentive mechanism can help better meet the demands of all parties,without doing too much damage to the fundamentalinterests of all users.