Hybrid Recruitment And Privacy Protection In Vehicular Crowdsensing

With the proliferation of smart sensing devices on vehicles,vehicular crowdsensing(VCS)has become an emerging paradigm.It has attracted much attention for its ability to collect data at a lower cost and on a larger scale in urban sensing applications such as traffic estimation and environmental monitoring.The mobility of vehicles allows for large-scale mobile sensing data,but it remains a challenging problem to recruit the right participating vehicles and to actively maximize the sensing benefits.In addition,the collection and processing of raw data uploaded by vehicles can result in the leakage of a large amount of sensitive information about the vehicles when performing traffic flow forecasting in VCS.Therefore,it presents a significant design challenge to balance the protection of vehicle owner privacy and the development of accurate traffic prediction in VCS.Based on previous research,the main work of this paper is as follows:(1)In this paper,a hybrid recruitment algorithm based on deep learning in vehicular crowdsensing(HR-DLVCS)is proposed to solve the problem of maximizing task completion rate under limited budget.It consists of two phases: an opportunistic vehicle recruitment phase and a participatory vehicle recruitment phase.In the first phase,a deep learning-based opportunistic vehicle recruitment algorithm(DL-OVR)is proposed to maximize the sensing task completion rate within a limited budget.It aims to recruit the most suitable vehicles to collect sensing data to complete the task according to their daily movement patterns.In the second stage,a sensing task density-based participatory vehicle recruitment algorithm(STD-PVR)is proposed to reduce the computational complexity of matching participatory vehicles with uncompleted sensing tasks.(2)In this paper,a traffic flow prediction scheme based on vehicular crowdsensing privacy protection(TFP-VCS)is proposed to solve the challenge of privacy protection and accurate traffic prediction.Firstly,location privacy protection is implemented based on local differential privacy Optimized Local Hashing(OLH)protocol to avoid the situation of untrusted third-party data collection platform leaking user privacy.Then an optimized prediction model PSO-XGBoost is used for traffic flow prediction,which can control the calculation amount of traffic flow prediction and achieve good traffic flow prediction effect.Simulation results show that the scheme in this paper can provide highly usable published datasets and accurate traffic flow prediction with guaranteed privacy.