Research On User Selection And Privacy Protection In Mobile Crowd Sensing

The rise and popularization of mobile terminals equipped with sensor devices such as smart phones,tablet computers,and in-vehicle sensing devices makes sensing data easily available whenever and wherever.Accordingly,the MCS(Mobile Crowd Sensing)technology has been gradually becoming an effective manner of real-time sensing,collection and sharing of information.Different from the traditional static sensor network,MCS does not need to deploy numerous sensor nodes in advance.And it only needs to use the mobile devices equipped with sensors carried by ordinary users as the basic sensing unit to sense the surrounding environment information to realize the collection of sensing data.User selection and privacy protection are the key issues in the research of MCS,where user selection refers to selecting appropriate sensing users to perform sensing tasks under certain constraints to achieve a certain optimization goal and privacy protection protecting users' privacy information,such as identity,location and data privacy in the process of sensing and sending the sensing data so as to prevent users from refusing to participate in sensing tasks for fear of privacy leakage.The main research works of this paper are as follows:(1)Due to the fact that all tasks of mobile crowd sensing require users to collect data,the quality of which collected by users determines the whole quality of sensing tasks,and various types of tasks require distinct sensors and distinct ways of collecting data,which making the quality of users' sensing of various types of tasks inconsistent.Based on the analysis,this paper proposes a user selection method based on the quality of historical task data.The method classifies tasks so that each type of task completed by users will have corresponding data quality.When selecting users,the platform chooses users according to the historical data quality of users' completing this type of task.Meanwhile provides a calculation method for updating the user's historical task data quality by dividing the sensing data quality's evaluation into absolute data quality's evaluation and relative data quality's evaluation.(2)Based on the VCS(Vehicular Crowd Sensing)system,this paper proposes a privacy protection scheme based on the anonymous mechanism and the RSU(Road Side Unit)agent.This scheme uses RSU as the vehicular agent.The vehicle encrypts its sensing data with a platform's public key and sends it to the RSU in a pseudonymous identity after collecting the sensing data.The RSU transmits the encrypted sensing data to the platform with its own identity after completing the anonymous authentication of the vehicle identity.In this scheme,neither the RSU nor the platform can obtain the user's real identity information,and other entities except the platform,including RSU,cannot obtain the data information reported by the vehicles,thus realizing the privacy protection of the user's identity and the secure transmission of data;the schema also supports privacy protection in the process of exchanging rewards.Except for the TA(Trust Authority),neither the RSU nor the platform can obtain the reward information corresponding to the vehicles.