PROJECT TITLE :

An Efficient Prediction-Based User Recruitment for Mobile Crowdsensing - 2018

ABSTRACT:

Mobile crowdsensing could be a new paradigm in which a cluster of mobile users exploit their good devices to cooperatively perform a large-scale sensing job. One in all the users' main issues is the price of information uploading, that affects their willingness to participate in a very crowdsensing task. During this Project, we have a tendency to propose an efficient Prediction-primarily based User Recruitment for mobile crowdsEnsing (PURE), which separates the users into 2 groups such as completely different worth plans: Pay as you go (PAYG) and Pay monthly (PAYM). By concerning the PAYM users as destinations, the minimizing cost drawback goes to recruiting the users that have the most important contact chance with a destination. We first propose a semi-Markov model to determine the chance distribution of user arrival time at points of interest (PoIs) and then get the inter-user contact chance. Next, an economical prediction-based user-recruitment strategy for mobile crowdsensing is proposed to attenuate the info uploading value. We have a tendency to then propose PURE-DF by extending PURE to a case in that we have a tendency to address the tradeoff between the delivery ratio of sensing knowledge and the recruiter range according to Delegation Forwarding. We conduct extensive simulations primarily based on three widely-used real-world traces: roma/taxi, epfl, and geolife. The results show that, compared with other recruitment methods, PURE achieves a lower recruitment payment and PURE-DF achieves the highest delivery efficiency.