Mobile Crowdsensing User Recruitment Algorithm Based on Combination Multi-Armed Bandit

In the mobile crowdsensing task assignment, under the premise that the data platform does not know the user's perceived quality or cost value, how to establish a suitable user recruitment mechanism is the key issue that this article needs to solve. It is necessary to try to ensure the efficiency and profit maximization of the mobile crowdsensing platform. Therefore, a mobile crowdsensing user recruitment algorithm based on a Combined Multi-Armed Bandit (CMAB) is proposed to solve the recruitment problem of known and unknown user costs. Firstly, the user recruitment process is modeled as a combined multi-arm bandit model. Each rocker is represented by a different user's choice, and the income obtained represents the user's perceived quality. Secondly, the Upper Confidence Bound (UCB) algorithm is proposed to update the user's perceived quality according to the completion of the task. Users' perceived quality values are sorted from high to low, and then the user with the largest ratio of perceived quality to recruitment cost is selected under budget conditions, tasks are assigned, and their perceived quality is updated. Finally, A large number of experimental simulations based on real data sets are carried out to verify the feasibility and effectiveness of the algorithm.