Participant-Quantity-Aware Online Task Allocation in Mobile Crowd Sensing
Guisong Yang,Dongsheng Guo,Buye Wang,Xingyu He,Jiangtao Wang,Gang Wang
DOI: https://doi.org/10.1109/jiot.2023.3305034
IF: 10.6
2023-01-01
IEEE Internet of Things Journal
Abstract:Task allocation, which can be divided into offline task allocation and online task allocation, is a significant issue in mobile crowd sensing (MCS). Unlike offline task allocation, in the online task allocation scenario, since participants arrive at the service dynamically, the quantity of participants in a specific time and space is uncertain, hence it could affect the quality and efficiency of task completion. However, due to the difficulty of predicting the quantity of real-time participants in a specific time and space accurately, the existing studies of online task allocation lack deep consideration of the quantity of participants. Therefore, this paper investigates a participant-quantity-aware online task allocation problem. First of all, in view of the difficulty of predetermining the participant quantity in MCS, a fuzzy time series analysis (FTSA) method is developed to predict the participant quantity available for each task in a specific time and space. Then, according to the predicted quantity, two reasonable attributes for each task, including the task’s threshold on participant’s sensing ability and the reward provided for participants to execute the task, can be calculated separately. On this basis, considering the participant’s willingness, the participant’s sensing ability, the sensor types of the participant’s device, and the participant’s time coverage jointly, we design an online task allocation algorithm based on an improved genetic algorithm (OTAGA) to allocate appropriate set of tasks to each participant who arrives in real-time, so as to maximize the platform utility and minimize the movement cost of the participant. Simulation results show that the proposed method is effective in terms of the accuracy of prediction, the platform utility and the movement cost of the participant.
computer science, information systems,telecommunications,engineering, electrical & electronic