Balancing Worker Utility and Recruitment Cost in Spatial Crowdsensing: A Nash Game Approach

Leveraging the power of mobile crowdsensing (MCS) to achieve unparalleled coverage of tasks by utilizing the crucial spatio-temporal co-relation between the workers and tasks results in the distributed sensing system named spatio-temporal crowdsensing system (STCS). An STCS outsources sensing tasks to ubiquitous mobile computing devices to exploit their multi-modal sensing capabilities, collects the sensing results and process those to provide meaningful services. While realizing sustainable crowdsensing services, an STCS struggles with two conflicting objectives- maximizing the assignment quality and minimizing the recruitment cost. The first one requires assigning highly reputed workers in the spatio-temporal task blocks which results in higher recruitment costs. Thus a tradeoff between the aformentioned objectives is required. Existing works bring this tradeoff either by formulating a single objective as the linear weighted sum of both of the objectives or incorporating one of these as a constraint to another, yielding to an unfair and skewed tradeoff. In this work, we formulate the worker recruitment problem in STCS as a Nash game to attain fairness between the objectives while bringing this tradeoff. The performance of the proposed system is evaluated using MATLAB and results show its effectiveness compared to the state-of-the-art works in terms of fairness and per cost utility gain.