Quality-Aware Task Assignment in Opportunistic Network-Based Crowdsourcing

Mobile crowdsourcing in opportunistic networks outsources location-based tasks, such as taking photos and surveying Wi-Fi signal characteristic at points of interests, to a crowd of workers. The performance of tasks assignment is generally evaluated by the makespan. However, not only the makespan, but also the quality of performed tasks is important. Therefore, in this paper, we propose two task assignment schemes. One is the quality-aware task assignment (QA-TA) that tries to maximize the quality of tasks for given a deadline; the other is the minimum quality threshold task assignment (MQT-TA) that tries to minimize the makespan for a given minimum quality requirement. To this end, we apply the optimal stopping, which is one of the widely used techniques in mathematics, to the algorithm designs. The simulations using real mobility traces demonstrates that the proposed schemes successfully achieve their design goals.