Online Worker Scheduling for Maximizing Long-Term Utility in Spatiotemporal Crowdsensing
Jiajun Wang, Xingjian Ding, Jianxiong Guo, Zhiqing Tang, Deying Li · 2024
With the continuous development of mobile networks and sensing devices, spatiotemporal crowdsourcing has gradually become a new intelligent sensing paradigm for data acquisition and sharing in the Industrial Internet of Things. How to reasonably allocate tasks to workers in a dynamic environment to maximize the platform utility has become a research hotspot. Many past works have made great efforts in this regard, but most of them only consider the long-term constraints of resources, and ignore the short-term ability constraints of workers to provide resources. In this paper, we consider a platform-centered online spatiotemporal crowdsourcing system, where mobile workers have long-term and short-term resource constraints, while the platform has a long-term budget constraint. We aim to find an online worker scheduling scheme to maximize the platform’s long-term utility without violating the constraints of workers and the platform. To address the problem, we first transform the long-term utility maximization problem into a real-time utility maximization problem by leveraging the Lyapunov optimization. Then, we design a centralized algorithm based on Markov approximation to solve the real-time optimization problem. Furthermore, we demonstrate that our proposed approach can achieve near-optimal performance for our problem. Finally, we evaluate our designs by numerical simulation experiments, and the results demonstrate the effectiveness of our algorithms.