Optimal Collaborative Uploading in Crowdsensing with Graph Learning

Yao Zhang,Tom H. Luan,Hui Wang,Liang Wang,Zhiwen Yu,Bin Guo,Yimin Zhao

DOI: https://doi.org/10.1109/icc45041.2023.10279250

2023-01-01

Abstract:It is pivotal and challenging for crowdsensing systems to guarantee the reliable uploading of sensory data from source devices (workers) to a centralized platform, in order to process sensing tasks accurately and fast. On one hand, with limited communication resources, uploading a massive amount of sensory data is not cost-effective. On the other hand, the disruption of uploading is inevitable because of stochastic network environments and worker dropout, resulting in extra wasting of resources. To address that, we focus on a collaborative uploading scenario and propose to reduce the uploading latency of sensory data by adaptive data allocation while retaining data integrity at the destination. A key technical challenge is to identify proper collaborative paths such that corresponding data allocation and uploading are reliable enough. As such, we formulate a joint optimization problem with the minimization goal of uploading latency by considering both path selection and data allocation. To mine helpful information from unstructured topology-aware data, we propose a new diffusion graph convolution module by forming information aggregation based on the diffusion process that characterizes the stochastic correlation of devices. After transforming the original problem into a primal-dual problem, an algorithm is then developed by adapting Advantage Actor-Critic (A2C) framework embedded with the diffusion graph convolution module. With extensive experiments, it is validated that the newly developed algorithm improves collaborative uploading by reducing uploading latency and also stabilizing the queue state of intermediate devices, compared to existing heuristic and learning-based methods.