DDPG-Based Task Offloading in Satellite-Terrestrial Collaborative Edge Computing Networks
Qing Dong, Xiaodong Xu, Shujun Han, Rui Liu, Xuefei Zhang
Abstract
In remote areas, terrestrial cellular networks are usually sparsely deployed, making it difficult for the terrestrial base station (TBS) to meet demands of all users. Hybrid satellite-terrestrial edge computing architecture based on low-orbit (LEO) satellites aggregates the rich computing and caching resource of cloud servers to the network edge, which can provide computing services to users in remote area. Existing research focuses on offloading tasks to the network edge or cloud server, ignoring the user's own computing power. In this article, we formulate a LEO satellite-terrestrial collaborative edge computing (LSTEC) network with a three-tier computing architecture, including a device layer, an MEC layer, and a satellite layer, aiming to minimize the weighted sum of latency and energy consumption. Due to the high complexity of modeling problem and fast fading of the satellite-terrestrial channel, we propose Deep Deterministic Policy Gradient (DDPG)-based task offloading (DDPGTO) algorithm for joint computation offloading decision and resource allocation. Simulation results demonstrate that the proposed scheme outperforms existing schemes in terms of latency and energy consumption utility.