Robust Offloading for Edge Computing-Assisted Sensing and Communication Systems: A Deep Reinforcement Learning Approach

• Published in: Sensors (Basel, Switzerland) Vol. 24; no. 8; p. 2489
• Main Authors: Shen, Li, Li, Bin, Zhu, Xiaojie
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 12.04.2024; MDPI
• Subjects: Adaptive learning; Algorithms; computation uncertainty; Data processing; Data transmission; Deep learning; deep reinforcement learning; Design; Edge computing; Efficiency; Energy consumption; Infrastructure; integrated communication and sensing; Internet of Things; Markov processes; mobile edge computing; Optimization; robust design; Signal processing; Wireless communications; Wireless telecommunications equipment; computation uncertainty; mobile edge computing; robust design; deep reinforcement learning; integrated communication and sensing
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s24082489

In this paper, we consider an integrated sensing, communication, and computation (ISCC) system to alleviate the spectrum congestion and computation burden problem. Specifically, while serving communication users, a base station (BS) actively engages in sensing targets and collaborates seamlessly with the edge server to concurrently process the acquired sensing data for efficient target recognition. A significant challenge in edge computing systems arises from the inherent uncertainty in computations, mainly stemming from the unpredictable complexity of tasks. With this consideration, we address the computation uncertainty by formulating a robust communication and computing resource allocation problem in ISCC systems. The primary goal of the system is to minimize total energy consumption while adhering to perception and delay constraints. This is achieved through the optimization of transmit beamforming, offloading ratio, and computing resource allocation, effectively managing the trade-offs between local execution and edge computing. To overcome this challenge, we employ a Markov decision process (MDP) in conjunction with the proximal policy optimization (PPO) algorithm, establishing an adaptive learning strategy. The proposed algorithm stands out for its rapid training speed, ensuring compliance with latency requirements for perception and computation in applications. Simulation results highlight its robustness and effectiveness within ISCC systems compared to baseline approaches.