Computation Offloading Management in Vehicular Edge Network under Imperfect CSI

• Published in: 2019 IEEE 2nd International Conference on Information Communication and Signal Processing (ICICSP) pp. 199 - 203
• Main Authors: Shang, Licheng, Wang, Xinhong, Wang, Ping, Ngoc Van, Nguyen
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2019
• Subjects: Autonomous vehicles; computation offloading; Delays; Heuristic algorithms; imperfect CSI; MEC; power control; Power system reliability; Probability; Reliability; Signal processing algorithms; Simulation; Vehicular ad hoc networks; vehicular network; Videos
• DOI: 10.1109/ICICSP48821.2019.8958488

Recently, image recognition and video processing play important roles in the rapidly expanding field of autonomous driving. However, the computing capability of vehicles cannot meet the latency requirement of these tasks. Mobile Edge Computing (MEC) is envisioned as an effective paradigm to augment the computing capability of vehicles for dealing with computationally intensive tasks. And the imperfect channel state information (CSI) due to the high mobility of vehicular environment need to be properly accounted for when designing offloading strategy for vehicular network. In this paper, we formulate a problem to minimize the cost of computation and communication under imperfect CSI considering the processing delay, return delay, priority of tasks and energy consumption. The problem can be solved as a modified graph matching problem and we optimize the offload strategy. A computation offloading management (COM) algorithm is proposed to solve this optimization problem. We focus on minimizing the overall overhead while guaranteeing the reliability of each vehicle-to-vehicle (V2V) communication link. Numerical results demonstrate that our proposed algorithm can reduce total overhead and guarantee low outage probability.