Energy-Efficient Task Computation at the Edge for Vehicular Services

TL;DR

Energy-efficient task computation at the edge for vehicular services addresses how to minimize total energy while meeting latency constraints in a multi-tier MEC for highly mobile vehicles. The authors analyze real-world car mobility traces from a European MNO, formulate a MDP, and develop two PPO-based MARL policies: LAPPO for static scenarios and MALAPPO for mobile scenarios. The approach achieves up to $47\%$ energy savings in static settings and $14\%$ in mobile settings compared with baseline schemes, while reducing user dissatisfaction and task interruptions. The work demonstrates practical MEC strategies that respect privacy by using base-station level location proxies and shows strong potential for sustainable, low-latency vehicular computing.

Abstract

Multi-access edge computing (MEC) is a promising solution for providing the computational resources and low latency required by vehicular services such as autonomous driving. It enables cars to offload computationally intensive tasks to nearby servers. Effective offloading involves determining when to offload tasks, selecting the appropriate MEC site, and efficiently allocating resources to ensure good performance. Car mobility poses significant challenges to guaranteeing reliable task completion, and today we still lack energy efficient solutions to this problem, especially when considering real-world car mobility traces. In this paper, we begin by examining the mobility patterns of cars using data obtained from a leading mobile network operator in Europe. Based on the insights from this analysis, we design an optimization problem for task computation and offloading, considering both static and mobility scenarios. Our objective is to minimize the total energy consumption at the cars and at the MEC nodes while satisfying the latency requirements of various tasks. We evaluate our solution, based on multi-agent reinforcement learning, both in simulations and in a realistic setup that relies on datasets from the operator. Our solution shows a significant reduction of user dissatisfaction and task interruptions in both static and mobile scenarios, while achieving energy savings of 47 percent in the static case and 14 percent in the mobile case compared to state-of-the-art schemes.

Figures (6)

• Figure 1: Multi-tier MEC architecture for V2X services.
• Figure 2: Mobility characteristics of cars in our dataset. (a) Per-hour ECDF of the bounding box diagonal traveled by cars. (b) Percentage of cars' connection time to the top20 and top5 (total and per BS) BSs, between 8 AM and 8 PM. (c) Per-hour overlap coefficient (statistics over cars).
• Figure 3: Ratio of unsatisfied users (left), offloading energy (center) and local energy (right) for LAPPO and baseline solutions in the static scenario.
• Figure 4: Offloading (left) and local (right) energy consumption as a function of $\alpha$ in the static scenario.
• Figure 5: Ratio of unsatisfied users as a function of the number of users in the static scenario.