Abstract
By pushing computation, cache, and network control to the edge, mobile edge computing (MEC) is expected to play a leading role in fifth generation (5G) and future sixth generation (6G). Nevertheless, facing ubiquitous fast-growing computational demands, it is impossible for a single MEC paradigm to effectively support high-quality intelligent services at end user equipments (UEs). To address this issue, we propose an air-ground collaborative MEC (AGC-MEC) architecture in this article. The proposed AGC-MEC integrates all potentially available MEC servers within air and ground in the envisioned 6G, by a variety of collaborative ways to provide computation services at their best for UEs. Firstly, we introduce the AGC-MEC architecture and elaborate three typical use cases. Then, we discuss four main challenges in the AGC-MEC as well as their potential solutions. Next, we conduct a case study of collaborative service placement for AGC-MEC to validate the effectiveness of the proposed collaborative service placement strategy. Finally, we highlight several potential research directions of the AGC-MEC.
Introduction
As we transition toward the 6G era, traditional terrestrial edge computing architectures are facing unprecedented pressure. The explosion of data-intensive applications, such as autonomous driving, ultra-high-definition video streaming, and immersive AR/VR, requires low-latency processing that static ground-based infrastructure often cannot provide.
Current Mobile Edge Computing (MEC) deployments are primarily "terrestrial and static," typically tethered to base stations or roadside units. This rigidity leads to two critical shortcomings. First, fixed servers struggle with spatio-temporal dynamics, such as "flash crowds" at sporting events or shifting traffic congestion patterns. Second, there are significant infrastructure blind spots; in scenarios like natural disasters or military operations, ground-based infrastructure may be damaged or entirely absent. While Unmanned Aerial Vehicles (UAVs) have been proposed as mobile servers, their individual limitations in battery life and compute power suggest that a single-paradigm approach is insufficient. The solution lies in a unified, collaborative ecosystem: Air-Ground Collaborative MEC (AGC-MEC).
The AGC-MEC Architecture
The proposed AGC-MEC architecture integrates diverse aerial and terrestrial assets into a cohesive computing blanket. It is structured into two primary layers:
- Aerial MEC Layer: This layer utilizes "near-static" high-altitude platforms like airships and balloons for broad, reliable coverage, alongside highly mobile UAVs that provide flexible, low-latency support for localized demand.
- Terrestrial MEC Layer: This includes traditional static assets like Base Stations (BS) and Roadside Units (RSU), as well as mobile vehicles and "opportunistic" powerful mobile users who can contribute spare computational cycles.
Collaboration within this architecture is categorized into four distinct modes based on the mobility of the nodes involved: Static-Air/Static-Ground, Static-Air/Mobile-Ground, Mobile-Air/Static-Ground, and Mobile-Air/Mobile-Ground. This multi-faceted collaboration ensures that computing resources can follow the user, regardless of infrastructure availability or movement patterns.
Optimization Methodology
To validate the AGC-MEC concept, the researchers addressed a Service Placement Optimization problem, aiming to minimize the total energy consumption of user equipment (UE). This was modeled as a Mixed Integer Non-Linear Programming (MINLP) problem, which is complex due to the combination of discrete placement decisions and continuous trajectory variables.
The solution employed a hybrid approach:
- Branch and Bound (BnB): Used to manage the discrete "yes/no" decisions regarding which services to place on which edge servers.
- Successive Convex Approximation (SCA): Utilized to optimize continuous variables, specifically the 3D trajectory of UAVs to ensure they maintain optimal proximity to user clusters.
The simulation environment modeled a 200m x 200m area with 10 users. Service requests followed a Zipf distribution to reflect realistic popularity patterns, with task sizes ranging from 100 to 1000 KB.
Performance Evaluation
The case study results demonstrate that the collaborative AGC-MEC approach significantly outperforms traditional single-layer or non-optimized strategies. The proposed algorithm showed rapid convergence, reaching a stable solution within just three iterations.
Key quantitative findings include:
- Energy Efficiency: The AGC-MEC algorithm reduced UE energy consumption by 65.96% compared to random placement, 49.35% compared to a "greedy" strategy, and a staggering 87.89% compared to purely local execution.
- Mobility as a Feature: Simulation data confirmed that UAVs successfully adjusted their trajectories to "chase" user clusters, proving that aerial mobility is a vital asset for maintaining high-quality service in dynamic environments.
Challenges and Future Directions
Despite its promise, AGC-MEC faces several hurdles. The computational complexity of MINLP solvers remains a challenge for real-time 6G applications, suggesting a shift toward AI-driven or heuristic solvers. Furthermore, integrating opportunistic mobile users as servers necessitates robust security frameworks; the authors suggest Blockchain as a potential solution for secure resource trading.
Standardization is also critical. Utilizing Network Function Virtualization (NFV) and Software Defined Networking (SDN) will be essential to create a generic, manageable computing platform that abstracts the underlying heterogeneity of the air-ground network for the end-user.
Summary updated