Collaborative Research:CISE-MSI:DP:CNS:Enabling On-Demand and Flexible Mobile Edge Computing with Integrated Aerial-Ground Vehicles

移动和物联网设备的迅速普及彻底改变了我们生活的方方面面。然而，这些设备产生的大量数据对无线电频谱有限的无线通信基础设施构成了重大挑战。此外，许多新兴的应用程序需要低延迟和计算密集型处理，这使得传统的以云为中心的方法不够用。为了应对这些挑战，该项目提出了一种名为“空中-地面智能车辆边缘”（AGILE）的创新解决方案，该解决方案利用空中和地面车辆的能力以及人工智能处理能力，创建一个按需、灵活且具有成本效益的移动边缘计算（MEC）系统。AGILE旨在提供无处不在的低延迟计算服务，以支持大量连接设备并实现高效的数据处理。该项目的重点是设计AGILE架构，该架构将空中和地面车辆集成到一个3D网络中，用于智能MEC服务供应。首先，研究了无人机（uav）和地面车辆之间的协同训练方案，以实现智能MEC服务的快速节能联合学习。其次，该项目解决了无人机定位、通信和计算资源分配的耦合问题，优化它们以满足按需提供MEC服务。最后，开发了无人机动态移动和资源重构方案，以自适应满足用户需求，并在不同地面车辆资源存在的情况下实现灵活的MEC服务提供。该项目将加强三所参与的少数族裔服务机构之间现有的研究合作，同时促进非洲裔美国人/黑人、西班牙裔和女性本科生和/或研究生的研究参与，为MEC和人工智能领域的知识和技能做出贡献。那些缺乏服务的学生将通过研究项目、课堂教学和高级设计项目从这个项目中受益。这种参与将有助于所有学院提高代表性不足的学生的保留率。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The rapid proliferation of mobile and Internet-of-Things devices has revolutionized various aspects of our lives. However, the enormous amount of data generated by these devices poses significant challenges for wireless-communication infrastructure, which has limited radio spectrum. Additionally, many emerging applications require low-latency and computation-intensive processing, making the traditional cloud-centric approach inadequate. To address these challenges, this project proposes an innovative solution called Aerial-Ground Intelligent vehicular Edge (AGILE) which leverages the capabilities of aerial and ground vehicles with artificial-intelligence-processing capabilities to create an on-demand, flexible, and cost-effective mobile-edge-computing (MEC) system. AGILE aims to provide ubiquitous and low-latency computing services to support massive connected devices and enable efficient data processing.The project focuses on designing the AGILE architecture, which integrates aerial and ground vehicles into a 3D network for intelligent MEC service provisioning. Firstly, the research investigates collaborative training schemes between unmanned aerial vehicles (UAVs) and ground vehicles to enable fast and energy-efficient federated learning for intelligent MEC services. Secondly, the project addresses the coupling issue of UAV positioning, communication, and computing-resource allocation, optimizing them for on-demand MEC service provisioning. Finally, dynamic UAV movement and resource-reconfiguration schemes are developed to adaptively meet user demand and to achieve flexible MEC service provisioning in the presence of varying ground-vehicle resources. This project will strengthen the existing research collaborations among the three participating minority-serving institutions, while fostering research involvement of African American/Black, Hispanic, and women undergraduate and/or graduate students with the knowledge and skills to contribute to the fields of MEC and artificial intelligence. Those underserved students will benefit from this project through research projects, classroom teaching, and senior-design projects. Such participation will help all institutes in improving underrepresented students' retention rates.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.