【Author】
Rahmadika, Sandi; Firdaus, Muhammad; Jang, Seolah; Rhee, Kyung-Hyune
【Source】SECURITY AND COMMUNICATION NETWORKS
【Abstract】Edge networks (ENs) in 5G have the capability to protect traffic between edge entry points (edge-to-edge), enabling the design of various flexible and customizable applications. The advantage of edge networks is their pioneering integration of other prominent technologies such as blockchain and federated learning (FL) to produce better services on wireless networks. In this paper, we propose an intelligent system integrating blockchain technologies, 5G ENs, and FL to create an efficient and secure framework for transactions. FL enables user equipment (UE) to train the artificial intelligence model without exposing the UE's valuable data to the public, or to the model providers. Furthermore, the blockchain is an immutable data approach that can be leveraged for FL across 5G ENs and beyond. The recorded transactions cannot be altered maliciously, and they remain unchanged by design. We further propose a dynamic authentication protocol for UE to interact with a diverse base station. We apply blockchain as a reward mechanism in FL to enable computational offloading in wireless networks. Additionally, we implement and investigate blockchain technology for FL in 5G UE.
【标题】支持区块链的 5G 边缘网络及其他：一种智能的Cross-Silo联邦学习方法
【摘要】5G 中的边缘网络 (EN) 能够保护边缘入口点（边缘到边缘）之间的流量，从而能够设计各种灵活且可定制的应用程序。边缘网络的优势在于它们开创性地整合了区块链和联邦学习 (FL) 等其他突出技术，以在无线网络上提供更好的服务。在本文中，我们提出了一个集成区块链技术、5G EN 和 FL 的智能系统，以创建一个高效且安全的交易框架。 FL 使用户设备 (UE) 能够训练人工智能模型，而无需将 UE 的有价值数据暴露给公众或模型提供者。此外，区块链是一种不可变的数据方法，可用于 5G EN 及以上的 FL。记录的交易不能被恶意更改，它们在设计上保持不变。我们进一步提出了一种用于 UE 与不同基站交互的动态认证协议。我们在 FL 中应用区块链作为奖励机制，以实现无线网络中的计算卸载。此外，我们在 5G UE 中实施和研究用于 FL 的区块链技术。
评论