Hierarchical Federated Learning With Distributed Clustering and Multichannel ALOHA

In an age characterized by seamless interconnectivity, the quantity of Internet of Things (IoT) devices has experienced substantial growth in recent years, with projections indicating further expansion. In this context, federated learning (FL) plays an important role in the future of wireless communications, offering numerous advantages over traditional centralized learning approaches, including data privacy preservation, reduced bandwidth usage, improved accuracy, and customization. However, selecting an appropriate wireless protocol and data transmission method for FL is crucial. In this work, we adopt the multichannel ALOHA protocol due to its asynchronous nature and simple implementation compared to other protocols. This article focuses on optimizing multichannel ALOHA communication within a hierarchical FL (HFL) system by creating a device-to-device (D2D) clustering scheme, which enables a single base station (BS) to serve more devices and drastically reduces the achievable error.