Federated Learning for Wireless Networks

Federated learning (FL) is a distributed machine learning algorithm that allows devices to train a shared machine learning model while keeping data locally. However, applying FL in wireless networks and optimizing its performance involves research topics such as wireless impairments, incentive mechanisms, and security and privacy. This book provides a comprehensive overview of FL for wireless networks, covering resource optimization, incentive mechanism, security and privacy, and applications.

Format: Paperback / softback
Length: 253 pages
Publication date: 03 December 2022
Publisher: Springer Verlag, Singapore

Recently, machine learning schemes have garnered significant attention as vital enablers for the next-generation wireless systems. Currently, wireless systems predominantly employ machine learning algorithms that centralize training and inference processes by transferring end-device data to a third-party centralized location. However, these schemes result in end-device privacy leakage. To address these concerns, one can utilize distributed machine learning at the network edge. In this context, federated learning (FL) emerges as one of the most significant distributed learning algorithms, enabling devices to train a shared machine learning model while maintaining data locally. Nevertheless, applying FL in wireless networks and optimizing its performance involves a diverse range of research topics. For instance, in FL, training machine learning models necessitate communication between wireless devices and edge servers via wireless links. Consequently, wireless impairments such as uncertainties among wireless channel states, interference, and noise significantly impact the performance of FL. On the other hand, federated-reinforcement learning harnesses distributed computation power and data to solve complex optimization problems that arise in various use cases, such as interference alignment, resource management, clustering, and network control. Traditionally, FL assumes that edge devices will unconditionally participate in the tasks when invited, which is not practical in reality due to the cost of model training. As such, building incentive mechanisms is indispensable for FL networks.

This book offers a comprehensive overview of FL for wireless networks. It is structured into three main parts: The first part provides a brief introduction to FL for wireless networks, while the second part offers a comprehensive review. The third part explores the application of FL in various wireless networks, including cellular networks, Internet of Things (IoT) networks, and vehicular networks.

Weight: 415g
Dimension: 235 x 155 (mm)
ISBN-13: 9789811649653
Edition number: 1st ed. 2021