Foundations and Trends® in Signal Processing > Vol 15 > Issue 4

Wireless for Machine Learning: A Survey

By Henrik Hellström, KTH Royal Institute of Technology, Sweden, hhells@kth.se | José Mairton B. da Silva Jr., KTH Royal Institute of Technology, Sweden | Mohammad Mohammadi Amiri, Massachusetts Institute of Technology, USA | Mingzhe Chen, Princeton University, USA | Viktoria Fodor, KTH Royal Institute of Technology, Sweden | H. Vincent Poor, Princeton University, USA | Carlo Fischione, KTH Royal Institute of Technology, Sweden

 
Suggested Citation
Henrik Hellström, José Mairton B. da Silva Jr., Mohammad Mohammadi Amiri, Mingzhe Chen, Viktoria Fodor, H. Vincent Poor and Carlo Fischione (2022), "Wireless for Machine Learning: A Survey", Foundations and Trends® in Signal Processing: Vol. 15: No. 4, pp 290-399. http://dx.doi.org/10.1561/2000000114

Publication Date: 09 Jun 2022
© 2022 H. Hellström et al.
 
Subjects
Statistical signal processing: estimation and regression,  Statistical/machine learning,  Signal processing for communications,  Sensor and multiple source signal processing,  Linear and nonlinear filtering,  Coding and compression,  Detection and estimation,  Communication system design,  Joint source/channel coding,  Modulation and signal design,  Pattern recognition and learning,  Wireless Communications
 

Free Preview:

Download extract

Share

Download article
In this article:
1. Introduction
2. Primer on Distributed Machine Learning
3. Analog Over-the-air Computation
4. Digital Communications
5. Open Problems
6. Applications
7. Conclusions
References

Abstract

As data generation increasingly takes place on devices without a wired connection, Machine Learning (ML) related traffic will be ubiquitous in wireless networks. Many studies have shown that traditional wireless protocols are highly inefficient or unsustainable to support ML, which creates the need for new wireless communication methods. In this monograph, we give a comprehensive review of the state-of-the-art wireless methods that are specifically designed to support ML services over distributed datasets. Currently, there are two clear themes within the literature, analog over-the-air computation and digital radio resource management optimized for ML. This survey gives an introduction to these methods, reviews the most important works, highlights open problems, and discusses application scenarios.

DOI:10.1561/2000000114
ISBN: 978-1-63828-006-4
124 pp. $85.00
Buy book (pb)
 
ISBN: 978-1-63828-007-1
124 pp. $145.00
Buy E-book (.pdf)
Table of contents:
1. Introduction
2. Primer on Distributed Machine Learning
3. Analog Over-the-air Computation
4. Digital Communications
5. Open Problems
6. Applications
7. Conclusions
References

Wireless for Machine Learning: A Survey

This monograph covers the topic of Wireless for Machine Learning (ML). Although the general intersection of ML and wireless communications is currently a prolific field of research that has already generated multiple publications, there is little review work on Wireless for ML.

As data generation increasingly takes place on devices without a wired connection, ML related traffic will be ubiquitous in wireless networks. Research has shown that traditional wireless protocols are highly inefficient or unsustainable to support ML, which creates the need for new wireless communication methods. This monograph gives an exhaustive review of the state-of-the-art wireless methods that are specifically designed to support ML services over distributed datasets. Currently, there are two clear themes within the literature, analog over-the-air computation and digital radio resource management optimized for ML. A comprehensive introduction to these methods is presented, reviews are made of the most important works, open problems are highlighted and application scenarios are discussed.

 
SIG-114