Learning to optimize: Training deep neural networks for wireless resource management

Haoran Sun, Xiangyi Chen, Qingjiang Shi, Mingyi Hong, Xiao Fu, Nikos D. Sidiropoulos

Research output: Chapter in Book/Report/Conference proceedingConference contribution

87 Scopus citations

Abstract

For decades, optimization has played a central role in addressing wireless resource management problems such as power control and beamformer design. However, these algorithms often require a considerable number of iterations for convergence, which poses challenges for real-time processing. In this work, we propose a new learning-based approach for wireless resource management. The key idea is to treat the input and output of a resource allocation algorithm as an unknown non-linear mapping and to use a deep neural network (DNN) to approximate it. If the nonlinear mapping can be learned accurately and effectively by a DNN of moderate size, then such DNN can be used for resource allocation in almost real time, since passing the input through a DNN to get the output only requires a small number of simple operations. In this work, we first characterize a class of 'learnable algorithms' and then design DNNs to approximate some algorithms of interest in wireless communications. We use extensive numerical simulations to demonstrate the superior ability of DNNs for approximating two considerably complex algorithms that are designed for power allocation in wireless transmit signal design, while giving orders of magnitude speedup in computational time.

Original languageEnglish (US)
Title of host publication18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-6
Number of pages6
ISBN (Electronic)9781509030088
DOIs
StatePublished - Dec 19 2017
Event18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017 - Sapporo, Japan
Duration: Jul 3 2017Jul 6 2017

Publication series

NameIEEE Workshop on Signal Processing Advances in Wireless Communications, SPAWC
Volume2017-July

Other

Other18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017
Country/TerritoryJapan
CitySapporo
Period7/3/177/6/17

Bibliographical note

Funding Information:
The authors gratefully acknowledge the support of NVIDIA Corporation with the donation of the Titan X Pascal GPU used for this research. The authors also wish to thank helpful discussions with Tom Luo from Chinese University of Hong Kong (Shenzhen).

Funding Information:
M. Hong is supported in part by NSF under grant CCF-1526078, AFOSR under grant 15RT0767. Q. Shi is supported by NSFC under grant 61671411. N. D. Sidiropoulos is supported by NSF under grant IIS-1447788.

Funding Information:
M. Hong is supported in part by NSF under grant CCF-1526078, AFOSR under grant 15RT0767. Q. Shi is supported by NSFC under grant 61671411. N. D. Sidiropoulos is supported by NSF under grant IIS-1447788

Publisher Copyright:
© 2017 IEEE.

Fingerprint

Dive into the research topics of 'Learning to optimize: Training deep neural networks for wireless resource management'. Together they form a unique fingerprint.

Cite this