Saved in:
Bibliographic Details
Main Authors: Ge, Yilun, Liao, Shuyao, Han, Shengqian, Yang, Chenyang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.09398
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912271170535424
author Ge, Yilun
Liao, Shuyao
Han, Shengqian
Yang, Chenyang
author_facet Ge, Yilun
Liao, Shuyao
Han, Shengqian
Yang, Chenyang
contents Incorporating mathematical properties of a wireless policy to be learned into the design of deep neural networks (DNNs) is effective for enhancing learning efficiency. Multi-user precoding policy in multi-antenna system, which is the mapping from channel matrix to precoding matrix, possesses a permutation equivariance property, which has been harnessed to design the parameter sharing structure of the weight matrix of DNNs. In this paper, we study a stronger property than permutation equivariance, namely unitary equivariance, for precoder learning. We first show that a DNN with unitary equivariance designed by further introducing parameter sharing into a permutation equivariant DNN is unable to learn the optimal precoder. We proceed to develop a novel non-linear weighting process satisfying unitary equivariance and then construct a joint unitary and permutation equivariant DNN. Simulation results demonstrate that the proposed DNN not only outperforms existing learning methods in learning performance and generalizability but also reduces training complexity.
format Preprint
id arxiv_https___arxiv_org_abs_2503_09398
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Precoder Learning by Leveraging Unitary Equivariance Property
Ge, Yilun
Liao, Shuyao
Han, Shengqian
Yang, Chenyang
Signal Processing
Machine Learning
Systems and Control
Group Theory
Incorporating mathematical properties of a wireless policy to be learned into the design of deep neural networks (DNNs) is effective for enhancing learning efficiency. Multi-user precoding policy in multi-antenna system, which is the mapping from channel matrix to precoding matrix, possesses a permutation equivariance property, which has been harnessed to design the parameter sharing structure of the weight matrix of DNNs. In this paper, we study a stronger property than permutation equivariance, namely unitary equivariance, for precoder learning. We first show that a DNN with unitary equivariance designed by further introducing parameter sharing into a permutation equivariant DNN is unable to learn the optimal precoder. We proceed to develop a novel non-linear weighting process satisfying unitary equivariance and then construct a joint unitary and permutation equivariant DNN. Simulation results demonstrate that the proposed DNN not only outperforms existing learning methods in learning performance and generalizability but also reduces training complexity.
title Precoder Learning by Leveraging Unitary Equivariance Property
topic Signal Processing
Machine Learning
Systems and Control
Group Theory
url https://arxiv.org/abs/2503.09398