Compression of Site-Specific Deep Neural Networks for Massive MIMO Precoding
Ghazal Kasalaee, Ali Hasanzadeh Karkan, Jean-François Frigon, and François Leduc-Primeau · 2025
Read Compression of Site-Specific Deep Neural Networks for Massive MIMO Precoding by Ghazal Kasalaee, Ali Hasanzadeh Karkan, Jean-François Frigon, and François Leduc-Primeau (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-08758.txt.
Description
Ghazal Kasalaee, Ali Hasanzadeh Karkan, Jean-François Frigon, and François Leduc-Primeau wrote a 2025 paper on making neural networks for wireless precoding cheaper to run. Massive MIMO systems use many antennas, and learned precoders can beat classic methods. Their computing and energy costs, though, block wider use.
The authors built an energy model that counts both memory access and calculation in learning hardware. They then compressed the networks with mixed precision training, where numbers are stored at varied bit widths, plus an automated search over network shapes. On ray traced data from varied base station sites, compressed precoders reached up to thirty five times the energy efficiency of a standard method at equal quality. Site conditions changed the gains, so the paper also sets a benchmark.
Catalog record
| CREATOR | Ghazal Kasalaee, Ali Hasanzadeh Karkan, Jean-François Frigon, and François Leduc-Primeau |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-08758