# A Low-Complexity Plug-and-Play Deep Learning Model for Generalizable Massive MIMO Precoding

arXiv: https://arxiv.org/abs/2601.21897 · HTML: https://mlsanity.com/paper/2601.21897/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 29 Jan 2026.

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Unfolding WMMSE using Graph Neural Networks for Efficient Power Allocation](https://mlsanity.com/paper/2009.10812/) (2020)
- [Minimum Energy Quantized Neural Networks](https://mlsanity.com/paper/1711.00215/) (2017)
- [Domain Generalization in Machine Learning Models for Wireless Communications: Concepts, State-of-the-Art, and Open Issues](https://mlsanity.com/paper/2303.08106/) (2023)
- [Learn to Rapidly and Robustly Optimize Hybrid Precoding](https://mlsanity.com/paper/2301.00369/) (2023)
- [Unsupervised Deep Learning for Massive MIMO Hybrid Beamforming](https://mlsanity.com/paper/2007.00038/) (2020)
- [Lightweight Neural Network with Knowledge Distillation for CSI Feedback](https://mlsanity.com/paper/2210.17113/) (2022)
- [SAGE-HB: Swift Adaptation and Generalization in Massive MIMO Hybrid Beamforming](https://mlsanity.com/paper/2401.10513/) (2024)
- [Capacity Achieving by Diagonal Permutation for MU-MIMO channels](https://mlsanity.com/paper/2211.09208/) (2022)
- [Learning Energy-Efficient Hardware Configurations for Massive MIMO Beamforming](https://mlsanity.com/paper/2308.06376/) (2023)
- [Toward Energy-Efficient Massive MIMO: Graph Neural Network Precoding for Mitigating Non-Linear PA Distortion](https://mlsanity.com/paper/2312.04591/) (2023)
- [Cache-Aided Massive MIMO: Linear Precoding Design and Performance Analysis](https://mlsanity.com/paper/1903.09336/) (2019)
- [Downlink Massive MU-MIMO with Successively-Regularized Zero Forcing Precoding](https://mlsanity.com/paper/2206.08771/) (2022)
- [Deep Learning-Aided Delay-Tolerant Zero-Forcing Precoding in Cell-Free Massive MIMO](https://mlsanity.com/paper/2210.05229/) (2022)
- [Joint Linear Precoding and DFT Beamforming Design for Massive MIMO Satellite Communication](https://mlsanity.com/paper/2211.08757/) (2022)
- [Deep-Learning Aided Channel Training and Precoding in FDD Massive MIMO with Channel Statistics Knowledge](https://mlsanity.com/paper/2303.11096/) (2023)
- [Generalizing Deep Learning-Based CSI Feedback in Massive MIMO via ID-Photo-Inspired Preprocessing](https://mlsanity.com/paper/2409.13494/) (2024)
- [A Low-Complexity Plug-and-Play Deep Learning Model for Massive MIMO Precoding Across Sites](https://mlsanity.com/paper/2502.08757/) (2025)
- [Compression of Site-Specific Deep Neural Networks for Massive MIMO Precoding](https://mlsanity.com/paper/2502.08758/) (2025)
- [Adversarial Training: Enhancing Out-of-Distribution Generalization for Learning Wireless Resource Allocation](https://mlsanity.com/paper/2506.21208/) (2025)

## Abstract (arXiv)

Massive multiple-input multiple-output (mMIMO) downlink precoding offers high spectral efficiency but remains challenging to deploy in practice because near-optimal algorithms such as the weighted minimum mean squared error (WMMSE) are computationally expensive, and sensitive to SNR and channel-estimation quality, while existing deep learning (DL)-based solutions often lack robustness and require retraining for each deployment site. This paper proposes a plug-and-play precoder (PaPP), a DL framework with a backbone that can be trained for either fully digital (FDP) or hybrid beamforming (HBF) precoding and reused across sites, transmit-power levels, and with varying amounts of channel estimation error, avoiding the need to train a new model from scratch at each deployment. PaPP combines a high-capacity teacher and a compact student with a self-supervised loss that balances teacher imitation and normalized sum-rate, trained using meta-learning domain-generalization and transmit-power-aware input normalization. Numerical results on ray-tracing data from three unseen sites show that the PaPP FDP and HBF models both outperform conventional and deep learning baselines, after fine-tuning with a small set of local unlabeled samples. Across both architectures, PaPP achieves more than 21× reduction in modeled computation energy and maintains good performance under channel-estimation errors, making it a practical solution for energy-efficient mMIMO precoding.