Deep Learning for Massive MIMO CSI Feedback
arXiv:1712.08919
Abstract
In frequency division duplex mode, the downlink channel state information (CSI) should be sent to the base station through feedback links so that the potential gains of a massive multiple-input multiple-output can be exhibited. However, such a transmission is hindered by excessive feedback overhead. In this letter, we use deep learning technology to develop CsiNet, a novel CSI sensing and recovery {mechanism} that learns to effectively use channel structure from training samples. CsiNet learns a transformation from CSI to a near-optimal number of representations (or codewords) and an inverse transformation from codewords to CSI. We perform experiments to demonstrate that CsiNet can recover CSI with significantly improved reconstruction quality compared with existing compressive sensing (CS)-based methods. Even at excessively low compression regions where CS-based methods cannot work, CsiNet retains effective beamforming gain.
arXiv comment5 pages, 2 figures, 3 tables; IEEE Wireless Communications Letters; the source code of this paper is available on GitHub: https://github.com/sydney222/Python_CsiNet
References in corpus (5)
- Message Passing Algorithms for Compressed Sensing
- Distributed Compressive CSIT Estimation and Feedback for FDD Multi-user Massive MIMO Systems
- Downlink Training Techniques for FDD Massive MIMO Systems: Open-Loop and Closed-Loop Training with Memory
- DR2-Net: Deep Residual Reconstruction Network for Image Compressive Sensing
- Deep Learning Based MIMO Communications