Dynamically Fine-Tuned Neural Compressor for FDD Massive MIMO CSI Feedback

Paper i proceeding, 2025

Efficient channel state information (CSI) compression is essential in frequency division duplexing (FDD) massive multiple-input multiple-output (MIMO) systems to mitigate excessive feedback overhead. While deep learning-based methods achieve superior performance, they struggle with distribution shifts, leading to degradation in their performance. To address this, we propose a model fine-tuning approach for CSI feedback, dynamically updating encoder/decoder networks using recent CSI samples. We adopt a full-model fine-tuning scheme, jointly updating the encoder and decoder model while accounting for the additional feedback overhead imposed by conveying the updated decoder parameters. Our results demonstrate that full-model fine-tuning significantly enhances the rate-distortion (RD) performance of neural CSI compression. Furthermore, we analyze how often the full-model fine-tuning should be applied in a new wireless environment and identify an optimal period interval for achieving the best RD trade-off.