arXiv：cs.LG· Jixing Zhou, Xinming Huang·· 4 小时前AI 评分27
SNR-Gated LSTM-Conditioned 扩散模型用于 MIMO 信道估计
SNR-Gated LSTM-Conditioned Diffusion Model for MIMO Channel Estimation
AI 导读
该论文提出一种时序条件扩散框架，在角域去噪完成 MIMO 信道估计，用 LSTM 网络对短观测序列编码作为条件信息。模型引入可学习的 SNR 门控后融合短路，通过带可训练中心和尺度的 sigmoid 门将网络输入注入最终解码阶段，并采用 DDIM 式确定性反向更新与 SNR 自适应截断和步数分配降低推理延迟。
正文
Abstract:Accurate and low latency channel estimation is critical for modern MIMO systems, particularly under mobility, where channels exhibit structured sparsity and strong temporal correlation. This paper proposes a time-series conditioned diffusion framework for channel estimation that performs denoising in the angular domain. Starting from least squares (LS) observations, we train a diffusion denoiser whose conditioning information is encoded by a long short-term memory (LSTM) network over a short observation sequence, enabling the model to exploit temporal dynamics beyond per-snapshot estimation. To robustly balance observation fidelity and learned generative priors across a wide signal-to-noise ratio (SNR) range, we introduce a learnable SNR-gated late-fusion shortcut that injects the network input into the final decoding stage through a sigmoid gate with trainable center and scale. To reduce inference latency, we adopt deterministic denoising diffusion implicit model (DDIM) style reverse updates with SNR-adaptive truncation and step allocation, which significantly reduces the number of reverse diffusion steps at high SNR while maintaining strong performance in low SNR regimes. Simulations on time-evolving standardized channel models demonstrate that the proposed method achieves consistent performance gains over existing diffusion-based channel estimation baselines, while retaining low latency through SNR-adaptive inference.
| Comments: | 5 pages, 5 figures. Accepted by and presented at the 2026 IEEE 104th Vehicular Technology Conference (VTC2026-Fall) |
| Subjects: | Machine Learning (cs.LG); Signal Processing (eess.SP) |
| Cite as: | arXiv:2610.08977 [cs.LG] |
| (or arXiv:2610.08977v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08977 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xinming Huang [view email]
[v1]
Tue, 6 Oct 2026 18:39:12 UTC (5,344 KB)
来源：arXiv：cs.LG · arxiv.org