MIMO channel estimation improves with adaptive gated diffusion model
SNR-Gated LSTM-Conditioned Diffusion Model for MIMO Channel Estimation
Machine Learning
Summary
Communicating wirelessly using multiple antennas requires knowing how signals travel in changing environments, which is hard when things move quickly. The authors propose a new method that uses a clever combination of a memory-based neural network and a noise-reducing diffusion process to better estimate these signal paths over time. Their approach adapts to different noise levels and speeds up computations when the signal is clearer. Tests on simulated data show this method works better and faster than previous similar methods.
What this means in practice
- •For wireless network engineers: Improve real-time channel estimation in MIMO systems under high mobility to enhance data transmission quality and reliability.
- •For communication system developers: Incorporate adaptive noise gating and diffusion models to reduce processing latency in signal path estimation for mobile devices and infrastructure.
Authors
Jixing Zhou, Xinming Huang
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.