GD4: Graph-based Discrete Denoising Diffusion for MIMO Detection

2026-05-01

AI Summary

This work addresses the challenge of efficiently obtaining high-quality suboptimal solutions for detection in underdetermined MIMO systems by proposing GD4, a graph-driven discrete denoising diffusion method. GD4 is the first to introduce discrete denoising diffusion into MIMO detection, directly modeling and performing denoising in the symbol space while leveraging graph neural networks to capture signal structure. The method supports single-step or few-step inference and achieves significantly better performance than existing diffusion-based detectors and classical baselines in both underdetermined and overdetermined scenarios, all under comparable computational overhead. By doing so, GD4 strikes a superior balance between detection accuracy and computational efficiency.

Abstract

In wireless communications, recovering the optimal solution to the multiple-input multiple-output (MIMO) detection problem is NP-hard. Obtaining high-quality suboptimal solutions with a favorable performance-complexity trade-off is particularly challenging in under-determined systems with $N_t$ transmit antennas and $N_r < N_t$ receive antennas. Recent diffusion-based MIMO detectors have shown promise, but they require extensive sampling iterations at inference time, and their performance degrades in under-determined scenarios. We propose GD4, a graph-based discrete denoising diffusion method for MIMO detection. Unlike existing diffusion-based detectors that operate in a continuous relaxed space, GD4 performs denoising directly in the discrete symbol space and enables fast inference with one or a few denoising evaluations. Numerical results show that, under a similar inference-time compute budget, GD4 produces higher-quality suboptimal solutions than existing diffusion-based detectors and some widely used classical baseline including box-constrained Babai point and the $K$-best box-constrained randomized Klein-Babai point in both under-determined and overdetermined settings.

Problem

• MIMO detection
• NP-hard
• under-determined systems
• performance-complexity trade-off
• suboptimal solutions

Innovation

• discrete denoising diffusion
• graph-based MIMO detection
• under-determined systems
• fast inference
• symbol space