A Transformer Inspired AI-based MIMO receiver

2025/10/23 by Rácz, András, Borsos, Tamás, Veres, András +1
#FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2510.20363

Abstract

We present AttDet, a Transformer-inspired MIMO (Multiple Input Multiple Output) detection method that treats each transmit layer as a token and learns inter-stream interference via a lightweight self-attention mechanism. Queries and keys are derived directly from the estimated channel matrix, so attention scores quantify channel correlation. Values are initialized by matched-filter outputs and iteratively refined. The AttDet design combines model-based interpretability with data-driven flexibility. We demonstrate through link-level simulations under realistic 5G channel models and high-order, mixed QAM modulation and coding schemes, that AttDet can approach near-optimal BER/BLER (Bit Error Rate/Block Error Rate) performance while maintaining predictable, polynomial complexity.

Citations

• A Neural Receiver for 5G NR Multi-user MIMO
• Convolutional Self-Attention-Based Multi-User MIMO Demapper
• DeepRx MIMO: Convolutional MIMO Detection with Learned Multiplicative\n Transformations
• Graph Neural Networks for Massive MIMO Detection
• Adaptive Neural Signal Detection for Massive MIMO
• Attention Is All You Need
• Deep MIMO Detection

Related

• SigT: An Efficient End-to-End MIMO-OFDM Receiver Framework Based on Transformer
• Reconfigurable AI Modules Aided Channel Estimation and MIMO Detection
• Efficient MIMO Detection with Imperfect Channel Knowledge - A Deep Learning Approach
• Learning to Search for MIMO Detection
• Adaptive Neural Signal Detection for Massive MIMO