ReQuestNet: A Foundational Learning model for Channel Estimation

2025-08-12

AI Summary

Channel estimation in 5G+ systems faces significant challenges due to joint modeling of dynamically configurable resources—such as variable resource block (RB) counts, MIMO layer numbers, PRG sizes, and DMRS patterns—together with unknown precoded channels. Method: This paper proposes CoarseNet-RefinementNet, a unified deep learning architecture based on cyclic equivariance, which jointly models spatio-temporal-frequency correlations across PRGs and MIMO layers for end-to-end channel estimation—without requiring auxiliary reference signals or prior channel statistics. Contribution/Results: The framework achieves up to 10 dB NMSE improvement over ideal linear MMSE at high SNR and demonstrates strong generalization to unseen channel conditions and real-time configuration changes. It is the first approach to simultaneously enable joint correlation modeling of unknown precoded channels and full compatibility with dynamic resource configurations, establishing a unified estimation framework for practical 5G+ deployments.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

In this paper, we present a novel neural architecture for channel estimation (CE) in 5G and beyond, the Recurrent Equivariant UERS Estimation Network (ReQuestNet). It incorporates several practical considerations in wireless communication systems, such as ability to handle variable number of resource block (RB), dynamic number of transmit layers, physical resource block groups (PRGs) bundling size (BS), demodulation reference signal (DMRS) patterns with a single unified model, thereby, drastically simplifying the CE pipeline. Besides it addresses several limitations of the legacy linear MMSE solutions, for example, by being independent of other reference signals and particularly by jointly processing MIMO layers and differently precoded channels with unknown precoding at the receiver. ReQuestNet comprises of two sub-units, CoarseNet followed by RefinementNet. CoarseNet performs per PRG, per transmit-receive (Tx-Rx) stream channel estimation, while RefinementNet refines the CoarseNet channel estimate by incorporating correlations across differently precoded PRGs, and correlation across multiple input multiple output (MIMO) channel spatial dimensions (cross-MIMO). Simulation results demonstrate that ReQuestNet significantly outperforms genie minimum mean squared error (MMSE) CE across a wide range of channel conditions, delay-Doppler profiles, achieving up to 10dB gain at high SNRs. Notably, ReQuestNet generalizes effectively to unseen channel profiles, efficiently exploiting inter-PRG and cross-MIMO correlations under dynamic PRG BS and varying transmit layer allocations.

Problem

• Develops ReQuestNet for 5G channel estimation
• Handles variable RBs, DMRS patterns, MIMO layers
• Outperforms MMSE with inter-PRG cross-MIMO correlations

Innovation

• Unified neural model for diverse 5G CE parameters
• CoarseNet and RefinementNet for layered CE
• Exploits inter-PRG and cross-MIMO correlations