Attention-Based SINR Estimation in User-Centric Non-Terrestrial Networks

2026-02-24

AI Summary

This work addresses the challenge of real-time signal-to-interference-plus-noise ratio (SINR) estimation in non-terrestrial networks (NTNs), where conventional approaches relying on pilot signals or high-complexity minimum mean square error (MMSE) beamforming fail to meet the stringent latency requirements of user-centric scheduling. To overcome this, the authors propose a low-complexity SINR estimation framework based on dual multi-head self-attention (DMHSA), which— for the first time—applies multi-head self-attention mechanisms to NTN interference modeling by separately processing channel state information (CSI) and user location data. This approach accurately predicts group-level SINR without explicit MMSE computation, supports ultra-low-dimensional location inputs, and integrates priority queue-based scheduling during both training and inference. Experimental results demonstrate up to a 3× reduction in computational complexity with CSI inputs and two orders of magnitude reduction with location-only inputs, while achieving root-mean-square errors consistently below 1 dB, thereby significantly enhancing the efficiency of high-throughput user group selection.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: State Estimation
• Machine Learning: Hardware-aware ML

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Security and Privacy: Large-scale security measurements

Abstract

The signal-to-interference-plus-noise ratio (SINR) is central to performance optimization in user-centric beamforming for satellite-based non-terrestrial networks (NTNs). Its assessment either requires the transmission of dedicated pilots or relies on computing the beamforming matrix through minimum mean squared error (MMSE)-based formulations beforehand, a process that introduces significant computational overhead. In this paper, we propose a low-complexity SINR estimation framework that leverages multi-head self-attention (MHSA) to extract inter-user interference features directly from either channel state information or user location reports. The proposed dual MHSA (DMHSA) models evaluate the SINR of a scheduled user group without requiring explicit MMSE calculations. The architecture achieves a computational complexity reduction by a factor of three in the CSI-based setting and by two orders of magnitude in the location-based configuration, the latter benefiting from the lower dimensionality of user reports. We show that both DMHSA models maintain high estimation accuracy, with the root mean squared error typically below 1 dB with priority-queuing-based scheduled users. These results enable the integration of DMHSA-based estimators into scheduling procedures, allowing the evaluation of multiple candidate user groups and the selection of those offering the highest average SINR and capacity.

Problem

• SINR estimation
• non-terrestrial networks
• computational overhead
• user-centric beamforming
• interference assessment

Innovation

• multi-head self-attention
• SINR estimation
• non-terrestrial networks
• low-complexity
• user-centric beamforming