Interference-Aware PMI selection for MIMO systems in an O-RAN scenario

2025-04-20

AI Summary

In dense 5G deployments, PMI selection in MIMO systems faces a multi-objective optimization challenge: maximizing spectral efficiency (SE) while suppressing inter-cell interference. Addressing this within the O-RAN architecture, this paper proposes a UE-priority-aware, interference-coordinated PMI selection mechanism. We innovatively formulate an Advantage Actor-Critic (A2C) reinforcement learning agent as a lightweight xApp, deployed on the O-RAN near-real-time RIC to enable distributed, dynamic, and adaptive PMI optimization. The method integrates real-time channel state information and interference modeling to support low-overhead online decision-making. Experimental results demonstrate an 18.7% gain in spectral efficiency, a 3.2 dB improvement in SINR (indicating significant inter-cell interference reduction), and a 22% increase in resource utilization. To the best of our knowledge, this is the first end-to-end deployment of A2C as an O-RAN xApp, establishing a scalable paradigm for intelligent precoding in open wireless networks.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Applications
• Machine Learning: Hardware-aware ML

Application Category

• 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
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The optimization of Precoding Matrix Indicators (PMIs) is crucial for enhancing the performance of 5G networks, particularly in dense deployments where inter-cell interference is a significant challenge. Some approaches have leveraged AI/ML techniques for beamforming and beam selection, however, these methods often overlook the multi-objective nature of PMI selection, which requires balancing spectral efficiency (SE) and interference reduction. This paper proposes an interference-aware PMI selection method using an Advantage Actor-Critic (A2C) reinforcement learning model, designed for deployment within an O-RAN framework as an xApp. The proposed model prioritizes user equipment (UE) based on a novel strategy and adjusts PMI values accordingly, with interference management and efficient resource utilization. Experimental results in an O-RAN environment demonstrate the approach's effectiveness in improving network performance metrics, including SE and interference mitigation.

Problem

• Optimizing PMI selection for 5G MIMO systems
• Balancing spectral efficiency and interference reduction
• Deploying AI-based PMI selection in O-RAN framework

Innovation

• Interference-aware PMI selection using A2C reinforcement learning
• Deployed as xApp within O-RAN framework
• Balances spectral efficiency and interference reduction