Challenger-Based Combinatorial Bandits for Subcarrier Selection in OFDM Systems

2025-10-06

AI Summary

This paper addresses the combinatorial pure-exploration problem of selecting the top-𝑚 users in a multi-user MIMO-OFDM downlink—formulated as a stochastic linear bandit with combinatorial actions. Method: We propose an efficient online scheduling algorithm based on a challenger mechanism, introducing a novel gap-index framework that dynamically maintains a compact “champion–challenger” shortlist to prioritize high-information-gain measurements and avoid exhaustive search. The approach integrates linear utility modeling, combinatorial pure-exploration principles, and adaptive shortlist updating. Contribution/Results: It achieves a tunable trade-off between computational overhead and identification accuracy in large-scale action spaces. Experiments on realistic OFDM systems demonstrate substantial reductions in measurement cost and runtime—achieving several-fold speedup over state-of-the-art linear bandit methods—enabling AI-driven real-time radio resource scheduling.

Technology Category

• Machine Learning: Online Learning & Bandits
• Search and Optimization: Combinatorial Optimization
• Planning, Routing, and Scheduling: Scheduling under Uncertainty

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper investigates the identification of the top-m user-scheduling sets in multi-user MIMO downlink, which is cast as a combinatorial pure-exploration problem in stochastic linear bandits. Because the action space grows exponentially, exhaustive search is infeasible. We therefore adopt a linear utility model to enable efficient exploration and reliable selection of promising user subsets. We introduce a gap-index framework that maintains a shortlist of current estimates of champion arms (top-m sets) and a rotating shortlist of challenger arms that pose the greatest threat to the champions. This design focuses on measurements that yield the most informative gap-index-based comparisons, resulting in significant reductions in runtime and computation compared to state-of-the-art linear bandit methods, with high identification accuracy. The method also exposes a tunable trade-off between speed and accuracy. Simulations on a realistic OFDM downlink show that shortlist-driven pure exploration makes online, measurement-efficient subcarrier selection practical for AI-enabled communication systems.

Problem

• Identifies top user-scheduling sets in MIMO downlink systems
• Solves combinatorial bandit problem with exponential action space
• Enables efficient online subcarrier selection for AI communications

Innovation

• Gap-index framework with champion-challenger shortlists
• Linear utility model for efficient user subset exploration
• Tunable speed-accuracy trade-off in bandit optimization