Towards Smarter Sensing: 2D Clutter Mitigation in RL-Driven Cognitive MIMO Radar

2025-02-07

AI Summary

Robust multi-target detection in dynamic 6G environments under unknown two-dimensional (2D) clutter interference remains challenging. Method: This paper proposes a reinforcement learning (RL)-enabled cognitive MIMO radar system. It jointly optimizes waveform design and beamforming online by integrating the SARSA RL algorithm with a Wald-type statistical detector within a planar array architecture, while modeling 2D autoregressive clutter for real-time adaptive suppression. Contribution/Results: The approach establishes an end-to-end cognitive closed loop—unifying sensing, decision-making, and response—overcoming limitations of conventional static waveforms and open-loop processing. Experiments demonstrate significantly higher detection probability than omnidirectional transmission under low-SNR, clutter-dominated scenarios, validating superior robustness and real-time adaptability to dynamic interference.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Motivated by the growing interest in integrated sensing and communication for 6th generation (6G) networks, this paper presents a cognitive Multiple-Input Multiple-Output (MIMO) radar system enhanced by reinforcement learning (RL) for robust multitarget detection in dynamic environments. The system employs a planar array configuration and adapts its transmitted waveforms and beamforming patterns to optimize detection performance in the presence of unknown two-dimensional (2D) disturbances. A robust Wald-type detector is integrated with a SARSA-based RL algorithm, enabling the radar to learn and adapt to complex clutter environments modeled by a 2D autoregressive process. Simulation results demonstrate significant improvements in detection probability compared to omnidirectional methods, particularly for low Signal-to-Noise Ratio (SNR) targets masked by clutter.

Problem

• Enhancing MIMO radar with RL
• Mitigating 2D clutter in dynamic environments
• Improving detection in low SNR conditions

Innovation

• Reinforcement learning enhances MIMO radar
• SARSA-based RL optimizes waveform adaptation
• 2D autoregressive process models clutter