Physically Parameterized Differentiable MUSIC for DoA Estimation with Uncalibrated Arrays

2024-11-06

arXiv.org

AI Summary

This paper addresses the degradation of direction-of-arrival (DoA) estimation performance in uncalibrated arrays caused by hardware impairments—specifically, antenna position errors and channel complex gain mismatches. To tackle this, we propose a physically parameterized, fully differentiable MUSIC algorithm that embeds the array’s physical model into a differentiable spectral estimation framework. The method supports both supervised and unsupervised learning, enabling end-to-end joint estimation of DoAs and hardware errors. Its key innovation lies in the first-ever full-chain differentiability of MUSIC, integrating principles from physics-informed neural networks (PINNs) to perform gradient propagation and subspace optimization directly in the complex domain. Simulation results demonstrate that the proposed approach accurately recovers hardware imperfections and achieves significantly higher DoA estimation accuracy than classical MUSIC—particularly under low signal-to-noise ratios and severe hardware mismatches—while exhibiting superior robustness.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: State Estimation
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• 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

Direction of arrival (DoA) estimation is a common sensing problem in radar, sonar, audio, and wireless communication systems. It has gained renewed importance with the advent of the integrated sensing and communication paradigm. To fully exploit the potential of such sensing systems, it is crucial to take into account potential hardware impairments that can negatively impact the obtained performance. This study introduces a joint DoA estimation and hardware impairment learning scheme following a model-based approach. Specifically, a differentiable version of the multiple signal classification (MUSIC) algorithm is derived, allowing efficient learning of the considered impairments. The proposed approach supports both supervised and unsupervised learning strategies, showcasing its practical potential. Simulation results indicate that the proposed method successfully learns significant inaccuracies in both antenna locations and complex gains. Additionally, the proposed method outperforms the classical MUSIC algorithm in the DoA estimation task.

Problem

• Estimates direction of arrival with uncalibrated arrays
• Learns hardware impairments using differentiable MUSIC algorithm
• Improves DoA estimation accuracy over classical MUSIC

Innovation

• Differentiable MUSIC algorithm for DoA estimation
• Joint DoA estimation and hardware impairment learning
• Supports supervised and unsupervised learning strategies