Radar-APLANC: Unsupervised Radar-based Heartbeat Sensing via Augmented Pseudo-Label and Noise Contrast

2025-11-11

AI Summary

Supervised radar-based heartbeat sensing relies on costly ground-truth physiological labels, while conventional unsupervised methods suffer from poor noise robustness. Method: This paper proposes the first unsupervised radar heartbeat sensing framework, eliminating the need for annotated physiological signals. Leveraging inherent spectral separation between heartbeat and noise components in FMCW radar signals, it constructs positive/negative sample pairs. A pseudo-label enhancement strategy—incorporating domain-specific radar priors—and a novel Noise-Contrastive Triplet (NCT) loss enable self-supervised training, enhancing discriminability between heartbeat and noise in feature space. Contribution/Results: The framework achieves performance comparable to state-of-the-art supervised methods on both the public Equipleth dataset and a newly collected in-house dataset, demonstrating the feasibility and practicality of unsupervised learning for non-contact physiological sensing.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Large-scale security measurements

Abstract

Frequency Modulated Continuous Wave (FMCW) radars can measure subtle chest wall oscillations to enable non-contact heartbeat sensing. However, traditional radar-based heartbeat sensing methods face performance degradation due to noise. Learning-based radar methods achieve better noise robustness but require costly labeled signals for supervised training. To overcome these limitations, we propose the first unsupervised framework for radar-based heartbeat sensing via Augmented Pseudo-Label and Noise Contrast (Radar-APLANC). We propose to use both the heartbeat range and noise range within the radar range matrix to construct the positive and negative samples, respectively, for improved noise robustness. Our Noise-Contrastive Triplet (NCT) loss only utilizes positive samples, negative samples, and pseudo-label signals generated by the traditional radar method, thereby avoiding dependence on expensive ground-truth physiological signals. We further design a pseudo-label augmentation approach featuring adaptive noise-aware label selection to improve pseudo-label signal quality. Extensive experiments on the Equipleth dataset and our collected radar dataset demonstrate that our unsupervised method achieves performance comparable to state-of-the-art supervised methods. Our code, dataset, and supplementary materials can be accessed from https://github.com/RadarHRSensing/Radar-APLANC.

Problem

• Developing unsupervised radar heartbeat sensing without labeled data
• Overcoming noise interference in non-contact physiological monitoring
• Eliminating dependency on expensive ground-truth physiological signals

Innovation

• Unsupervised radar heartbeat sensing via pseudo-label contrast
• Noise-Contrastive Triplet loss using positive and negative samples
• Adaptive noise-aware label selection for pseudo-label augmentation