Exploring FMCW Radars and Feature Maps for Activity Recognition: A Benchmark Study

2025-03-07

AI Summary

To address user discomfort, privacy concerns, and degraded performance under low-light or long-range conditions inherent in wearable sensors and video-based human activity recognition (HAR), this paper proposes a contactless HAR framework using 60 GHz frequency-modulated continuous-wave (FMCW) radar. We introduce a novel tensor-based representation that directly feeds raw 3D radar feature maps—comprising range-Doppler, azimuth, and elevation dimensions—into temporal models, preserving structural information lost in conventional image-like preprocessing. A radar-optimized ConvLSTM architecture is designed for end-to-end recognition. Furthermore, we establish the first millimeter-wave radar benchmark dataset featuring seven activities captured from multiple viewpoints. Experiments demonstrate state-of-the-art performance: 90.51% accuracy (F1 = 87.31%) in cross-scenario evaluation and 89.56% accuracy (F1 = 87.15%) in leave-one-subject-out validation—significantly outperforming CNN, LSTM, and SVM baselines. The framework exhibits strong privacy preservation, robustness, and generalization capability.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Video Understanding & Activity Analysis
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

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

Abstract

Human Activity Recognition has gained significant attention due to its diverse applications, including ambient assisted living and remote sensing. Wearable sensor-based solutions often suffer from user discomfort and reliability issues, while video-based methods raise privacy concerns and perform poorly in low-light conditions or long ranges. This study introduces a Frequency-Modulated Continuous Wave radar-based framework for human activity recognition, leveraging a 60 GHz radar and multi-dimensional feature maps. Unlike conventional approaches that process feature maps as images, this study feeds multi-dimensional feature maps -- Range-Doppler, Range-Azimuth, and Range-Elevation -- as data vectors directly into the machine learning (SVM, MLP) and deep learning (CNN, LSTM, ConvLSTM) models, preserving the spatial and temporal structures of the data. These features were extracted from a novel dataset with seven activity classes and validated using two different validation approaches. The ConvLSTM model outperformed conventional machine learning and deep learning models, achieving an accuracy of 90.51% and an F1-score of 87.31% on cross-scene validation and an accuracy of 89.56% and an F1-score of 87.15% on leave-one-person-out cross-validation. The results highlight the approach's potential for scalable, non-intrusive, and privacy-preserving activity monitoring in real-world scenarios.

Problem

• Develops FMCW radar-based framework for human activity recognition.
• Addresses limitations of wearable sensors and video-based methods.
• Achieves high accuracy using ConvLSTM model on multi-dimensional feature maps.

Innovation

• Uses 60 GHz FMCW radar for activity recognition
• Processes multi-dimensional feature maps as data vectors
• Achieves high accuracy with ConvLSTM model