A Parameter Estimation and Deep Learning Hybrid Extraction Network for Multidirectional Human Activity Recognition Based on mmWave Radar
- 4 November 2024
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (5) , 5769-5782
Abstract
In the realm of human activity recognition (HAR) based on radar, the prevailing methods have been characterized by excessive complexity and a singular focus on a specific motion direction, posing challenges for practical deployment. This paper introduces a lightweight parameter estimation and deep learning hybrid extraction network (PDHE-Net) for achieving multi-directional HAR based on mmWave radar. The network consists of a lightweight deep learning feature extraction (LDE) module, a parameter estimation module, and a classification module. Specifically, the LDE module consists of multiple layers of group convolution and Ghost module, aimed at extracting deep features from the time-Doppler (TD) maps. Parameter estimation module is employed to capture direction-independent features from the TD map. Ultimately, multi-dimensional features extracted by the LDE module and parameter estimation module are classified to realize multi-directional HAR. To verify the performance of the proposed method, experimental data was collected, comprising six categories of activities wherein targets moved in multi-directions. The experimental results demonstrate that the proposed PDHE-Net achieved a recognition accuracy of 96.67% on the dataset, outperforming the state-of-the-art methods by 2.92%, while significantly reducing complexity.Keywords
Funding Information
- Municipal Government of Quzhou (2022D005, 2022D008, 2023D032)
- Higher Education Discipline Innovation Project (B17008)
- Chengdu “Revelation and Leadership” Science and Technology Project (2023-JB00-00032-GX)
This publication has 47 references indexed in Scilit:
- A Survey of mmWave-Based Human Sensing: Technology, Platforms and ApplicationsIEEE Communications Surveys & Tutorials, 2023
- Convolutional neural network for people counting using UWB impulse radarJournal of Instrumentation, 2021
- Human Motion Serialization Recognition With Through-the-Wall RadarIEEE Access, 2020
- ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks2020
- GhostNet: More Features From Cheap Operations2020
- Review of research on lightweight convolutional neural networks2020
- Activity Classification Using Raw Range and I & Q Radar Data with Long Short Term Memory Layers2018
- MobileNetV2: Inverted Residuals and Linear Bottlenecks2018
- Performance Analysis of Classification Algorithms for Activity Recognition Using Micro-Doppler Feature2017
- Radar‐based fall detection based on Doppler time–frequency signatures for assisted livingIET Radar, Sonar & Navigation, 2015