A Novel Detection and Recognition Method for Continuous Hand Gesture Using FMCW Radar

Conventionally, a pulse radar is used for middle-sized or large-sized ships to detect other ships or obstacles located at a long distance. However, it is hardly equipped for most of the small-sized ships due to mounting and maintenance costs. Therefore, FMCW(frequency modulated continuous wave) radar is suggested as an alternative for the small-sized ships. Since it operates with low power and has good range resolution for relatively close objects, it is eligible for the small-sized ships. In previously proposed FMCW radar system, it only estimates distance and velocity of a target ship placed in the direction of main beam and is hard to detect several ships simultaneously. Thus, we suggest the method for detecting several ships at the same time by applying MUSIC(multiple signal classification) algorithm to FMCW radar signal received by a phased array antenna. In addition, by combining digital beam forming with the MUSIC algorithm, better angle resolution is achievable. 논문 15-40-06-19 The Journal of Korean Institute of Communications and Information Sciences '15-06 Vol.40 No.06 http://dx.doi.org/10.7840/kics.2015.40.6.113

In this paper proposed is a cost-efficient signal processing algorithm of a frequency modulation continuous wave (FMCW) radar for the close range under indoor environments. Recently, the use of an FMCW radar has been expanded to short distance from long distance due to easy production and inexpensive designing costs. In order to obtain satisfying resolution of the close range radar, it still requires a wide modulation bandwidth, which results in high costing wideband RF devices. This paper proposes a cost-efficient signal processing method using correlations between a reflected signal in frequency-domain and a lookup table which holds the frequency information depending on distances. The proposed algorithm shows the similar performance to a wideband system just by updating a firmware of low costing narrowband RF FMCW radar devices without additional hardware modification.

For hand gesture recognition (HGR) sensing based on frequency modulated continuous wave (FMCW) radar, it is challenging to cope with the ambiguity and variability of gesture motions performed by different peoples due to the private habits or position, which inevitably reduce the robustness of HGR sensing. To solve this problem, we propose a novel dual-flow fusion deformable residual network (DFDRN). More detailed gesture information, time frequency maps and spectrum map videos, are fed into DFDRN, which improves the generalization and motion modeling ability of HGR through the extensive feature space. To further improve the robustness of HGR sensing, the deformable block is exploited to replace the convolution. To focus on the important features of dual-flow fusion, the convolutional block attention module (CBAM) is embedded in the backbone of DFDRN. Experimental results show that the accuracy of DFDRN, which can reach 97.5%, is obviously better than the state-of-the-art HGR methods.

In this paper, we compare the performances of impulse radio ultra-wideband (IR-UWB) and frequency modulation continuous wave (FMCW) radars in measuring noncontact vital signs such as respiration rate and heart rate. These two type radars have been widely used in various fields and have shown their applicability to extract vital signs in noncontact ways. IR-UWB radar can extract vital signs using distance information. On the other hand, FMCW radar requires phase information to estimate vital signs, and the result can be enhanced with Multi-input Multi-output (MIMO) antenna topologies. By using commercial radar chipsets, the operation of radars under different conditions and frequency bands will also affect the performance of vital sign detection capabilities. We compared the accuracy and signal-to-noise (SNR) ratios of IR-UWB and FMCW radars in various scenarios, such as distance, orientation, carotid pulse, harmonics, and obstacle penetration. In general, the IR-UWB radars offer a slightly better accuracy and higher SNR in comparison to FMCW radar. However, each radar system has its own unique advantages, with IR-UWB exhibiting fewer harmonics and a higher SNR, while FMCW can combine the results from each channel.

A novel lightweight hand gesture recognition approach that is based on Frequency-Modulated Continuous Wave (FMCW) RADAR, which aims to minimize computational complexity and memory usage as well as maintain a high recognition performance, is proposed in this paper. Most of the existing methods utilize two-dimensional (2D) or three-dimensional (3D) features that are combined with complex neural network structures, which result in high computational costs. The proposed approach in contrast extracts four components, which include range, Doppler, azimuth, and elevation, as the one-dimensional (1D) time-series features. These features are fed into a neural network that comprises of a multi-branch temporal convolutional network, depthwise separable convolutions, and a channel attention mechanism in order to enhance the classification performance. The experiments were conducted with 9 hand gestures that were collected from 9 participants. The proposed system achieved a high accuracy of 99.38% with only 44.6K parameters and 1.84M FLOPs. Extensive ablation studies and comparative experiments against the existing models demonstrated that the proposed method effectively balances the performance and computational efficiency. This study validates the expressive capability of 1D features for hand gesture recognition and suggests practical applicability in resource-constrained environments, such as embedded systems.

Recently, hand gesture detection has gradually become a research hotspot. We propose a Region-based Faster Convolutional Neural Network (F-RCNN) gesture detection method based on Frequency Modulated Continuous Wave (FMCW) radar using 3-Dimensions Deep Convolutional Generative Adversarial Networks (3-DCGAN). Specifically, this paper adopts FMCW radar for hand gesture data acquisition, and estimates the distance of the hand gesture using the regularity of the change of echo frequency and emission frequency of radar signals. Then the semantic label maps of the generated images of distance are sent to the 3-DCGAN to extend datasets. After that, the original images and the images generated by the 3-DCGAN are simultaneously sent to F-RCNN for training. The results show that the proposed approach increases the mAP by 3% compared to the baseline F-RCNN. Besides, the proposed method not only effectively solves the problem of small amount of hand gesture data, but also the manpower and material resources consumed by collecting data.

Phase noise can severely degrade the radar performances by producing strong sidelobes which may mask small targets. Reducing those sidelobes by improving the radar analog front-end severely increases the system complexity. Another approach consists in developing waveforms more robust to phase noise. Even if the phase noise phenomenon is well known, it has been intensively investigated only in Frequency Modulated Continuous Wave (FMCW) radars, not so much in Phase Modulated Con-tinuous Wave (PMCW) radars. In this paper, we will provide a detailed analysis of the impact of phase noise on FMCW radars and on different code sequences for PMCW radars. This analysis will be used to compare the impact of phase noise on different waveforms. Both range profile and Doppler profile will be investigated. We will show that radar waveforms can be designed to provide low sidelobes in presence of phase noise. Our model will be validated by simulation.

Frequency-modulated continuous wave (FMCW) radar has seen a significant increase in use due to its ability to operate in low-power and mobile environments suited for low peak power solid state transmitters. Though there are many current applications for the FMCW technique, the majority are focused on measuring “hard” targets (e.g., cars, humans, etc.) and it is not heavily used in the measurement of volume targets such as precipitation. Moreover, current FMCW radars used for volumetric measurements are generally focused on very specific use cases such as vertical profiling. This work presents DARMA, a low-power, dual-polarized, FMCW radar with a phased array antenna capable of being deployed on unmanned aircraft systems (UAS). It will be used to test the feasibility of measuring and classifying complex weather events on a mobile platform which will provide different perspectives at the lowest kilometer of precipitation volumes.

In order to reduce the probability of automobile safety incidents, the in-vehicle occupant monitoring is indispensable. However, occupant monitoring using frequency-modulated continuous wave (FMCW) radar can be challenging due to the interference from passengers’ posture, movement, and the presence of multiple people. This paper proposes an improved method for generating point clouds using FMCW radar. The approach involves point cloud clustering, post-processing operations such as segmentation, merging, and filtering of the clustered point cloud to match the actual in-vehicle environment, and a state machine combination step. Experimental results show that the proposed method can achieve high recognition accuracy in scenarios with multiple passengers who are moving and sitting in a relaxed manner.

Human body detection is very important in the research of automotive safety technology. The extraction and analysis of human micro-motion based on frequency-modulated continuous wave (FMCW) radar is gradually receiving attention. Aimed at the modulation effect of human micro-motion on FMCW radar, a human running model is proposed to study human radar characteristics. According to the scattering characteristics of rigid bodies, the analytical expression of human running radar echoes is established. By using time–frequency analysis, the micro-Doppler features in the radar echoes are extracted during the running period. Under running conditions, the micro-Doppler characteristics of key components are studied. This model is applied to the real FMCW radar verification platform, and the runners are measured at a distance of 10 m. The fit rate of all parts of the human body can reach above 90%. The overall fit rate of the human model can reach up to 90.6%. The model proposed is a realistic and simple human kinematic model. This model, which can realize the real simulation of a running human body and provide strong support for human target radar echo analysis, can fill the deficiency of FMCW radar technology in the complex motion model.

This letter presents a novel method to reduce measurement time for remote heart rate measurement using millimeter-wave frequency-modulated continuous wave (FMCW) radar. The proposed method is based on the sampling theory and the frame structure of recent FMCW radar systems. The range displacement from heartbeat is measured by 60 GHz FMCW radar with a fixed frame interval, 50 ms. With additional chirp repetition in each frame, the heartbeat signal can be sampled with increased or decreased interval. By adopting three different sampling rates from a single radar measurement, the resolution of heart rate measurement is improved in a short time, 2.5 s. The proposed method is evaluated with a commercial off-the-shelf FMCW radar system and three different human subjects, for no respiration and natural status. To evaluate the performance of the proposed method, the heart rate result is compared with electrocardiogram (ECG) sensor and conventional signal processing method. The results show that the proposed method estimates the heart rate with lower error and shorter measurement time.

Ground Penetrating Radar (GPR) application in road surface detection has been greatly developed in the past few decades, which enables rapid and economical estimation of pavement thickness and other physical properties in non-destructive testing (NDT) and non-contact testing (NCT). In recent years, with the rapid development of microwave and millimeter-wave solid-state devices and digital signal processors, the cost of Frequency-Modulated Continuous-Wave (FMCW) radar has dropped significantly, with smaller size and lighter weight. Thereafter, FMCW GPR is considered to be applied during pavement inspection. To improve the precision of FMCW GPR for NDT and NCT of pavement thickness, a Chirp Z-transform (CZT) algorithm is introduced to FMCW GPR and investigated in this paper. A FMCW + CZT GPR at 2.5 GHz with a bandwidth of 1 GHz was built, and laboratory and field experiments were carried out. The experimental results demonstrate that the FMCW + CZT GPR radar can obtain the sample thickness with low error and recognize subtle thickness variations. This method realizes the high precision thickness measurement of shallow asphalt pavement by FMCW radar with a narrow bandwidth pulse signal and would provide a promising low-cost measurement solution for GPR.

Since respiration and heartbeat are important vital signs of humans, it is meaningful to measure them throughout the day with comfortable experience. This paper proposes the use of the frequency-modulated continuous wave (FMCW) radar for non-contact and non-invasive vital signs measurement. Specifically, the fast Fourier transform is applied to the intermediate frequency signal of the radar to obtain the human chest position, and the phase information on the measured position is collected. Then, the extracted phase is processed by the extended differential and cross-multiply algorithm. The eigenvalues selection based on singular spectrum analysis method is proposed to suppress the interference and noise. Finally, a novel VItal SIigns signals recONstruction (VISION) scheme is further proposed to extract the intrinsic mode functions (IMF) components and reconstruct the respiratory and heartbeat signals effectively. The test platform is built and the experiments are carried out to validate the feasibility and effectiveness of the proposed method, and the measured results are very close to those achieved by contact sensors.

Compressed Sensing (CS) has provided a viable approach to undersample a sparse signal and reconstruct it perfectly. In this paper, the simulation results of a frequency-modulated continuous-wave (FMCW) radar, which employs a CS based data acquisition and reconstruction algorithm to recover a sparse 2-D target frame using fewer number of scans are presented. A 16-element antenna array based on digital beamforming approach is used on the receiver end to obtain random spatial measurements of the target frame, which is the key to compressed sensing. A linear relationship is established between the total received FMCW beat signal for each scan and the 2-D sparse target frame using a basis transform matrix. Simulations of the proposed radar are performed in MATLAB and the reconstruction results for different noise levels are presented.

While traditional electrocardiogram (ECG) monitoring provides vital clinical information, its electrode-based setup restricts patient movement. To address this limitation, contactless ECG monitoring using frequency-modulated continuous-wave (FMCW) radar and deep learning has been developed. However, such approaches face challenges owing to the limited availability of training data and inherent discrepancies between radar and ECG signals. This article introduces a novel approach to transforming high-fidelity ECG signals from millimeter-wave (mmWave) radar signals reflecting cardiac mechanical activity. The proposed method uses a cascade framework with a cross-modal autoencoder trained using joint waveforms, spectrograms, and deep feature losses. This strategy enables the model to leverage a pretrained ECG-to-ECG autoencoder and a cardiac event (CE) predictor for learning general ECG representations while simultaneously capturing time- and frequency-domain features from limited data. We evaluated the effectiveness of the proposed autoencoder model in terms of signal quality and CE integrity using ablation studies on data from 20 healthy participants. The model achieved high transformation accuracy with a cross correlation of 0.914 and average timing errors below 31 ms for critical ECG features. These findings demonstrate the feasibility and effectiveness of the proposed FMCW radar-based contactless ECG monitoring method, particularly in overcoming the limitations imposed by small datasets and domain discrepancies.