RESUMO
This study introduces a contactless blood pressure monitoring approach that combines conventional radar signal processing with novel deep learning architectures. During the preprocessing phase, datasets suitable for synchronization are created by integrating Kalman filtering, multiscale bandpass filters, and a periodic extraction method in the time domain. These data comprise data on chest micro variations, encapsulating a complex array of physiological and biomedical information reflective of cardiac micromotions. The Radar-based Stacked Deformable convolution Network (RSD-Net) integrates channel and spatial self attention mechanisms within a deformable convolutional framework to enhance feature extraction from radar signals. The network architecture systematically employs deformable convolutions for initial deep feature extraction from individual signals. Subsequently, continuous blood pressure estimation is conducted using self attention mechanisms on feature map from single source coupled with multi-feature map channel attention. The performance of model is corroborated via the open-source dataset procured using a non-invasive 24 GHz six-port continuous wave radar system. The dataset, encompassing readings from 30 healthy individuals subjected to diverse conditions including rest, the Valsalva maneuver, apnea, and tilt-table examinations. It serves to substantiate the validity and resilience of the proposed method in the non-contact assessment of continuous blood pressure. Evaluation metrics reveal Pearson correlation coefficients of 0.838 for systolic and 0.797 for diastolic blood pressure predictions. The Mean Error (ME) and Standard Deviation (SD) for systolic and diastolic blood pressure measurements are -0.32 ±6.14 mmHg and -0.20 ±5.50 mmHg, respectively. The ablation study assesses the contribution of different structural components of the RSD-Net, validating their significance in the overall of model performance.
Assuntos
Determinação da Pressão Arterial , Pressão Sanguínea , Aprendizado Profundo , Radar , Processamento de Sinais Assistido por Computador , Humanos , Determinação da Pressão Arterial/métodos , Pressão Sanguínea/fisiologia , Adulto , Masculino , Feminino , Redes Neurais de Computação , AlgoritmosRESUMO
The potential of microwave Doppler radar in non-contact vital sign detection is significant; however, prevailing radar-based heart rate (HR) and heart rate variability (HRV) monitoring technologies often necessitate data lengths surpassing 10 s, leading to increased detection latency and inaccurate HRV estimates. To address this problem, this paper introduces a novel network integrating a frequency representation module and a residual in residual module for the precise estimation and tracking of HR from concise time series, followed by HRV monitoring. The network adeptly transforms radar signals from the time domain to the frequency domain, yielding high-resolution spectrum representation within specified frequency intervals. This significantly reduces latency and improves HRV estimation accuracy by using data that are only 4 s in length. This study uses simulation data, Frequency-Modulated Continuous-Wave radar-measured data, and Continuous-Wave radar data to validate the model. Experimental results show that despite the shortened data length, the average heart rate measurement accuracy of the algorithm remains above 95% with no loss of estimation accuracy. This study contributes an efficient heart rate variability estimation algorithm to the domain of non-contact vital sign detection, offering significant practical application value.
Assuntos
Aprendizado Profundo , Frequência Cardíaca , Radar , Determinação da Frequência Cardíaca , AlgoritmosRESUMO
An increasing number of studies on non-contact vital sign detection using radar are now beginning to turn to data-driven neural network approaches rather than traditional signal-processing methods. However, there are few radar datasets available for deep learning due to the difficulty of acquiring and labeling the data, which require specialized equipment and physician collaboration. This paper presents a new model of heartbeat-induced chest wall motion (CWM) with the goal of generating a large amount of simulation data to support deep learning methods. An in-depth analysis of published CWM data collected by the VICON Infrared (IR) motion capture system and continuous wave (CW) radar system during respiratory hold was used to summarize the motion characteristics of each stage within a cardiac cycle. In combination with the physiological properties of the heartbeat, appropriate mathematical functions were selected to describe these movement properties. The model produced simulation data that closely matched the measured data as evaluated by dynamic time warping (DTW) and the root-mean-squared error (RMSE). By adjusting the model parameters, the heartbeat signals of different individuals were simulated. This will accelerate the application of data-driven deep learning methods in radar-based non-contact vital sign detection research and further advance the field.
Assuntos
Parede Torácica , Humanos , Radar , Movimento (Física) , Movimento , Simulação por ComputadorRESUMO
Recently, synchrosqueenzing transform (SST)-based time-frequency analysis (TFA) methods have been developed for achieving the highly concentrated TF representation (TFR). However, SST-based methods suffer from two drawbacks. The first one is that the TFRs are unsatisfactory when dealing with the multicomponent signals, the instantaneous frequencies (IFs) of which are closely adjacent or intersected. Besides, the exhaustive adjustment of window length is required for SST-based methods to obtain the optimal TFR. To tackle these problems, in this article, we first analyze the concentration of TFRs for SST-based methods. A deep learning (DL)-based end-to-end replacement scheme for SST-based methods, named TFA-Net, is then proposed, which learns complete basis functions to obtain various TF characteristics of time series. The 2-D filter kernels are subsequently used for energy concentration. Different from the two-step SST-based methods where the TF transform and energy concentration are separated, the proposed end-to-end architecture makes the basis functions used for extracting TF features more beneficial to energy concentration. The comprehensive numerical experiments are conducted to demonstrate the effectiveness of the TFA-Net. The applications of the proposed method to real-world vital signs, undersea voices and micro-Doppler signatures show its great potential in analyzing nonstationary signals.
RESUMO
Frequency estimation of 2-D multicomponent sinusoidal signals is a fundamental issue in the statistical signal processing community that arises in various disciplines. In this article, we extend the DeepFreq model by modifying its network architecture and apply it to 2-D signals. We name the proposed framework 2-D ResFreq. Compared with the original DeepFreq framework, the 2-D convolutional implementation of the matched filtering module facilitates the transformation from time-domain signals to frequency-domain signals and reduces the number of network parameters. The additional upsampling layer and stacked residual blocks are designed to perform superresolution. Moreover, we introduce frequency amplitude information into the optimization function to improve the amplitude accuracy. After training, the signals in the test set are forward-mapped to 2-D accurate and high-resolution frequency representations. Frequency and amplitude estimation are achieved by measuring the locations and strengths of the spectral peaks. We conduct numerical experiments to demonstrate the superior performance of the proposed architecture in terms of its superresolution capability and estimation accuracy.
RESUMO
The progress brought by the deep learning technology over the last decade has inspired many research domains, such as radar signal processing, speech and audio recognition, etc., to apply it to their respective problems. Most of the prominent deep learning models exploit data representations acquired with either Lidar or camera sensors, leaving automotive radars rarely used. This is despite the vital potential of radars in adverse weather conditions, as well as their ability to simultaneously measure an object's range and radial velocity seamlessly. As radar signals have not been exploited very much so far, there is a lack of available benchmark data. However, recently, there has been a lot of interest in applying radar data as input to various deep learning algorithms, as more datasets are being provided. To this end, this paper presents a survey of various deep learning approaches processing radar signals to accomplish some significant tasks in an autonomous driving application, such as detection and classification. We have itemized the review based on different radar signal representations, as it is one of the critical aspects while using radar data with deep learning models. Furthermore, we give an extensive review of the recent deep learning-based multi-sensor fusion models exploiting radar signals and camera images for object detection tasks. We then provide a summary of the available datasets containing radar data. Finally, we discuss the gaps and important innovations in the reviewed papers and highlight some possible future research prospects.
RESUMO
Traditional monopulse angle estimations are mainly based on phase comparison and amplitude comparison methods, which are commonly adopted in narrowband radars. In modern radar systems, wideband radars are becoming more and more important, while the angle estimation for wideband signals is little studied in previous works. As noise in wideband radars has larger bandwidth than narrowband radars, the challenge lies in the accumulation of energy from the high resolution range profile (HRRP) of monopulse. In wideband radars, linear frequency modulated (LFM) signals are frequently utilized. In this paper, we investigate the monopulse angle estimation problem for wideband LFM signals. To accumulate the energy of the received echo signals from different scatterers of a target, we propose utilizing a cross-correlation operation, which can achieve a good performance in low signal-to-noise ratio (SNR) conditions. In the proposed algorithm, the problem of angle estimation is converted to estimating the frequency of the cross-correlation function (CCF). Experimental results demonstrate the similar performance of the proposed algorithm compared with the traditional amplitude comparison method. It means that the proposed method for angle estimation can be adopted. When adopting the proposed method, future radars may only need wideband signals for both tracking and imaging, which can greatly increase the data rate and strengthen the capability of anti-jamming. More importantly, the estimated angle will not become ambiguous under an arbitrary angle, which can significantly extend the estimated angle range in wideband radars.