Dataset Of Signals Research Articles

Efficient Convolutional Networks for Robust Automatic Modulation Classification in OFDM-Based Wireless Systems

Orthogonal frequency-division multiplexing (OFDM) is commonly deployed in Internet of Things (IoT) systems to achieve high data rates with reasonable complexity, where noncooperative protocols encode signals with different modulations for multiple users. However, conventional modulation recognition is limited to single-carrier communication systems, and there is an urgent need to design an effective method to blindly identify the unknown modulation format of received signals. In this article, we propose an automatic modulation classification method for the OFDM systems with the presence of channel deterioration. Our method first leverages a data reconstruction mechanism to arrange signals into high-dimensional data arrays and then exploits an efficient convolutional network, namely OFDMsym-Net, to learn underlying radio characteristics. OFDMsym-Net is designed by two kinds of processing modules, which manipulate one-dimensional asymmetric convolution filters to extract the intracorrelation within an OFDM symbol and the intercorrelation between different symbols. Moreover, a sophisticated structure with addition and concatenation layers is developed inside every module to improve learning efficiency. Based on simulation results achieved on a synthetic dataset of OFDM signals, our proposed method shows the classification robustness under various channel impairments. It reveals to have an overall accuracy of 95.41% at 10-dB signal-to-noise ratio, being superior to several state-of-the-art approaches.

Study on the Tribological Properties of Polymer Interface and the Mapping Mechanism of Friction Noise in a Wide Temperature Range

In order to study the tribological properties and frictional noise mapping mechanism of polymer interface at a wide temperature range, tribological and frictional noise acquisition tests were carried out on a Ball-on-Three-Plates Tribology Accessory. The friction coefficient and noise signal data sets were collected for 15 minutes from the friction surfaces of polytetrafluoroethylene and 0Cr18Ni9 austenitic steel pairs at six temperatures and three constant loading-speed contact points in a wide temperature range of -150°C-25°C. The wear rate of the polymer was measured and the wear state of the polymer interface was studied by field emission environmental scanning electron microscopy (SEM). Based on the analysis of the time-frequency domain characteristics of friction noise and the qualitative mapping relationship between the wide-temperature domain tribological properties and friction noise, the theoretical and experimental basis is provided for establishing online early warning monitoring of the tribological performance of polymer-metal friction pairs based on the time-frequency domain characteristics of friction noise.

Learning and controlling the source-filter representation of speech with a variational autoencoder

Understanding and controlling latent representations in deep generative models is a challenging yet important problem for analyzing, transforming and generating various types of data. In speech processing, inspiring from the anatomical mechanisms of phonation, the source-filter model considers that speech signals are produced from a few independent and physically meaningful continuous latent factors, among which the fundamental frequency f0 and the formants are of primary importance. In this work, we start from a variational autoencoder (VAE) trained in an unsupervised manner on a large dataset of unlabeled natural speech signals, and we show that the source-filter model of speech production naturally arises as orthogonal subspaces of the VAE latent space. Using only a few seconds of labeled speech signals generated with an artificial speech synthesizer, we propose a method to identify the latent subspaces encoding f0 and the first three formant frequencies, we show that these subspaces are orthogonal, and based on this orthogonality, we develop a method to accurately and independently control the source-filter speech factors within the latent subspaces. Without requiring additional information such as text or human-labeled data, this results in a deep generative model of speech spectrograms that is conditioned on f0 and the formant frequencies, and which is applied to the transformation speech signals. Finally, we also propose a robust f0 estimation method that exploits the projection of a speech signal onto the learned latent subspace associated with f0.

Fractional Dynamics Foster Deep Learning of COPD Stage Prediction.

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of death worldwide. Current COPD diagnosis (i.e., spirometry) could be unreliable because the test depends on an adequate effort from the tester and testee. Moreover, the early diagnosis of COPD is challenging. The authors address COPD detection by constructing two novel physiological signals datasets (4432 records from 54 patients in the WestRo COPD dataset and 13824 medical records from 534 patients in the WestRo Porti COPD dataset). The authors demonstrate their complex coupled fractal dynamical characteristics and perform a fractional-order dynamics deep learning analysis to diagnose COPD. The authors found that the fractional-order dynamical modeling can extract distinguishing signatures from the physiological signals across patients with all COPD stages-from stage 0 (healthy) to stage 4 (very severe). They use the fractional signatures to develop and train a deep neural network that predicts COPD stages based on the input features (such as thorax breathing effort, respiratory rate, or oxygen saturation). The authors show that the fractional dynamic deep learning model (FDDLM) achieves a COPD prediction accuracy of 98.66% and can serve as a robust alternative to spirometry. The FDDLM also has high accuracy when validated on a dataset with different physiological signals.

A YOLOX-Based Automatic Monitoring Approach of Broken Wires in Prestressed Concrete Cylinder Pipe Using Fiber-Optic Distributed Acoustic Sensors.

Wire breakage is a major factor in the failure of prestressed concrete cylinder pipes (PCCP). In the presented work, an automatic monitoring approach of broken wires in PCCP using fiber-optic distributed acoustic sensors (DAS) is investigated. The study designs a 1:1 prototype wire break monitoring experiment using a DN4000 mm PCCP buried underground in a simulated test environment. The test combines the collected wire break signals with the previously collected noise signals in the operating pipe and transforms them into a spectrogram as the wire break signal dataset. A deep learning-based target detection algorithm is developed to detect the occurrence of wire break events by extracting the spectrogram image features of wire break signals in the dataset. The results show that the recall, precision, F1 score, and false detection rate of the pruned model reach 100%, 100%, 1, and 0%, respectively; the video detection frame rate reaches 35 fps and the model size is only 732 KB. It can be seen that this method greatly simplifies the model without loss of precision, providing an effective method for the identification of PCCP wire break signals, while the lightweight model is more conducive to the embedded deployment of a PCCP wire break monitoring system.

Comparative karyotype analysis of eight Cucurbitaceae crops using fluorochrome banding and 45S rDNA-FISH.

To have an insight into the karyotype variation of eight Cucurbitaceae crops including Cucumissativus Linnaeus, 1753, Cucumismelo Linnaeus, 1753, Citrulluslanatus (Thunberg, 1794) Matsumura et Nakai, 1916, Benincasahispida (Thunberg, 1784) Cogniaux, 1881, Momordicacharantia Linnaeus, 1753, Luffacylindrica (Linnaeus, 1753) Roemer, 1846, Lagenariasicerariavar.hispida (Thunberg, 1783) Hara, 1948 and Cucurbitamoschata Duchesne ex Poiret, 1819, well morphologically differentiated mitotic metaphase chromosomes were prepared using the enzymatic maceration and flame-drying method, and the chromosomal distribution of heterochromatin and 18S-5.8S-26S rRNA genes (45S rDNA) was investigated using sequential combined PI and DAPI (CPD) staining and fluorescence in situ hybridization (FISH) with 45S rDNA probe. Detailed karyotypes were established using the dataset of chromosome measurements, fluorochrome bands and rDNA FISH signals. Four karyotype asymmetry indices, CVCI, CVCL, MCA and Stebbins' category, were measured to elucidate the karyological relationships among species. All the species studied had symmetrical karyotypes composed of metacentric and submetacentric or only metacentric chromosomes, but their karyotype structure can be discriminated by the scatter plot of MCA vs. CVCL. The karyological relationships among these species revealed by PCoA based on x, 2n, TCL, MCA, CVCL and CVCI was basically in agreement with the phylogenetic relationships revealed by DNA sequences. CPD staining revealed all 45S rDNA sites in all species, (peri)centromeric GC-rich heterochromatin in C.sativus, C.melo, C.lanatus, M.charantia and L.cylindrica, terminal GC-rich heterochromatin in C.sativus. DAPI counterstaining after FISH revealed pericentromeric DAPI+ heterochromatin in C.moschata. rDNA FISH detected two 45S loci in five species and five 45S loci in three species. Among these 45S loci, most were located at the terminals of chromosome arms, and a few in the proximal regions. In C.sativus, individual chromosomes can be precisely distinguished by the CPD band and 45S rDNA signal patterns, providing an easy method for chromosome identification of cucumber. The genome differentiation among these species was discussed in terms of genome size, heterochromatin, 45S rDNA site, and karyotype asymmetry based on the data of this study and previous reports.

Efficient Energy Management Strategy for an Electric Vehicle Powered by a Hybrid Energy Storage System Based on Hybrid GBDT-RSA Approach

This paper proposes an efficient energy management scheme for an EV with a hybrid energy storage system like super capacitor and battery based on hybrid optimization method. The proposed hybrid approach is a parallel performance of gradient boosting decision tree algorithm and reptile search algorithm. The major purpose is to diminish the variance among real and reference power on battery and super capacitor. The HESS is divided into two segments: (1) Determine the super capacitor reference voltage, which is affected by load dynamics. (2) Maximize the power flow via the HESS. The super capacitor reference voltage is first calculated by assessing the load dynamics on real time, such as vehicle dynamics, motor characteristics, and regenerative braking systems. The super capacitor’s input parameters include load current, battery current, and state of charge. The RSA is created on proposed system by combining the probable data set of HESS control signals. The GBDT algorithm is trained and forecasts optimal HESS parameters using the RSA’s practiced data set. The proposed approach also improves battery current magnitude, super capacitor voltage, battery current ranges, and battery power. It optimizes the HESS parameter provides specific solutions. The proposed hybrid system is implemented on MATLAB and its performance is related to that of existing approaches.

Cross Dataset Analysis for Generalizability of HRV-Based Stress Detection Models.

Stress is an increasingly prevalent mental health condition across the world. In Europe, for example, stress is considered one of the most common health problems, and over USD 300 billion are spent on stress treatments annually. Therefore, monitoring, identification and prevention of stress are of the utmost importance. While most stress monitoring is carried out through self-reporting, there are now several studies on stress detection from physiological signals using Artificial Intelligence algorithms. However, the generalizability of these models is only rarely discussed. The main goal of this work is to provide a monitoring proof-of-concept tool exploring the generalization capabilities of Heart Rate Variability-based machine learning models. To this end, two Machine Learning models are used, Logistic Regression and Random Forest to analyze and classify stress in two datasets differing in terms of protocol, stressors and recording devices. First, the models are evaluated using leave-one-subject-out cross-validation with train and test samples from the same dataset. Next, a cross-dataset validation of the models is performed, that is, leave-one-subject-out models trained on a Multi-modal Dataset for Real-time, Continuous Stress Detection from Physiological Signals dataset and validated using the University of Waterloo stress dataset. While both logistic regression and random forest models achieve good classification results in the independent dataset analysis, the random forest model demonstrates better generalization capabilities with a stable F1 score of 61%. This indicates that the random forest can be used to generalize HRV-based stress detection models, which can lead to better analyses in the mental health and medical research field through training and integrating different models.

A SOM-CNN Algorithm for NLOS Signal Identification

Aiming at the problem of non-line of sight (NLOS) signal recognition for Ultra Wide Band (UWB) positioning, we utilize the concepts of Neural Network Clustering and Neural Network Pattern Recognition. We propose a classification algorithm based on self-organizing feature mapping (SOM) neural network batch processing, and a recognition algorithm based on convolutional neural network (CNN). By assigning different weights to learning, training and testing parts in the data set of UWB location signals with given known patterns, a strong NLOS signal recognizer is trained to minimize the recognition error rate. Finally, the proposed NLOS signal recognition algorithm is verified using data sets from real scenarios. The test results show that the proposed algorithm can solve the problem of UWB NLOS signal recognition under strong signal interference. The simulation results illustrate that the proposed algorithm is significantly more effective compared with other algorithms.

Indoor Geomagnetic Positioning Using Direction-Aware Multiscale Recurrent Neural Networks

Accurate indoor positioning has become an indispensable technology for location-based service in indoor environments. Geomagnetic field fingerprinting for indoor positioning is an attractive alternative to Wi-Fi and Bluetooth because of the omnipresent signals and the infrastructure-free mode. However, heterogeneous devices and users with fluctuant geomagnetism and massive crowdsourced data covering large-scale indoor scenes may result in a low degree of discernibility and degrade the rerecognizable performance for pedestrian positioning. To overcome the problem, a deep-learning-based indoor geomagnetic positioning method with direction-aware multiscale recurrent neural networks (DM-RNNs) is presented. First, the direction information is taken into the fingerprint construction and localization process to increase spatial identification, which is necessary for indoor pedestrian navigation. Second, instead of using a single holistic feature from sequence fingerprints directly, we employ the different scale-based feature extraction units for variational anomalies of the signal by using multiscale RNNs, increasing the model adaptability and generality for random pedestrian motion and device heterogeneity. Third, the ensemble learning mechanism is adopted in the model to obtain robust localization results. Furthermore, we employ the convenient visual-inertial odometry (VIO) to collect the geomagnetic signal dataset and utilize the sequence augmentation approach to generate synthesized trajectories from multiple single-point magnetic values for training and testing the model. Extensive experiments in three different indoor scenes demonstrate that the proposed approach outperforms state-of-the-art competing schemes by a wide margin, reducing mean localization error by more than 30% and obtaining over 79% direction estimation precision in different indoor scenarios.

A Scalable and Transferable Federated Learning System for Classifying Healthcare Sensor Data.

With the development of Internet of Medical Things, massive healthcare sensor data (HSD) are transmitted in the Internet, which faces various security problems. Healthcare data are sensitive and important for patients. Automatic classification of HSD has significant value for protecting the privacy of patients. Recently, the edge computing-based federated learning has brought new opportunities and challenges. It is difficult to develop a lightweight HSD classification system for edge computing. In particular, the classification system should consider the dynamic characteristics of HSD, e.g., the change of data distributions and the appearance of initially unknown classes. To solve these problems, the paper proposes a scalable and transferable classification system, called SCALT. It is a one-classifier-per-class system based on federated learning. It comprises a one-dimensional convolution-based network for feature extraction, and an individual mini-classifier for each class. It is easy to be scaled when new class appears since only a mini-classifier will be trained. The feature extractor is updated only when it is transferred to a new task. SCALT has a parameter protection mechanism, which can avoid catastrophic forgetting in sequential HSD classification tasks. We conduct comprehensive experiments to evaluate SCALT on three different physiological signal datasets: Electrocardiogram, Electroencephalogram and Photoplethysmograph. The accuracies on the three datasets are 98.65%, 91.10% and 89.93% respectively, which are higher than the compared state-of-the-art works. At last, an application of applying SCALT to protect the privacy of patients is presented.

Blind Detection of Broadband Signal Based on Weighted Bi-Directional Feature Pyramid Network.

With the development of wireless technology, signals propagating in space are easy to mix, so blind detection of communication signals has become a very practical and challenging problem. In this paper, we propose a blind detection method for broadband signals based on a weighted bi-directional feature pyramid network (BiFPN). The method can quickly perform detection and automatic modulation identification (AMC) on time-domain aliased signals in broadband data. Firstly, the method performs a time-frequency analysis on the received signals and extracts the normalized time-frequency images and the corresponding labels by short-time Fourier transform (STFT). Secondly, we build a target detection model based on YOLOv5 for time-domain mixed signals in broadband data and learn the features of the time-frequency distribution image dataset of broadband signals, which achieves the purpose of training the model. The main improvements of the algorithm are as follows: (1) a weighted bi-directional feature pyramid network is used to achieve a simple and fast multi-scale feature fusion approach to improve the detection probability; (2) the Efficient-Intersection over Union (EIOU) loss function is introduced to achieve high accuracy signal detection in a low Signal-Noise Ratio (SNR) environment. Finally, the time-frequency images are detected by an improved deep network model to complete the blind detection of time-domain mixed signals. The simulation results show that the method can effectively detect the continuous and burst signals in the broadband communication signal data and identify their modulation types.

Graph learning from band-limited data by graph Fourier transform analysis

A graph provides an effective means to represent the statistical dependence or similarity among signals observed at different vertices. A critical challenge is to excavate graphs underlying observed signals, because of non-convex problem structure and associated high computational requirements. This paper presents a new graph learning technique that is able to efficiently infer the graph structure underlying observed graph signals. The key idea is that we reveal the intrinsic relation between the frequency-domain representation of general band-limited graph signals, and the graph Fourier transform (GFT) basis. Accordingly, we derive a new closed-form analytic expression for the GFT basis, which depends deterministically on the observed signals (as opposed to being solved numerically and approximately in the literature). Given the GFT basis, the estimation of the graph Laplacian, more explicitly, its eigenvalues, is convex and efficiently solved using the alternating direction method of multipliers (ADMM). Simulations based on synthetic data and experiments based on public weather and brain signal datasets show that the new technique outperforms the state of the art in accuracy and efficiency.

An accurate automated speaker counting architecture based on James Webb Pattern

Speaker counting is an important research area in sound forensics. There are limited speaker counting papers in the literature, as it is challenging to collect datasets. This work aims to collect a new overlapping speech signal dataset for speaker counting and propose a novel feature engineering model. In this work, textural feature extraction is based on the iconic James Webb space telescope; hence, this pattern is named James Webb Pattern (JWPat). A new speaker counting speech dataset comprising 3,121 speeches divided into 32 classes (the class number corresponded to the number of speakers) was collected. A new framework that mimics the deep learning model has been proposed to classify the collected speech classes. The proposed feature engineering model is self-organized and uses various mother wavelet functions to generate features at both low and high levels. We have obtained the best classification accuracy of 86.74% using the symlet4 mother wavelet function. Using our proposed framework, eight classification results have been calculated with accuracy ranging from 75.94% to 86.74%. This range is over 10% accuracy, and it demonstrates the effect of the mother wavelet function on the classification performance. Moreover, the feature extraction capability of the mirror of the James Webb telescope has been demonstrated. Our proposed method yielded 86.74% accuracy on a large dataset and indicated the success of our proposed model.

Photoplethysmography Data Reduction Using Truncated Singular Value Decomposition and Internet of Things Computing

Biometric-based identity authentication is integral to modern-day technologies. From smart phones, personal computers, and tablets to security checkpoints, they all utilize a form of identity check based on methods such as face recognition and fingerprint-verification. Photoplethysmography (PPG) is another form of biometric-based authentication that has recently been gaining momentum, because it is effective and easy to implement. This paper considers a cloud-based system model for PPG-authentication, where the PPG signals of various individuals are collected with distributed sensors and communicated to the cloud for authentication. Such a model incursarge signal traffic, especially in crowded places such as airport security checkpoints. This motivates the need for a compression–decompression scheme (or a Codec for short). The Codec is required to reduce the data traffic by compressing each PPG signal before it is communicated, i.e., encoding the signal right after it comes off the sensor and before it is sent to the cloud to be reconstructed (i.e., decoded). Therefore, the Codec has two system requirements to meet: (i) produce high-fidelity signal reconstruction; and (ii) have a computationallyightweight encoder. Both requirements are met by the Codec proposed in this paper, which is designed using truncated singular value decomposition (T-SVD). The proposed Codec is developed and tested using a publicly available dataset of PPG signals collected from multiple individuals, namely the CapnoBase dataset. It is shown to achieve a 95% compression ratio and a 99% coefficient of determination. This means that the Codec is capable of delivering on the first requirement, high-fidelity reconstruction, while producing highly compressed signals. Those compressed signals do not require heavy computations to be produced as well. An implementation on a single-board computer is attempted for the encoder, showing that the encoder can average 300 milliseconds per signal on a Raspberry Pi 3. This is enough time to encode a PPG signal prior to transmission to the cloud.

A Novel Method for Few-Shot Specific Emitter Identification in Non-Cooperative Scenarios

Obtaining larger category-label-containing training signal datasets in non-cooperative scenarios is difficult. Moreover, employing smaller labeled signal datasets for specific emitter identification is technically challenging. Therefore, we propose a novel method for few-shot SEI. We first design a bispectral analysis and Radon transformation-based signal preprocessing scheme to obtain feature vectors that effectively characterize the radio frequency fingerprints. The feature vectors are then fed to a network model for feature learning. Moreover, a meta-learning algorithm is applied to the network model to adapt to few-shot feature learning. The conventional meta-learning algorithm is improved to develop a novel algorithm involving latent embedding optimization for meta-learning. The proposed method extracts low-dimensional key features from high-dimensional input data and evaluates the distance and degree of feature dispersion. The resulting information is employed in sample point prediction. The algorithm effectively achieves few-shot SEI and offers stable and efficient recognition after training with a minimum of forty samples. This method identifies emitter individuals under multiple modulation types and exhibits scalability in identifying the emitter numbers. Moreover, it offers adaptability in identifying the emitter individuals under multiple propagation channel types.

Augmentation-Assisted Robust Fringe Detection on Unseen Experimental Signals Applied to Optical Feedback Interferometry Using a Deep Network

Commercialization and industrial deployment of optical feedback interferometry or self-mixing interferometry (SMI) based displacement instruments is held back due to inaccurate fringe detection under different optical feedback or speckle induced by operating conditions. In this work, we propose using deep neural 2D networks, which have renowned generalization performance on unseen data. Nonetheless, training deep neural networks requires very large data which in our application would imply acquiring large datasets of experimental signals by operating such interferometers under as many operating conditions as possible. To circumvent this time- and resource-consuming process, we propose a novel data augmentation scheme which increases the amount of training data needed by a deep network for fringe detection/classification in SMI signals. Interestingly, this has enabled the trained deep network to acquire excellent generalization capability where it has learnt to detect SMI fringes belonging to weak-, and very strong-optical feedback regime, even when it was only trained on moderate-, and strong-feedback regime signals. Consequently, our trained model has shown robust performance for simulated weak-, moderate-, and strong-optical feedback regime SMI signals affected by additive noise. Various experimental SMI signals, acquired under different sensor- and optical-conditions, have also been successfully processed. We also implemented an established fringe detection method for comparison. Our work presents very good generalization capability as compared to this established method. Our novel augmentation scheme is generic and can be applied for other interferometric signals. We have released our dataset and implementation, with the hope that this will assist community in accelerating the commercialization of optical feedback interferometry leveraging the full potential of deep learning.

A Real-time ECG CTG based Ensemble Feature Extraction and Unsupervised Learning based Classification Framework for Multi-class Abnormality Prediction

Cardiovascular diseases (CVDs) are a leading cause of death worldwide. Early detection and diagnosis of these diseases can greatly reduce complications and improve outcomes for high-risk individuals. One method for detecting CVDs is through the use of electrocardiogram (ECG) monitoring systems, which use various technologies such as the Internet of Things (IoT), mobile applications, wireless sensor networks (WSN), and wearable devices to acquire and analyze ECG data for early diagnosis. However, despite the prevalence of these systems in the literature, there is a need for further optimization and improvement of their classification accuracy. In an effort to address this challenge, a novel heterogeneous unsupervised learning model for real-time ECG classification was proposed. The main goal of this work was to reduce the error rate and improve the classification accuracy of the system. This study presents a framework for the classification of multi-class abnormalities in electrocardiograms (ECGs) using an ensemble feature extraction technique and unsupervised learning. The framework utilizes a real-time electrocardiogram-cardiotocography (ECG-CTG) system to extract features from the ECG signal, and then employs an ensemble of feature extraction techniques to enhance the discrimination of the extracted features. The extracted features are then used in an unsupervised learning-based classification algorithm to classify the ECG signals into different classes of abnormalities. The proposed framework is evaluated on a dataset of ECG signals and the results show that it can effectively classify ECG signals with high accuracy and low computational complexity.

Accelerometer-Based Bed Occupancy Detection for Automatic, Non-Invasive Long-Term Cough Monitoring

We present a new machine learning based bed-occupancy detection system that uses the accelerometer signal captured by a bed-attached consumer smartphone. Automatic bed-occupancy detection is necessary for automatic long-term cough monitoring, since the time which the monitored patient occupies the bed is required to accurately calculate a cough rate. Accelerometer measurements are more cost effective and less intrusive than alternatives such as video monitoring or pressure sensors. A 249-hour dataset of manually-labelled acceleration signals gathered from seven patients undergoing treatment for tuberculosis (TB) was compiled for experimentation. These signals are characterised by brief activity bursts interspersed with long periods of little or no activity, even when the bed is occupied. To process them effectively, we propose an architecture consisting of three interconnected components. An occupancy-change detector locates instances at which bed occupancy is likely to have changed, an occupancy-interval detector classifies periods between detected occupancy changes and an occupancy-state detector corrects falsely-identified occupancy changes. Using long short-term memory (LSTM) networks, this architecture was demonstrated to achieve an AUC of 0.94. When integrated into a complete cough monitoring system, the daily cough rate of a patient undergoing TB treatment was determined over a period of 14 days. As the colony forming unit (CFU) counts decreased and the time to positivity (TPP) increased, the measured cough rate decreased, indicating effective TB treatment. This provides a first indication that automatic cough monitoring based on bed-mounted accelerometer measurements may present a non-invasive, non-intrusive and cost-effective means of monitoring long-term recovery of TB patients.

A Study on 3D Human Pose Estimation Using Through-Wall IR-UWB Radar and Transformer

In this paper, we propose a human pose estimation algorithm for an impulse radio ultra-wideband (IR-UWB) radar based on the transformer-based deep learning model. We have built an IR-UWB radar system with an 8-by-8 multiple-input multiple-output (MIMO) antenna array. The IR-UWB radar system in our paper is advantageous for the through-wall detection application since it operates on a very low frequency range (i.e., 0.45 to 3.55 GHz) compared to other existing works on RF-based human pose estimation. Moreover, the human pose estimation by an IR-UWB radar has not been studied very well in other existing works since all existing works have used a frequency-modulated continuous wave (FMCW) radar or a WiFi device. We propose a 3D-TransPOSE algorithm for the 3D human pose estimation from the IR-UWB radar signals. The proposed algorithm is designed based on the transformer architecture. While the transformer has actively been studied in the natural language processing (NLP) or vision domains, no prior work has applied the transformer model to the RF-based human pose estimation problem. The attention mechanism of the proposed algorithm is able to focus on the relevant time segments of the IR-UWB radar signals for detecting the human pose, eliminating the needs of converting radar signals to a voxelized 3D image. We have gathered a large dataset of IR-UWB radar signals labeled with 3D human skeletons, and shown that the proposed algorithm is able to detect human skeletons with a high accuracy.