micro-Doppler Signatures Research Articles

Polarimetric MDS of pedestrian

Radar observation and signature identification of human motions have a variety of applications in social security and rescue operations. Both simulation and real radar experiment are conducted to investigate the polarimetric micro-Doppler signatures (MDSs) of a pedestrian. In the simulation, the motion-captured dataset developed by Carnegie Mellon University motion graphic laboratory is first used, then both Feldberechnung bei Korpern mit beliebiger Oberflache (FEKO) and MATLAB are used to calculate the radar scattering of pedestrian’ arms, based on which, horizontal-horizontal (HH) and horizontal-vertical (HV) polarimetric MDSs are analysed. Simulation results clearly show that in the time–frequency diagram, HH micro-Doppler (m-D) of arms is rising, whereas HV m-D is falling, which is verified by practical Ka-band radar experiment on a pedestrian. The work shows that by using polarimetric radar the m-D signature of human motion can be much well detected and identified which can be further explored for classification of different people.

Animal Lameness Detection With Radar Sensing

Lameness is a significant problem for performance horses and farmed animals, with severe impact on animal welfare and treatment costs. Lameness is commonly diagnosed through subjective scoring methods performed by trained veterinary clinicians, but automatic methods using suitable sensors would improve efficiency and reliability. In this letter, we propose the use of radar micro-Doppler signatures for contactless and automatic identification of lameness, and present preliminary results for dairy cows, sheep, and horses. These proof-of-concept results are promising, with classification accuracy above 85% for dairy cows, around 92% for horses, and close to 99% for sheep.

Automatic Data-Driven Frequency-Warped Cepstral Feature Design for Micro-Doppler Classification

Micro-Doppler signature analysis and speech processing share a common approach as both rely on the extraction of features from the signal's time-frequency distribution for classification. As a result, features, such as the mel-frequency cepstrum coefficients (MFCCs), which have shown success in speech processing, have been proposed for use in micro-Doppler classification. MFCCs were originally designed to take into account the auditory properties of the human ear by filtering the signal using a filter bank spaced according to the mel-frequency scale. However, the physics underlying radar micro-Doppler is unrelated to that of human hearing or speech. This work shows that the mel-scale filter bank results in the loss of frequency components significant to the classification of radar micro-Doppler. A novel method for frequency-warped cepstral feature design is proposed as a means for optimizing the efficacy of features in a data-driven fashion specifically for micro-Doppler analysis. It is shown that the performance of the proposed frequency warped cepstral coefficients outperforms MFCC based on both simulated and measured data sets for four-class and eight-class human activity classification problems.

Radar‐ID: human identification based on radar micro‐Doppler signatures using deep convolutional neural networks

Human identification is crucial in various applications, including terrorist attack preventing, criminal seeking, defence and so on. Traditional human identification methods are usually based on vision, biological features, radio-frequency identification cards and so on. In this study, the authors propose an identification method based on radar micro-Doppler signatures using deep convolutional neural networks (DCNNs) for the first time, which can identify human in non-contact, remote and no lighting status. They employ a K-band Doppler radar to acquire the raw signals due to its stationary clutter rejection and movement detection ability as well as its short wavelength which can generate larger Doppler shift. Then short-time Fourier transform is applied to the raw signals to characterise micro-Doppler signatures. They adopt the DCNNs to deal with the spectrograms for human identification problem. The DCNNs can learn the necessary features and classification conditions from raw micro-Doppler spectrograms without employing any explicit features. While the traditional supervised learning techniques relying on the extracted features require domain knowledge of each problem. It is shown that this method can achieve average accuracy ∼97.1% for 4 people, 90.9% for 6 people, 89.1% for 8 people, 85.6% for 10 people, 77.4% for 12 people, 72.6% for 16 people and 68.9% for 20 people.

Indoor Person Identification Using a Low-Power FMCW Radar

Contemporary surveillance systems mainly use video cameras as their primary sensor. However, video cameras possess fundamental deficiencies, such as the inability to handle low-light environments, poor weather conditions, and concealing clothing. In contrast, radar devices are able to sense in pitch-dark environments and to see through walls. In this paper, we investigate the use of micro-Doppler (MD) signatures retrieved from a low-power radar device to identify a set of persons based on their gait characteristics. To that end, we propose a robust feature learning approach based on deep convolutional neural networks. Given that we aim at providing a solution for a real-world problem, people are allowed to walk around freely in two different rooms. In this setting, the IDentification with Radar data data set is constructed and published, consisting of 150 min of annotated MD data equally spread over five targets. Through experiments, we investigate the effectiveness of both the Doppler and time dimension, showing that our approach achieves a classification error rate of 24.70% on the validation set and 21.54% on the test set for the five targets used. When experimenting with larger time windows, we are able to further lower the error rate.

Personnel Recognition and Gait Classification Based on Multistatic Micro-Doppler Signatures Using Deep Convolutional Neural Networks

In this letter, we propose two methods for personnel recognition and gait classification using deep convolutional neural networks (DCNNs) based on multistatic radar micro-Doppler signatures. Previous DCNN-based schemes have mainly focused on monostatic scenarios, whereas directional diversity offered by multistatic radar is exploited in this letter to improve classification accuracy. We first propose the voted monostatic DCNN (VMo-DCNN) method, which trains DCNNs on each receiver node separately and fuses the results by binary voting. By merging the fusion step into the network architecture, we further propose the multistatic DCNN (Mul-DCNN) method, which performs slightly better than VMo-DCNN. These methods are validated on real data measured with a 2.4-GHz multistatic radar system. Experimental results show that the Mul-DCNN achieves over 99% accuracy in armed/unarmed gait classification using only 20% training data and similar performance in two-class personnel recognition using 50% training data, which are higher than the accuracy obtained by performing DCNN on a single radar node.

Detection of multiple micro‐drones via cadence velocity diagram analysis

Many studies have demonstrated the capability of radar micro-Doppler signature for classifying micro-drones. However, most existing works on radar classification of drones are based on the assumption that the received signal is only reflected from a single drone. When multiple drones are present simultaneously, the existing methods of drone classification fail due to the superimposition of the micro-Doppler features of multiple drones. In this Letter, a method for detection of multiple drones is proposed. First the time–frequency spectrogram is converted into the cadence-velocity diagram (CVD), which expresses how the curves in the time-frequency-domain repeat. Then the cadence frequency spectrum (CFS), as the basis vector of the training data from each class, of the CVD is extracted. Finally, the K-means classifier is used to recognise the component of multiple micro-drones based on the CFS. The experimental results on real radar data demonstrate that the proposed method is capable of dealing with multiple drones with satisfactory classification accuracy.

Automatic Measurement of Blade Length and Rotation Rate of Drone Using W-Band Micro-Doppler Radar

For the classification of target, one should utilize different features of target that others do not have. In case of drone as a target, classification requires input parameters, such as a number of blades, blade length, frame size, and rotations per second (RPS). In this paper, we have introduced an algorithm for automatic calculation of blade length and RPS using micro-Doppler signature. We used time–frequency domain for the analysis of received signal and the variation in micro-Doppler frequency with time gives relationship with blade length and RPS. For experimental purpose, W-band homodyne micro-Doppler radar is developed using designed low-loss high-gain modules of transmitter, low noise front-end, and downconverter. The transmitter output power is 25 dBm, front-end gain is 51 dB over 4 GHz bandwidth, and conversion loss of downconverter is −12 dB. Validation of the proposed algorithm has been done by analyzing experimental data for single rotor, double rotors, and quad rotors cases, and measurement results achieve average error of 1.3% (with maximum 6.1% error) in estimation of blade length using W-band Doppler radar.

Detection of Multiple Movers Based on Single Channel Source Separation of Their Micro-Dopplers

Studies have demonstrated the usefulness of micro-Doppler signatures for classifying dynamic radar targets such as humans, helicopters, and wind turbines. However, these classification works are based on the assumption that the propagation channel consists of only a single moving target. When multiple targets move simultaneously in the channel, the micro-Dopplers, in their radar backscatter, superimpose thereby distorting the signatures. In this paper, we propose a method to detect multiple targets that move simultaneously in the propagation channel. We first model the micro-Doppler radar signatures of different movers using dictionary learning techniques. Then, we use a sparse coding algorithm to separate the aggregate radar backscatter signal from multiple targets into their individual components. We demonstrate that the disaggregated signals are useful for accurately detecting multiple targets.

Dictionary Learning With Low Computational Complexity for Classification of Human Micro-Dopplers Across Multiple Carrier Frequencies

Recently, several machine learning algorithms have been applied for classifying micro-Doppler signatures from different human motions. However, these algorithms must demonstrate versatility in handling diversity in test and training data to be used for real-life scenarios. For example, situations may arise where the propagation channel or the presence of interference sources in the test site will permit only specific frequency bands of radar operation. These bands may differ from those used previously while training. In this paper, we examine the performances of three sparsity driven dictionary learning algorithms—synthesis, deep, and analysis—for learning unique features extracted from training data gathered across multiple carrier frequencies. These features are subsequently used for classifying test data from another distinct carrier frequency. Our experimental results, from measurement data, show that the dictionary learning algorithms are capable of extracting meaningful representations of the micro-Dopplers despite the rich frequency diversity in the data. In particular, the deep dictionary learning algorithm yields a high classification accuracy of 91% with a very low computational time for testing.

Intrinsic Mode Chirp Multicomponent Decomposition with Kernel Sparse Learning for Overlapped Nonstationary Signals Involving Big Data

We focus on the decomposition problem for nonstationary multicomponent signals involving Big Data. We propose the kernel sparse learning (KSL), developed for the T‐F reassignment algorithm by the path penalty function, to decompose the instantaneous frequencies (IFs) ridges of the overlapped multicomponent from a time‐frequency representation (TFR). The main objective of KSL is to minimize the error of the prediction process while minimizing the amount of training samples used and thus to cut the costs interrelated with the training sample collection. The IFs first extraction is decided using the framework of the intrinsic mode polynomial chirp transform (IMPCT), which obtains a brief local orthogonal TFR of signals. Then, the IFs curves of the multicomponent signal can be easily reconstructed by the T‐F reassignment. After the IFs are extracted, component decomposition is performed through KSL. Finally, the performance of the method is compared when applied to several simulated micro‐Doppler signals, which shows its effectiveness in various applications.

Deep Neural Network Initialization Methods for Micro-Doppler Classification With Low Training Sample Support

Deep neural networks (DNNs) require large-scale labeled data sets to prevent overfitting while having good generalization. In radar applications, however, acquiring a measured data set of the order of thousands is challenging due to constraints on manpower, cost, and other resources. In this letter, the efficacy of two neural network initialization techniques—unsupervised pretraining and transfer learning—for dealing with training DNNs on small data sets is compared. Unsupervised pretraining is implemented through the design of a convolutional autoencoder (CAE), while transfer learning from two popular convolutional neural network architectures (VGGNet and GoogleNet) is used to augment measured RF data for training. A 12-class problem for discrimination of micro-Doppler signatures for indoor human activities is utilized to analyze activation maps, bottleneck features, class model, and classification accuracy with respect to training sample size. Results show that on meager data sets, transfer learning outperforms unsupervised pretraining and random initialization by 10% and 25%, respectively, but that when the sample size exceeds 650, unsupervised pretraining surpasses transfer learning and random initialization by 5% and 10%, respectively. Visualization of activation layers and learned models reveals how the CAE succeeds in representing the micro-Doppler signature.

An Adaptive S-Method to Analyze Micro-Doppler Signals for Human Activity Classification.

In this paper, we propose the multiwindow Adaptive S-method (AS-method) distribution approach used in the time-frequency analysis for radar signals. Based on the results of orthogonal Hermite functions that have good time-frequency resolution, we vary the length of window to suppress the oscillating component caused by cross-terms. This method can bring a better compromise in the auto-terms concentration and cross-terms suppressing, which contributes to the multi-component signal separation. Finally, the effective micro signal is extracted by threshold segmentation and envelope extraction. To verify the proposed method, six states of motion are separated by a classifier of a support vector machine (SVM) trained to the extracted features. The trained SVM can detect a human subject with an accuracy of 95.4% for two cases without interference.

Modelling and extraction technique for micro-doppler signature of aircraft rotor blades

The process of detecting and distinguishing between different aircrafts has been a major point of interest in Defence applications. Micro-Doppler effect is one such phenomenon unique for aircrafts with different rotor dynamics and design. In this paper, we focus on deducing a mathematical model for micro-Doppler signature, of aircraft rotor blades assumed to be rotating in a plane perpendicular to the flying direction, induced on the incident radar signal. Also, we use the Wigner-Ville Distribution (WVD) to extract this signature from the radar return. This mathematical model is compared with the simulation results obtained from MATLAB, to validate the results and show the accurateness of the developed model.

Experimental Analysis of Small Drone Polarimetry Based on Micro-Doppler Signature

We present a polarimetric analysis of small drones from different aspect angles. Polarimetric analysis can provide more information of a target, since the returned radar signal is affected by different wave polarization. The analysis is performed with micro-Doppler signature (MDS) to investigate micromotions of the drone detected by the radar. We measured operating small drones in an anechoic chamber from two aspect angles, 0° and 90°. An outdoor experiment was carried out with metal clutters for verification in real environment. The indoor analysis result shows that copolarized antenna receives signals better than cross polarized when the aspect angle is 0°, and vice versa. We also verified that cross-polarized antenna receives MDS from the drone better than copolarized antenna, from outdoors when an aspect angle is almost 90°. By utilizing the polarimetric characteristic of the drone at this frequency band, it is preferable to use a polarimetric radar for drone detection.

Separation of Overlapped Non-Stationary Signals by Ridge Path Regrouping and Intrinsic Chirp Component Decomposition

In some applications, it is necessary to analyze multi-component non-stationary signals whose components severely overlap in the time-frequency (T-F) domain. Separating those signal components is desired but very challenging for existing methods. To address this issue, we propose a novel non-parametric algorithm called ridge path regrouping (RPRG) to extract the instantaneous frequencies (IFs) of the overlapped components from a T-F representation (TFR). The RPRG first detects the ridges of a multi-component signal from a TFR and then extracts the desired IFs by regrouping the ridge curves according to their variation rates at the intersections. After the IFs are obtained, component separation is achieved by using the intrinsic chirp component decomposition (ICCD) method. Different from traditional T-F filter-based methods, the ICCD can accurately reconstruct overlapped components by using a joint-estimation scheme. Finally, applications of separating some simulated and experimental micro-Doppler signals are presented to show the effectiveness of the method.

Micro-Doppler Estimation and Analysis of Slow Moving Objects in Forward Scattering Radar System

Micro-Doppler signature can convey information of detected targets and has been used for target recognition in many Radar systems. Nevertheless, micro-Doppler for the specific Forward Scattering Radar (FSR) system has yet to be analyzed and investigated in detail; consequently, information carried by the micro-Doppler in FSR is not fully understood. This paper demonstrates the feasibility and effectiveness of FSR in detecting and extracting micro-Doppler signature generated from a target’s micro-motions. Comprehensive theoretical analyses and simulation results followed by experimental investigations into the feasibility of using the FSR for detecting micro-Doppler signatures are presented in this paper. The obtained results verified that the FSR system is capable of detecting micro-Doppler signature of a swinging pendulum placed on a moving trolley and discriminating different swinging speeds. Furthermore, human movement and micro-Doppler from hand motions can be detected and monitored by using the FSR system which resembles a potential application for human gait monitoring and classification.

Looking inside

looking inside

The STFT-Based Estimator of Micro-Doppler Parameters

A two-stage technique for estimating micro-Doppler signal parameters has been proposed. In the first stage, rough parameter estimations are performed by regression of instantaneous frequency estimate obtained from the short-time Fourier transform. Afterwards, rough estimates are refined in the second stage. The proposed technique has better performance with respect to current state-of-the-art algorithms, and it reaches the Cramer–Rao lower bound for sinusoidal frequency-modulated signal parameters.

Practical investigation of multiband mono‐ and bistatic radar signatures of wind turbines

The negative effects wind farm clutter has on the performance of radar systems for air traffic control and air surveillance is well known in the radar research community and several mitigation techniques have been proposed to address this problem. These include bi- and multistatic radar systems providing multiple views of the area under surveillance, and hence potential additional information that can be used to improve the receiver performance. This study presents the analysis of a set of experimental data collected simultaneously by two radar systems, one operating at S-band and one at X-band, of echoes from an operational wind farm in the UK near Oxford. This analysis presents several parameters extracted from the time domain data and the Doppler spectra, such as Doppler centroid and bandwidth of the micro-Doppler signature as well as amplitude statistics of the time domain returns. These parameters are characterised using data recorded at mono- and bistatic nodes, as well as at different polarisation combinations.