In a smart home environment, assisted living has been a topic of great research over the past decade. Human motion analysis is considered as a key technology for living states recognition in an assisted living system. Recent research has proved that rich information can be obtained from human movements, such as the motion category, moving patterns, and human identity. In this paper, a nonintrusive human movement sensing system is established with a mono-static ultrawide bandwidth radar. Then, we propose a well-designed joint motion classification (MCL) and person identification (PID) convolutional neural network (named as “JMI-CNN”). To recognize human motions and identities simultaneously, the network employs a multitask learning scheme as well as the attention mechanism and the hierarchical feature reuse strategies. We report the experimental result on the data from 15 individuals, each performing six motions. It shows that the model achieves a promising performance of 80.57% on the joint task, while the accuracy for MCL and PID are 98.50% and 80.92%, respectively. Moreover, we carry out ablation studies to evaluate the design principles of the proposed method. Discussions on the impact of signal noise ratio and slow-time window length are also conducted.

Joint Motion Classification and Person Identification via Multitask Learning for Smart Homes

A complete list of the titles in this series appears at the end of this volume. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Inverse Synthetic Aperture Radar Imaging with MATLAB® Algorithms: With Advanced SAR/ISAR Imaging Concepts, Algorithms, and MATLAB® Codes

For a conventional narrow-band radar system, the detectable information of the target is limited, and it is difficult for the radar to accurately identify the target type. In particular, the classification probability will further decrease when part of the echo data is missed. By extracting the target features in time and frequency domains from multi-wave gates sparse echo data, this paper presents a classification algorithm in conventional narrow-band radar to identify three different types of aircraft target, i.e., helicopter, propeller and jet. Firstly, the classical sparse reconstruction algorithm is utilized to reconstruct the target frequency spectrum with single-wave gate sparse echo data. Then, the micro-Doppler effect caused by rotating parts of different targets is analyzed, and the micro-Doppler based features, such as amplitude deviation coefficient, time domain waveform entropy and frequency domain waveform entropy, are extracted from reconstructed echo data to identify targets. Thirdly, the target features extracted from multi-wave gates reconstructed echo data are weighted and fused to improve the accuracy of classification. Finally, the fused feature vectors are fed into a support vector machine (SVM) model for classification. By contrast with the conventional algorithm of aircraft target classification, the proposed algorithm can effectively process sparse echo data and achieve higher classification probability via weighted features fusion of multi-wave gates echo data. The experiments on synthetic data are carried out to validate the effectiveness of the proposed algorithm.

Aircraft Target Classification for Conventional Narrow-Band Radar with Multi-Wave Gates Sparse Echo Data

The precession period was used to identify space targets in radar target recognition, especially the axisymmetric ballistic missile warheads and the similar shaped decoys. There are many precession period extraction methods based on RCS sequence. However, these methods have many restrictions and often yield poor results under noise condition. Aiming at extracting the precession period from RCS sequence, this paper designed a one-dimensional convolutional neural network. The precession period extraction is converted to a signal parameter estimation problem where the RCS sequence is the input signal and the period is the expected parameter. The proposed method was trained and validated on simulated RCS sequences and compared with spectral method, including CAUTOC, CAMDF, CAUTOC/CAMDF and trigonometric fitting method. The results showed that the proposed method yielded more accurate estimation results. Moreover, it can tell that there is no valid period by yielding a value that is not in the range [2, N /2]where $N$ is the length of the RCS sequence.

Precession Period Extraction of Axisymmetric Space Target from RCS Sequence via Convolutional Neural Network

The motion of ballistic midcourse targets include orbit motion and micro-motion,compared with orbit motion,the micro-motion of ballistic targets causes the target more rapid change relative to the angle of radar sight,which is benefit to the inverse synthetic aperture radar(ISAR) imaging.According to the micro-motion characteristic of spin symmetric warhead,a new micro-motion imaging algorithm based on the matching pursuit(MP) sparse decomposition is proposed,the imaging plane and accumulation angle are analyzed.Compared with the traditional range-Doppler,Wigner-Ville imaging algorithm,the result of simulation indicates that the new algorithm has a better imaging precision and with no cross interference,which is a kind of effective algorithm for the micro-motion target imaging.

A novel ISAR imaging method of ballistic midcourse targets based on MP sparse decomposition

Nutation and geometrical parameters estimation of cone-shaped target based on micro-Doppler effect

The paper proposed a method to estimate 5 parameters for the question of rotor target recognition, which included the rotation rate, blade number, rotor disk attitude angle, rotor disk radius and the distance of blade root from the rotor disk center. The method generated the electromagnetic scattering data obtained from different radar based on physical optics and edge diffraction, the period of the echo response of the rotor target was obtained by the autocorrelation, the ISAR image sequence was gained by the range-instantaneous doppler algorithm, the blade number could be obtained from the image and the rotation rate was obtained, the paper also deduced the nonlinear three variables equation combined of rotor disk attitude angle and rotor disk radius, so the cross range scaling of the rotor ISAR image was realized, the projection relationship of the rotor target on the ISAR image was analyzed, the way to get the distance of blade root from the rotor disk center was proposed. Finally, the method was proved to be effective by the simulation.

http://ieeexplore.ieee.org/iel7/7400327/7405778/07405896.pdf

Feature Parameters Estimation of the Rotor Target Based on ISAR Image Sequences

The invention relates to a short-distance target radar detection method based on stepped frequency signals. Antennas adopt a bistatic mode, one antenna is used for transmitting detection signals, another antenna is used for receiving target scattering echo signals, the detection signals transmitted every time are single-frequency continuous waves, the detection signals are transmitted in a stepped frequency mode, the times of transmitting the detection signals are the frequency number of the detection signals N=B/delta f, wherein B refers to frequency bandwidth, and delta f refers to the step length of the frequency. Corresponding to the detection signals with N frequencies, the received target scattering echo signals are si, i=0, 1,..., N-1, the target scattering echo signals are subjected to inverse Fourier transform, and one-dimensional imaging of a detection target is realized according to the propagation speed of electromagnetic waves.

Chang-Yong Shu

Papers

Nutation and geometrical parameters estimation of cone-shaped target based on micro-Doppler effect

Feature Parameters Estimation of the Rotor Target Based on ISAR Image Sequences

Short-distance target detection radar based on stepped frequency signals