Abstract
Feature representation based on the high resolution range profile (HRRP) is the key technology in radar automatic target recognition(RATR). In this paper, we design a deep-u-blind denoising network(DUBDNet) to extract features with high-noise-stability. The fully convolutional DUBDNet is based on autoencoder and employs fusion layers to transfer input features to high dimensional space. Then radar HRRP shift-robust convolutional neural network(RSRNet) is proposed as the classifier. In the experiment, two radar sensors are used to measure HRRP signals of warplanes and civil airplanes. RSRNet performs high robustness to HRRP time-shift sensitivity via testing translation data. This is also a proof that convolutional neural network(CNN) is shift-robust on HRRP target recognition. Trained with noise-to-noise, DUBDNet can achieve blind-denoising in low signal-to-noise ratio(SNR) and significantly improve the correct recognition rate of targets. When the SNR of input HRRP signals is less than 5 dB, DUBDNet can increase the SNR by 10 dB. When the input SNR is -15 dB, output SNR can be increased by 15 dB and the correct recognition rate of targets can be increased by 15%.
Highlights
high resolution range profile (HRRP) represents the distribution of target centers in line sight of radar sensors
Because of low computation complexity and containing structure information of targets, HRRP is frequently used in radar automatic target recognition(RATR)
DUBDNet is a fully convolutional denoising network based on autoencoder, we propose a novel training mode mapping from noise to noise and achieve blind- denoising
Summary
HRRP represents the distribution of target centers in line sight of radar sensors. HRRP is to be obtained and stored by high-resolution radars. Because of low computation complexity and containing structure information of targets, HRRP is frequently used in radar automatic target recognition(RATR). With the development of HRRP target recognition, many methods are proposed and improved in RATR [1]. HRRP datasets are mainly obtained by cooperative radar sensors, which has high-SNR and uniform azimuthaltitude angle. In real war applications, the data source of radar sensor is non-cooperative and has unknown noise, which would be faced with harsh natural environment such as sea clutter, rainfall, hurricane and so on. How to address unknown noise in complex high-noise environment is a challenge [2], [3]. The noise environment is always complex and the algorithm needs to be designed based on the
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
