TY - JOUR
T1 - Enhancing micro-Doppler classification using Superlet based time-frequency distribution
AU - Mignone, Luca
AU - Ilioudis, Christos
AU - Clemente, Carmine
AU - Ullo, Silvia
N1 - © 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
PY - 2023/12/1
Y1 - 2023/12/1
N2 - Classical time-frequency (TF) distributions, as the short time Fourier transform (STFT) or the continuous wavelet transform (CWT), aim to enhance either the resolution in time or frequency, or attempt to strike a balance between the two. In this article, we demonstrate how a super resolution technique, the superlet-based TF distribution, named superlet transform (SLT), can boost the performance of existing classification algorithms relying on information extraction from the micro-Doppler signature. SLT is applied to provide a TF distribution with finer resolutions that would boost the performance of micro-Doppler classification approaches based on TF distributions (TFDs). This work shows the effectiveness of the integration of SLT in the processing pipeline with verification on real radar data.
AB - Classical time-frequency (TF) distributions, as the short time Fourier transform (STFT) or the continuous wavelet transform (CWT), aim to enhance either the resolution in time or frequency, or attempt to strike a balance between the two. In this article, we demonstrate how a super resolution technique, the superlet-based TF distribution, named superlet transform (SLT), can boost the performance of existing classification algorithms relying on information extraction from the micro-Doppler signature. SLT is applied to provide a TF distribution with finer resolutions that would boost the performance of micro-Doppler classification approaches based on TF distributions (TFDs). This work shows the effectiveness of the integration of SLT in the processing pipeline with verification on real radar data.
KW - time-frequency representation
KW - micro-Doppler
KW - superlet transform
KW - human motion recognition
U2 - 10.1109/TAES.2023.3312064
DO - 10.1109/TAES.2023.3312064
M3 - Article
SN - 0018-9251
VL - 59
SP - 9831
EP - 9838
JO - IEEE Transactions on Aerospace and Electronic Systems
JF - IEEE Transactions on Aerospace and Electronic Systems
IS - 6
ER -