Super-Resolution Time Delay Estimation of OFDM Passive Radar Based on Fully Convolutional Networks

This article studies the time delay resolution of radar, which refers to the ability to distinguish the delay of multiple adjacent targets. We propose a novel coarse-fine joint super-resolution time delay estimation (TDE) strategy for passive radar using orthogonal frequency division multiplexing (O...

Full description

Saved in:
Bibliographic Details
Published in:IEEE sensors journal 2023-11, Vol.23 (21), p.26347-26356
Main Authors: Wang, Chujun, Wan, Xianrong, Yi, Jianxin, Cheng, Feng
Format: Article
Language:English
Subjects:
Delay effects
Delays
Doppler effect
Estimation
Fully convolutional network (FCN)
OFDM
Orthogonal Frequency Division Multiplexing
orthogonal frequency division multiplexing (OFDM)
Oversampling
Parameters
passive radar
Radar
Reference signals
Sidelobes
Signal resolution
super-resolution
Symbols
time delay estimation (TDE)
Time lag
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article studies the time delay resolution of radar, which refers to the ability to distinguish the delay of multiple adjacent targets. We propose a novel coarse-fine joint super-resolution time delay estimation (TDE) strategy for passive radar using orthogonal frequency division multiplexing (OFDM) signals. First, the modulation symbol domain (MSD) method is employed to estimate the coarse delay and Doppler parameters of the targets. The delay dimension is then oversampled by reference signal compensation (RSC) methods. To achieve super-resolution and high accuracy of the target delay, we present a method based on a 1-D fully convolutional network (1-D FCN). This method takes 1-D segments of coarse time delay parameters in the oversampling delay dimension as inputs, which reduces computational complexity. The simulation and experimental results show that the proposed method outputs low sidelobes, which greatly reduces the influence of strong targets on the delay estimation of adjacent weak targets. Besides, it can also easily realize fractional delay estimation.
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2023.3317955