RRN-ATR-Net dataset

Citation Author(s):
zhuofeng
Wu
Guangdong University of Technology
Zhenyu
Liu
Guangdong University of Technology
Submitted by:
zhuo wu
Last updated:
Wed, 02/07/2024 - 21:43
DOI:
10.21227/2q85-nz22
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

RRN-ATR-Net is a radar echo data set containing the micro-Doppler signatures (m-Ds) of different aerial targets, with a total sample size of 1200. It is available for researchers interested in this field but having difficulty collecting data. The data set is acquired using Texas Instruments' AWR1642BOOST radar sensor and the DCA1000EVM high-speed data acquisition card. The target types subject to acquisition include quadrotor (Phantom3s), fixed-wing (Cessna182), helicopter (T-REX450), and bionic bird (Gogo Bird1020). These data are collected in the anechoic chamber to eliminate unnecessary interference. The acquisition method is to place the target statically in a fixed position within a range of 1.5 to 6 meters away from the radar, allowing only the blade or wing to move. 300 signal samples are collected for each type.

Instructions: 

Use the following matlab code to read and parse the data, as an example of a short-time Fourier transform of the signal:
%%
clear all;close all;clc;
num = 1;
Rawdata = load(strcat('fixed_wing\',num2str(num),'.mat'));
rawdata = Rawdata.target_bin_part;
nwin = 64;
wind = hamming(nwin);
nlap = nwin-2;
nfft = 128;
STFT_imag=spectrogram(rawdata,wind,nlap,nfft,10000,'centered','yaxis');
x = [0 0.4096];
y = [-5000 5000];
figure
imagesc(x,y,db(abs(STFT_imag)))
axis xy;
colormap('jet')
%%

Funding Agency: 
Guangdong Basic and Applied Basic Research Foundation under Grant;Guangzhou Key Research and Development Project under Grant; Foshan Social Science and Technology Research Project;Guangdong Province Social Development Science and Technology Collaborative
Grant Number: 
2023A1515012873;2023B01J0011;2220001018608;2023A1111120003