University of Boumerdes / Institute of Electrical and Electronic Engineering
Radar and Satellite Communication Systems (EE 515) - M03 / Telecom
EE515 : LAB 4 : Moving Target Indicator (MTI) radar
Because the small moving targets are hidden by the heavily cluttered environments, they can
not be detected in the time domain. Instead, the signal can be detected in the frequency domain
using Doppler frequency analysis.
To do this, return data must be collected and processed using two dimensional (2D) signal analysis
for both target speed and distance. 2D signal processing is required for moving target indicator (MTI)
The basic idea behind the MTI is to filter the clutter at a DC or very near DC, while keeping
the other spectrum region flat. As shown in Figure 1, a three-pulse (double, second-order) canceller can
be formed by cascading two first-order sections using a transfer function :
H(z) = 1 – 2z–1 +1 z–2
Figure 1. Moving target indicator A) with a double delay line canceller
B) Its equivalent three-pulse cancellers.
1) Show that : | |
Matlab code to filter a signal x using double delay line canceller:
h=[1 -2 1] ;
y = filter(h,1,x,[ ],2) ;
% To plot the transfer function Hf versus a frequency vector freq :
freq = linspace(0,M*PRF,nt); %M is the number of the PRFs and nt is the number of
%samples
Hf = freqz(h,1,freq,PRF);
plot(freq, abs(Hf)); grid on; xlabel('Doppler Frequency (Hz)'); ylabel('Magnitude');
title('Frequency Response of the Three-Pulse Canceller');
1
Teacher: I. HAMOUDI
University of Boumerdes / Institute of Electrical and Electronic Engineering
Radar and Satellite Communication Systems (EE 515) - M03 / Telecom
2) Generating the input signal samples Xmn :
You can generate n complex envelopes Xmn = (I [m,n ], Q [m,n]) corresponding to the mth PRI (Pulse
Repetition Interval T) (i.e : (m-1)T ≤ t mT).
A matlab function [t,tm,I,Q,bt,fs]=MTILAB(a,r,v,m), similar to the one used to generate Pulse
Doppler echo signals except that only one sample is taken for each range gate. So : fs = 1/tp (Only
video signal is modeled : no need to the carrier component!).
Figure 2. Data Collection for MTI Processing example.
a) Write an m-file to generate an M N matrix as shown in figure 2. Using the MTILAB function,
you should obtain samples Xmn = (I [m,n ], Q [m,n]) where m = 1 : M and n = 1 : N.
b) Since if X is a matrix, filter operates on the columns of X, you can use y = filter(h,1,X,[ ],2).
c) Apply the FFT for each column of length M. You can also use the function :
Periodogram. (see help topic fom more details). A Doppler spectrum is to be obtained for each
range gate.
2
Teacher: I. HAMOUDI
