Chapter 14
14 - 1 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• Moving Target Indication (MTI) radar: A delay line
canceller filter to isolate moving targets from
nonmoving background
- Ambiguous velocity
- Unambiguous range
• Pulsed Doppler radar: Doppler data are extracted
by the use of range gates and Doppler filters.
- Unambiguous velocity
- Unambiguous or ambiguous range
14 - 2 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• High-PRF: unambiguous Doppler frequency, highly
ambiguous range
- solve TX-RX coupling problem of CW system
- Range blind during TX time periods
• Low-PRF: unambiguous range, highly Doppler
frequency
- circumvents the TX-RX coupling
- Introduce Doppler blind zones (ground clutter)
• Medium-PRF: ambiguous Doppler frequency,
ambiguous range
- circumvents the TX-RX coupling
- Improve noise-limited detection relative
to low-PRF waveform
- Minimize the number of introduced blind
zones relative to low-PRF system.
14 - 3 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• Range: range is obtained from transmit-to-receive
pulse delay T
- ,
• Range Resolution: Pulse width must be shorter
than the propagation time from target 1 to target 2
and back
• Unambiguous range
, T: pulse repetition interval (PRI)
- There are ways to get around this by using a
staggered PRI (Multi-PRF)
2R cT = R ¬ ct 2 =
1µs 150m ÷ 1ns 15cm ÷
2 R
2
R
1
– ( ) ct = t ¬ 2 R
2
R
1
– ( ) c =
R A ct 2 =
R
unamb
cT 2 =
R
ct
2
------- =
Transmit
pulse
t
Target 1
Return
Target 2
Return
R
1
ct
1
2 =
R
2
ct
2
2 =
Transmit
pulse
t
Combined returned
from target 1 and 2
T
Unambiguous Range (R < cT/2)
Ambiguous Range (R > cT/2)
R
ct'
2
-------- =
R A
14 - 4 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Pulsed Doppler Power Spectrum
f
d
2V
c
ì
---------- | cos =
: angle between the platform velocityand the
line of sight (LOS)
|
14 - 5 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
- No reference signal
- TX phase is reserved
- detection of moving target by
suppressing fixed targets
14 - 6 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Range gate
1
2 3 4
switch sampling
Sampling
Analog
Digital
FFT
(filter bank)
Doppler Information
Range Information
14 - 7 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Power Spectrum Density (Pulsed)
• As the antenna scans, the beam dwell time is finite.
• : interpulse period; : pulse period T
i
t
p
N+1 = 5
14 - 8 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Power Spectrum Density (Pulsed) 2
• A five-burst waveform: the return from a scatter at a
slant range .
- contains four pulse samples
- contains only one pulse sample
• The shape of both the spectrum and ambiguity function
for is important determine performance of MTI
R
T
R
T
R
AMB
±
R
T
R
AMB
+
t ÷
N=4
N = 5
N+1=5
N+1=4
N+1=4
N+1=1
14 - 9 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Power Spectrum Density (Pulsed RF)
14 - 10 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• Until now, we have not said a great deal about
filtering of the return signal except to say that
matched filtering is desirable and for
• In some cases, we need a better idea about
bandwidth for estimation S/N ratio.
• Recall that we previously developed a radar
equation of the form
previously we consider this to be a single pulse.
• Integration of pulses: Depending on scan rate &
PRF, we may receive more than 1 pulse from a
target. We can use that our advantage
- : number of pulses for
integration during dwelling time.
- : beamwidth, : antenna scan rate
- : PRF
B
IF
1 t =
R
max
P
t
G
2
ì
2
o
4t ( )
3
kTBFL S
o
N
o
( )
min
-----------------------------------------------------------------
1 4 /
=
n
B
u
B
u
s
·
( ) f
P
· =
u
B
u
s
·
f
P
u
B
u
s
·
• An example: If , ,
PRF=300Hz
u
s
·
5rpm = u
B
1.5° =
u
s
·
5rpm round/per min. ( ) =
5 360 × 1 60 ( ) = 30° sec =
n
B
1.5
30
----------
\ .
| |
300 × 15 pulses = =
14 - 11 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Pulse Integration
• Two techniques
- Predetection Integration
- Postdetection Integration
• Predetection Integration is coherent but
somewhat more difficult to implement than
postdetection
• Postdetection is incoherent but some
improvement in (S
o
/N
o
) can be obtained.
IF
Envelope
detection
Video
Amp.
Predetection Postdetection
coherent
integration
signal
Noise
P
signal
nv ( )
2
·
n
2
·
P
signal
nP
n in a pulse ( )
·
Predetection
Postdetection
14 - 12 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Pulse Integration
• Recall we were developing alternate expression for the
- : single pulse S/N required for
prespecification .
- :efficiency factor; n: # of pulses integrated
- Note that for a pulse radar is a peak power, we can
also express in terms of average power
, where : Pulse width, :
PRI; : duty cycle.
- ; : Energy per pulse
R
max
P
t
G
2
ì
2
onE
i
n ( )
4t ( )
3
kTBFL S
o
N
o
( )
min
-----------------------------------------------------------------
1 4 /
=
S
o
N
o
( )
min
P
FA
E
i
n ( )
P
t
P
avg
P
t
t T ( ) tP
t
f
p
= = t T
t T
P
t
P
avg
tf
p
= Et P
avg
f
p
=
R
max
P
av
G
2
ì
2
onE
i
n ( )
4t ( )
3
kT Bt ( )FL S
o
N
o
( )
min
f
p
-------------------------------------------------------------------------------
1 4 /
=
EtG
2
ì
2
onE
i
n ( )
4t ( )
3
kT Bt ( )FL S
o
N
o
( )
min
------------------------------------------------------------------------
1 4 /
=
• We can express
- for ideal predetection ;
- : effective #
pulses integrated
• Example: , ,
find . If 1000
pulses are integrated (postdetection
square law)

S N ( )
min
S
o
N
o
( )
min
nE
i
n ( ) ( ) =
E
i
n ( ) 1 =
n
1 2 / –
E
i
n ( ) 1 s s
I
i
n ( ) nE
i
n ( ) =
P
FA
10
12 –
= P
D
0.9 =
S
o
N
o
( )
min
15.8dB ~
S N ( )
min
S
o
N
o
( )
min
nE
i
n ( ) ( ) =
15.8 10 130 ( ) log – = 5.34 – dB =
R
max
P
t
G
2
ì
2
o
4t ( )
3
kTBFL S N ( )
min
-----------------------------------------------------------
1 4 /
=
P
t
G
2
ì
2
onE
i
n ( )
4t ( )
3
kTBFL S
o
N
o
( )
min
-----------------------------------------------------------------
1 4 /
=
14 - 13 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
- No reference signal used by the receiver is phase
coherent to the output phase of the transmitter.
- A free-running pulsed transmitter
- Automatic Frequency Control (AFC) Local OSC is
made to track the transmitter frequency
- IF signal is bandpass filtered and amplified by IF
amplifier
- Square law (noncoherent) detection noncoherent
integrated signal processor CFAR
÷
÷
÷
frequency coherent
Bandpass Filtered
• Problems encountered in detecting
small-RCS in expected background
clutter environments
- WB Filter high noise
- Radar designer is left with only a few
techniques to minimize the
performance limits imposed by return
from background clutter.
- constrain parameters: operating
frequency, maximum permitted
antenna dimension dimensions...
÷
to TX frequency
14 - 14 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• The phase of TX waveform is preserved is a reference signal
• The use of STRALO and COHO reference signals to store the
phase of the later signal processing identifies the radar.
÷
No filter bank
• Relative complexity between coherent
and noncoherent systems
- If it were not for performance,
noncoherent configuration would be
used extensively used in search radar
applications.
exploitation of different Doppler shift to
isolate desired target responses from
large dominating (in amplitude)
background returns
- Relative motion between desired
target and its background
• Some techniques through which
noncoherent radar can be used to
accomplish Doppler shift-aided
detection of targets
14 - 15 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Pulsed Coherent MTI
• The detection of moving targets are improved by suppression of
fixed targets.
• This is expanded to incorporate Doppler processing as one
possible form of MTI implementation
• is deified as one uses simple band reject to reject the return
from fixed targets Enhanced detection of the moving target ÷
• Doppler filter
- A relative narrow bandwidth
clutter is rejected.
Doppler shift)
- Post-Doppler processing stage
Noncoherent integration
- Rejection notches in the
passband should be placed in
that are to be rejected and
should be as wide as required to
achieve the desired clutter
cancellation.
÷
Noncoherent
integration
14 - 16 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
MTI filter (Delay line canceller)
Two techniques are available for realizing MTI filter
• Delay line canceller
- Digital filter (Multipulse canceller)
- A real-time delay is equal to the PRI
- Digital implementations can provide the desired
passband with the flexibility of passband
programmability The preferred choice
• Range gate and filter
÷
14 - 17 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Range gate and filter
• output of phase detector (I or Q) is
provided as input to N sample-and-
hold circuits
• Each sample-and-hold circuit
lumped constant (active) bandpass
filter
• : lower corner frequency. :
higher corner frequency. : PRF
• Noncoherent MTI
- A-scope range video
presentations contain “butterflies”
at the slant range of the moving
target.
- Each butterfly is created by the
fluctuating amplitude of the sum
of the return from both the
background and the target
÷
f
L
f
H
f
R
14 - 18 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
Noncoherent MTI
• Noncoherent MTI
- A-scope range video presentations contain
“butterflies” at the slant range of the moving
target.
- Each butterfly is created by the fluctuating
amplitude of the sum of the return from both
the background and the target
A-scope
14 - 19 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
A pulsed Doppler radar exploits Doppler
shift to obtain velocity information from a
pulsed waveform
- High-PRF pulsed Doppler airbone radar
- Medium-PRF pulsed Doppler airbone
Two approaches
- Analog filtering
- Digital filtering
Analog filtering
- processing at IF
- Output of IF amplifier is gated
- gated output is then filtered by a bank of
contiguous narrowband (NB) filters
- Automatic detection and signal-sorting.
select so that overlap occurs at -3dB
3dB
BW
f
IF
14 - 20 Chapter 14: MTI and Pulsed Doppler Radar
Dr. Sheng-Chou Lin
• selection of digital approach for most modern pulsed
• Spectral leakage
- uniformly weighted set of N samples
- Other Doppler shift High sidelobes in the same
range cell
- aperture weighting is used to minimize the spectral
leakage.
÷
Sin nx ( ) x ( ) sin
÷
• Digital filtering (FFT)
- In-phase (I) and quadrature (Q) video
- Analog-to-digital conversion (A/D)
- N-point fast Fourier transform (FFT)
- N contiguous filter passbands (span
from zero to PRF)
- The number of range cells (m) and
Doppler cells (N) increases the
amount of hardware grows
dramatically.
- Modern digital technology leads to
hardware efficient realizations
• N-point FFT which produces a finite
impulse response filter is the desired
matched-filter for a finite sequence
(step scan algorithms)
• NB filter (infinite impulse response) in
the analog approach can be only an
approximation of the desired matched-
filter.
÷