MTI AND PUI.

SE DOPPLER RADAR
\y
threshold from the 16 range cells and a clutter threshold from the clutter map arc calculated for
filters 2 and 8 adjacent to the zero-velocity filter, and the larger of the two is used as the
threshold.
Tlie advantage of using two prfs to detect targets in rain is illustrated by Fig. 4.28. The
prf's differ by about 20 percent. A typical rain spectrum with a nonzero average velocity is
depicted on the bottom line. (Precipitation at S band might typically have a spectral width of
about 25 to 30 knots centered anywhere from -60 to +60 knots, depending on the wind
conditions and the antenna pointing. 42) The narrow spectrum of the moving aircraft is at the
right of the figure. Because of foldover it is shown as occupying filters 6 and 7 on prf-1, and
fillers 7 and 8 on prf-2. With prf-2, the aircraft velocity is shown competing with the rain
clutter; but with prf-1 it appears in one filter (No. 6) without any rain clutter). Thus, by using
I wo prfs which are alternated every 10 pulses (one-half beamwidth), aircraft targets will usually
appear in at least one filler free from rain except for a small region (approximately 30 knots
wide) when the target's radial velocity is exactly that of the rain.
Interference from other radars might appear as one large return among the ten returns
normally processed In the MTD, an interference eliminator compares the magnitude of each of
the It ) pulses against the average magnitude of the ten. Ifany pulse is greater than five times t h e
average, all information from that range-azimuth cell is discarded. A saturation detector detects
whether any of the 10 pulses within a processing interval saturates the A/D converter and if it
docs, the entire 10 pulses are discarded.
The output of the MTD is a hit report which contains the azimuth, range, and amplitude of
the target return as well as the filter number and prf. On a particular scan, a large aircraft might
be reported from more than one doppler filter, from several coherent processing intervals, and
from adjacent range gates. As many as 20 hit-reports might be generated by a single large
target.113 A post processor groups together all reports which appear to originate from the same
target and interpolates to find the best azimuth, range, amplitude, and radial velocity. The
target amplitude and doppler are used to eliminate small cross section and low-speed angel
echoes before the target reports are delivered to the automatic tracking circuits. The tracking
circuits also eliminate false hit-reports which do not form logical tracks. The output of the
automatic tracker is what is displayed on the scope. Since the MTD processor eliminates a large
amount of the clutter and has a low false-delection rate, its output can be reliably icmoted v i a
narrow bandwidth telephone circuits.

4.8 LIMITATIONS TO MTI PERFORMANCE

The improvement in signal-to-clutter ratio of an MTI is affected by factors other than the
design of the doppler signal processor. Instabilities of the transmitter and receiver, physical
motions of the clutter, the finite time on target (or scanning modulation), and limiting in the
receiver can all detract from the performance of an MTI radar. Before discussing these effects,
some definitions will be stated.

MTI improvement factor. The signal-to-clutter ratio at the output of the MTI system divided by
the signal-to-clutter ratio at the input, averaged uniformly over all target radial veloci; ties of
interest.
Siihclutter visibility. The ratio by which the target echo power may be weaker than the
coincident clutter echo power and still be detected with specified detection and
false-alarm probabilities All target radial velocities are assumed equally likely. A
subclutter visibility of, for example, 30 dB implies that a moving target can be detected
in the presence of clutter even though the clutter echo power is 1000 times the target
echo
 INTRODUCTION TO RADAR SYSTEMS

of the doppler filter-bank. The fast Fourier transform algorithm is used to implement the
doppler filter-bank. Since the first two pulses of a three-pulse cancelcr are meaningless, only
the last eight of the ten pulses output from thecanceler are passed to the filter-bank. Following
the filter-bank, weighting is applied in the frequency domain to reduce the filter sidelobes.
Although unweighted filters with -13.2-dB sidelobes might be satisfactory for most ground
clutter, the frequency spectrum of rain and other windblown clutter often requires lower
sidelobe levels. The magnitude operation forms ( I 2 + Q1)"1. Separate thresholds are applied to
each filter. The thresholds for the nonzero-velocity resolution cells are established by
summing the detected outputs of the signals in the same velocity filter in 16 range cells, eight on
either side of the cell of interest. Thus, each filter output is averaged over one mile in range to
establish the statistical mean level of nonzero-velocity clutter (such as rain) or noise. The filter
thresholds are determined by multiplying the mean levels by an appropriate constant to obtain
the desired false-alarm probability. This application of an adaptive threshold to each doppler
filter at each range cell provides a constant false-alarm rate (CFAR) and results in subweather
visibility in that an aircraft with a radial velocity sufficiently different from the rain so as to fall
into another filter can be seen even if the aircraft echo is substantially less than the weather echo.
A digital clutter map is generated which establishes the thresholds for the zero-velocity
cells. The map is implemented with one word for each of the 365,000 range-azimuth cells. The
original MTD stored the map on a magnetic disc memory. The purpose of the zero-velocity
filter is to recover the clutter signal eliminated by the MT1 delay-line canceler and to use this
signal as a means for detecting targets on-crossing trajectories with zero velocities that would
normally be lost in the usual MT1. Only targets larger than the clutter would be so detected. In
each cell of the ground clutter map is stored the average value of the output of the
zero-velocity filter for the past eight scans (32 seconds). On each scan, one-eighth of the output
of the zero-velocity filter is added to seven-eighths of the value stored in the map. Thus, the map
is built up in a recursive manner. About 10 to 20 scans are required to establish steady-clutter
vMues. As rain moves into the area, or as propagation conditions result in changing
ground-clutter levels, the clutter map changes accordingly. The values stored in the map are
multiplied by an appropriate constant to establish the threshold for zero-relative-velocity
targets. This eliminates the usual MTI blind speed at zero radial velocity and permits the
detection of crossing targets in clulter if the target cross section is sufficiently large.
Thus, in Fig. 4.28 the zero-velocity filter (No. 1) threshold is determined by the output of
the clutter map. The thresholds for filters 3 through 7 are obtained from the mean level of ihe
signals in the 16 range cells centered around the range cell of interest.
Both a mean-level 12 3 4
Filter number b 6 i7

prf -1
2WWWY1
prf-2

Aircraft True Figure 4.28 Detection of aircraft in

J^\ Rain
closes aircraft rain using two prf's with a doppler
-velocity
niter bank, illustrating Ihe efleci of
0 Radial doppler foldover. [From Muehe.43
velocity Courtesy IEEE.)