Page 261 - Radar Technology Encyclopedia

251 loss, collapsing loss, crossover

replaced with t = 1/B, and collapsing loss will not depend on
n
the pulse compression ratio but only on the ability of the sub- 10
sequent processing to retain the resulting resolution. In pulsed
doppler systems, the effect of mismatched range gate or 8
receiver bandwidth appears directly as a matching loss, rather
than as the lower collapsing loss applicable to systems using 6
noncoherent integration. DKB CFAR loss L g (dB)
Ref.: Blake (1980), p. 49; Barton (1988), pp. 77–79. 4
Table L9 2
Equations for Collapsing Ratio P = 10 -x
fa
0
Cases for which P /r remains constant: 0.5 CFAR ratio = x/m e 1.0 1.5
fa
(a) Restricted CRT sweep speed,
d + st Figure L23 Universal curve for CFAR loss in single-hit detec-
s, where d is spot diameter and t r = --------------
st tion, for steady or Rayleigh target (after Gregers-Hansen).
is pulse width
The general equation for m is
(b) Restricted video bandwidth, e
1 B
B , where B is IF signal band- r = 1 + ------------ = 1 + --------- m + k
v
v
width 2B t 2B v m = -------------
e
k
1 +
(c) Collapsing of coordinates onto the display: where values of k are shown in Table L10, and m is the num-
ber of reference cells used to form the threshold. DKB
2D r /c is time-delay interval per
2D r
display cell r = --------- Table L10
ct
Constants Determining Number of Effective CFAR Refer-
t
w e v is elevation scan sector, q e
w t ence Samples
e v
is elevation beamwidth r = ----------
q
e
Square-law detector k = 1
t
w a v is elevation scan sector, q a w t Linear envelope detector k = 0.09
a v
is elevation beamwidth r = -----------
q
a Log detector k = 0.65
Cases for which P remains constant: Greatest-of CFAR, square-law detector k = 0.37
fa
Greatest-of CFAR, linear envelope k = 0.5
(d) Excessive IF bandwidth, B >
1
1/t , followed by matched video r = 1 + ------ detector
Bt
Greatest-of CFAR, log detector k = 1.26
(e) Receiver outputs mixed at r = m
Hard-limiting CFAR with Dicke fix
video, m is number of receivers B w
(add 1-dB limiter loss) m = ------- – 1
e B
(f) IF filter followed by gate of n
1 t g
width t and by video integra- r = ------ + ----- Hard-limiting dispersive or pulse-com- m = Bt 1–
g
tion Bt t e
pression CFAR (add 1-dB limiter loss)
Constant-false-alarm-rate (CFAR) loss is the result of
Ref.: Gregers-Hansen, W., “Constant False-Alarm Rate Processing in Search
using a CFAR circuit to establish the detection threshold, Radars,” IEE Intl. Radar Conf. Radar-73, Oct. 1973, pp. 325–332; Bar-
rather than using a fixed threshold based on exact knowledge ton (1988), pp. 88–92.
of the noise (or interference) level and statistics. It is defined
Conversion loss refers to the loss in signal-to-noise ratio S/N
as the increase in signal power required to achieve a given
in passing through the mixer in a superheterodyne receiver.
detection performance using the CFAR process on noise (or
The loss does not appear directly in the radar equation but is
interference) of unknown level and with Rayleigh distributed,
included in the receiver noise factor. DKB
relative to that required for a fixed threshold on known level.
Ref.: Van Voorhis (1948), p. 18.
For a class of conventional, cell-averaging CFAR, the loss is
given by Fig. L23, where the parameter x is the negative Crossover loss results from the squinting of the beam axis
- 6
exponent of P (e.g., x = 6 for P = 10 ), and the CFAR from the tracking axis in conical scan or other types of lobe-
fa
fa
ratio is x/m , where m is the effective number of reference switching trackers. It appears as a component of pattern-prop-
e
e
samples. agation factor in the radar equation, and directly in the equa-