Note: Descriptions are shown in the official language in which they were submitted.
~l~223~a~
This invention relates -to a Doppler tracking processor
and time of closest approach detector for use in a semi-active
or active continuous wave (CW) radar system. An example of
tha-t is -the application of the invention to CW semi-active
guided missiles. So, before presenting the invention, its
context will be introduced in the following text.
With guided missiles used against moving targets, such
as aircraft, it is necessary to detonate the missile at the time
of closest approach (TCA) to the target because of the
unliklihood of the missile scoring a direct hi-t on the target.
The problem consists in detecting accurately the time of closest
approach (TCA) to the target by processing the Doppler signal in
the seeker of a CW radar semi-ac-tive missile when -the TCA detector
has been armed by other missile subsystems, such as a safety-
arming (S&A) device which maintains the missile warhea~ in an
unarmed condition until the missile has been.intentionally
launched and has traveled a safe distance from the launching
aircraft.
Guidance and control of missiles are based on different
techniques of major importance (beam rider guidance, command
guidance, preset guidance and homing guidance). One of them,
the homing guidance, generates steering signals from information
received at the missile seeker from the target.
Homing guidance systems are of three types: active,
semi-active and passive systems. An active homing system beams
a signal at the target and generates steering commands from the
reflected signal. This homing device on the missile reveals
the presence of -the missile to the target. In a semi-active
3~3
homing system, a remote transmitterl located outside the missile
(on an airplane, a ship or any other equipment with an appropriate
antenna to illuminate the scenario), bounces signals off the
target to the missile. The remote illuminating transmitter
reveals i-ts position and reveals the existence of a missile
to the target. A passive homing system receives radiated
energy emanating from the target. The passive homing device
does not reveal the position or the existence of the missile
to the target by emitting any energy (as a target, a missile
radiates energy...). In these three systems radiated
energy may be radio, optical (IR, UV and visible light) and
sound signals.
The realisation of the present invention is concerned
with a C~ radar "semi-active" misslle. Even if -the missile
has only a receiver, it is called "semi-active" since it
requires an external illumination to guide on the target.
The energy transmitted from the illuminator transmitter
external to the missile is received by rear and front
antennas on the missile. The rear signal is the sample of
the radiated energy from the transmitter used as a reference.
The front signal is the energy from the illuminator that has
been reflected by the target. These microwave signals are
mixed and low-pass filtered to deliver a signal at lower
frequency. This signal is -termed the video Doppler since
it is not necessarily a narrow-band signal and because it
includes relative missile-target information with noise and
interference.
~L%2;~311~
The target signal i5 the radi.o energy returned to
a radar by a -target, also known as echo signal or video
signal. Video is pertaining to the demodu].ated radar received
output -that is applied to a radar ind.ica-tor. In low-altitude
environments, sea clu-tter is often dominant. Wi-th targets
composed of several reflectors, forward scatter from reflectors
already passed, and backscat-ter from reflectors yet to be
passed are sometimes importan-t contributors to the Doppler.
It is known to use a fractional Doppler gate (FDG) to
process -the Doppler signal to activa-te the warhead detonatio
properly at TCA. In that prior arrangement the TCA Doppler
detection is accomplished by a tunable band-pass filter ~constant
bandwidth), adjusted at a fractional Doppler frequency, preceded
by a fast automatic gain control unit to normalize the entire
signal at its inpu-t, and followed by an envelope detector and a
high-pass filter (differentiator) in such a way as to detect
only amplitude increases greater than a fixed threshold of the
signal at the output of the band-pass filter. This TCA Doppler
detector indicates when the doppler energy has rolled off to a
fractional frequency within the constant bandwidth of the
band-pass. The time-frequency plane is related to the TCA in
the time~distance plane, by the physi.cal and the geometrical
charac-teri~ation of the intercept (-this will be explained later).
The FDG center frequency is tuned at about one-half the pre-
intercept Doppler frequency, 0.5fdO.
Assuming that all the other arming subsystems operate
properly, the present invention subsystem achieves a better
signal processing than by the FDG sinceo
2~23(~
1. It is not sensitive to amplitu~evariation, as it reacts
as a frequency discriminator.
2. It has to detect the Doppler frequency decrease from its
pre-intercept value (fdO) down to a fraction of fd which
corresponds to the TCA. This avoids premature TCA detection.
The present invention involves tracking the Doppler
signal before the intercept and during the Doppler roll-off in
such a way as to detect timely any frequency of the roll-off up
to one half of the pre-intercept frequency in the time-frequency
plane. Thus, using the relation between the time-frequency plane
and the time-distance plane, this processor can deliver an appro-
priate and accurate TCA detection when all the other sub-systems
have enabled it to output.
In accordance with a broad aspect of the invention there is
provided a Doppler tracking processor and time of closest approach
detector system compris;~ng first means for deriving a reference
signal proportional to a predetermined fraction of the frequency
of a Doppler signal, second means for deriving a second signal
~hich is dyn3mic and ;~s proportional to-the i~stantaneou-s-~oppler
frequency of the Doppler signal during Doppler roll-off, and
comparator means for comparing said second signal and said re~r-
ence signall said comparator producing an output signal if sai'~
second signal drops below said reference signal.
The prior art and the present invention will now be
further described in conjunction with the accompanying drawings,
in which:
2~
Figure 1 is a time history of the spectral content of
the Doppler signal in relation to the sequence of events that
leads to fuzin~,
Figure 2 is a diagram which illustrates the relation
between the time-fre~uency plane, part (a), and the t.ime~distance
plane, part (b), of an intercept when -the radial to transverse
speed ratio is high,
Figure 3 is a block diagram of a prior ar-t TCA Doppler
detector using a frac-tional Doppler gate,
Figure 4 is a more detailed block diagram of a
fractional Dopp].er gate,
Figure 5 is a block diagram of a Doppler tracking
processor and TCA detector according to the present invention,
Figures 6A and 6B, which fit together as shown in
Figure 6C, comprise a schematic diagram of the Doppler processor
and TCA detector according to the invention, and
Figure 7 is a diagram illustrating the difference
between dynamic detection an~ static detection of TC~ used by
the tracking sys-tem of the invention and the fractional Doppler
gate, respectively.
Figure 2 is a diagram which illustra-tes the relation
between the time-frequency plane, part (a), and the -time-distance
plane, part (b), of an intercept (upper portion of part a) shows
missile-target at TCA) when the radial to transverse speed
ratio is high.
The time of closest approach (TCA) is defined as the
time where the missile-target distance (ranye) i5 at its
minimum value (whichever value it is). This minimum value, -the
2~
miss distance (dm), occurs at TCA as illustra-ted by the range
curve of Figure 2. Also TCA is the time corresponding to -the
closest point of an approach (CPA).
TCA: -the -time reference that corresponds to the CPA
CPA: the spacial point of the missile trajectory o~ an
engagement where the missile-targe-t distance is
at the minimum value.
dm: -the minimum value of the missile~target distance~
The basic fuze presently known is the fractional Doppler
gate (FDG) whose operation depends on the Doppler behaviour of
an intercept. The FDG is an electronic circuit which receives
energy from the target at the closest-point-of-approach (CPA),
converts this energy to a high energy pulse, and supplies the
pulse to the warhead detonator.
~hen the closing on relative velocity Vc is almost
radial, the,Doppler frequency (fd) can be e~pressed in a simple
equation:
d c (1 + cos e) = -
~r
'~ (1 + cos e) ~ where ~0 = TX (transmit)
wavelengh
where
Vr -,Vector representing relative radial velocity between
the target and missile (since Vc ~ Vr, as for Figure 2)
fO = illuminator transmitter frequency of -the CW carrier
c = ~elocity of propagation
= Angle between the relative velocity vector and the
line-of-sight from the missile to target
36;~
When the missile ls at a position alGng the trajectory
where the range from missile to target is greater than ten times
the miss distance, the angle ~ can be treated as æero and
f = Vrfo = 2Vr
c o
This ~alue is fdO, the pre-intercept Doppler frequency.
When the missile is abeam of -the target (CPA) the angle ~ is 90
degrees and
V f
r o
f = --
It follows that the Doppler frequency at CPA is
approximately one-half the Doppler frequency of the guidance
portion of the flight just before the intercept region (when
the range from the missile to the target is ten times the miss
distance) .
The FDG can be regarded as a narrow-band filter tuned
to receive energy at the half-Doppler frequency. This function is
accomplished by mixing the Doppler signal with a local oscilla-tor
reference signal and amplifying the resulting difference
frequency in a tuned amplifier.
Figure 3 is a block diagram of a complete Doppler
processing system of the prior art type, and will now be
briefly described.
The video input (video Doppler) is applied to a speedgate
10. ~s the missile is launched and accelerates, the missile
longitudinal accelerometer (not shown) programs the missile
-- 7 --
9~2~3~
speedgate 10 to correspond with the actual missile-target velocity
of closure. The output of speedgate 10 is detected at 12 and,
via an AFC loop 13, controls the frequency of a local oscillator
in the speedgate 10 so that the Doppler signa:L is cen-tered in the
speedgate pass-band intermediate frequency (IF). The frequency
of a local oscillator in the fractional doppler gate (FDG)
1~ is controlled by a voltage derived from the AFC of the speed-
gate. When the Doppler in the speedga-te begins the rapid shift
which occurs at intercept, the rate of change of -the Doppler
frequency exceeds the tracking capability of the speedgate.
This actuates a control circuit (within what is labelled
"Detector" at Figure 3) that keeps (holds) -the voltage that
tunes the FDG local oscillator to the desired frequency during
an intercept. Thus, the difference between the FDG pass-band
intermediate frequency and the FDG local oscillator frequency is
adjusted to be approximately one half the Doppl.er frequency
present in the speedgate during the guidance phase of -the flight.
And this difference is maintained so that the equivalen-t pass-
band center frequency is at approximately one half the previous
value of the Doppler (fdO) when the speedgate loses the signal
due to the rapid change of frequency of -the Doppler signal
during the intercept phase of the flight.
The TCA Doppler detection is then accomplished by a
tunable band-pass filter 17 (constant bandwidth with two states,
selectable as function of Doppler frequency range) in FDG 14,
adjusted at a fractional Doppler frequency, preceded by a fast
~2223C~8
automatic gain control unit 15 to normalize the entire signal
at its input, and followed b~ an envelope detector 16 and a
high~pass filter (differentiator) 18 in such a way as to detect
only amplitude increases greater than a fixed threshold of the
signal at the output of the band-pass filter 17 in FDG 14. This
TCA Doppler detector indicates when the Doppler energy has
rolled off to a fractional frequency within the constant
bandwid-th (for a specific range of fdO) of the band-pass.
The center frequency of FDG 14 is tuned at about one-half
the pre-intercept Doppler frequency, 0.5fdO.
The output of high-pass filter 18 triggers what is
equivalent to a monostable 20 which provides one input -to
fuze switch 22. Fuze switch 22 (an AND gate) provides a
fuzing pulse to detonate the warhead (not shown) provided
the other fuze inputs are present. These other inputs are
Range Arming (RA), Intercept Arm (IA) and Guard Gate
Activation (GGA) as shown by the sequence of events of Figure
l(b). It is not proposed to d~scuss the components of
s
Figure 3 relating to these other fuzing inputs ~ they are
not particularly germane to the present discussion.
The fractional Doppler gate (FDG) 14 is shown in more
detail in Figure 4 and is seen to comprise a fast AGC 15 to
which the video is applied as an input. A voltage controlled
oscillator (VCO) 24 receives the vol-tage from AFC loop 13
(see Figure 3) of the speed gate 10. The outputs of fast AGC 15
and VCO 24 are fed to a mixer 26, the output of which feeds an
IF filter 28. The IF filter 28 feeds a demodulator (low-pass
3~a3
envelope detector) 16 which is followed by high-pass filter
(differentiator) 18, and comparator and holding circuit 30
(or the comparator 20 in Figure 3), the output of which is
fed to -the AND gated ~u~e switch (22 in Figure 3).
The fractional Doppler gate (FDG) 14 is a selective
gate tuned at approximatively 0.5 fdO. The E'DG center frequency
is controlled by the output of the automatic frequency con-trol
(AFC) loop 13 of the speedga-te 10. The ratio be-tween the center
frequencies of -the FDG and the speedgate is approximatively se-t
.lO to one-half.
The mixer 26 with the VCO 2~ and the IF filter 28 are
equivalent to a voltage controlled band-pass fil-ter. The
center frequency of this band-pass fil-ter is voltage adjusted to
be at 0.5 fdO. The fast AGC at the input of the IF filter 28
normalizes the energy of the video signal. The high-pass filter
(differentiator) 1~ at the output of the demodulator removes the
mean value of the noise a-t 0.5 fdO prior -to intercep-t. The
threshold of the comparator 30 protec-ts the FDG against false
intercept detections.
At intercept, the ratio of the missile-target distance
to the illuminator-missile distance is very small. Then the
missile-target closing velocity can be factored in radial (parallel
to the CW propagation from the illuminator to -the missile and the
target) and transverse componentsO This forms an orthogonal
basis ~rom which any intercept can be modelled. For any given
radial to transverse velocity ratio, the invention could deliver
a pulse at the exact TCA of an intercept using the relation
between the time-frequency plane and the time-distance plane.
-- 10 --
23~
This model can be divided in two simple cases that are
intrinsic to it. For a radial velocity much smaller than the
transverse velocity, the ~CA occurs when the instantaneous
Doppler frequency is equal to zero. This is similar to the
Doppler roll-off of an active system even if the missile is
semi-active. For a transverse velocity much smaller than the
radial velocity, the TCA occurs when the instantaneous Doppler
frequency is at one-half its pre-intercep-t value. This case is
illustrated at Figure 2.
For this particular case, a TCA detector outputs a
pulse exactly at TCA when it is adjusted at one-half Doppler.
In the TCA detector according to the present invention, Figure 5,
a comparator 52 verifies continually the relation between
two voltages. One of them is the low-pass filtered (~-~ 0.3s)
output of the AFC of the missile speedgate. It is attenuated
by a factor of two by a fractional ratio converter 50 to
represent one-half of the preroll-off reference. The other
is the low-pass filtered ( ~~~3 ms) voltage of the frequency loop
of the tracking filter, i.e. the output of filter 51. This
2Q vol-tage represents the instantaneous Doppler frequency.
Then, before TCA -the comparator 52 output is at zero and at
TCA the comparator turns on rapidly (r ~ 1 ~s). This activates
the monostable 20 to produce a pulse of predetermined duration
which is applied to fuze switch 22.
In most all engagements -the missile-target velocity is
essentially radial. For a radial component larger than 60% of -the
magnitude of the closing velocity, the one-half Doppler occurs within
10~ of ~ u before TCA. At Figure 2 -the parameter ~ u~ the
-- 11 --
~2~3~
intercept time constant, is defined as the unit of time equal to
the ratio of the miss distance to the closing velocity.
A TCA detector can be used in a boost-glide missile.
During glide, the missile slows down and so cloes the Doppler
frequency tracked by the speedgateO The rate of change of the
Doppler frequency is of the order of a few ~z/ms. The FDG center
frequency follows this change at one-half rate. At intercept the
rate of change of the Doppler frequency is much higher, ~ the
order of a few kHz/ms. This ra-te of change is too large for
the AFC of -the speedgate. The energy drops off i-ts band-pass.
The intercept arm (IA, Figure 3) is then activated. The voltage
at the AFC of the speedgate, when the IA is activated, is
proportional to the frequency of the Doppler just before -the
roll-off. This is due to the fact that the time cons-tant of
the AFC is quite large ( r ~ 1OO ms) and that the IA time
constant is small ( ~ ~ 1 ms). The activation of the IA
forces the FDG to hold its previous value which is -therefore
the 0.5 fdO (one-half the pre-intercept Doppler frequency).
During the roll-off, the Doppler shifts toward the FDG
2Q center frequency. If the signal to noise ratio a-t the output of
the band-pass filter 17 of the FDG is larger than O dB, this in-
cominy signal generates an increase of the output level. The FDG is
activated when the rate of this increasing voltage is large
enough to build, at the differentia-tor 18 output, a voltage larger
-than 1.2 Y (-the threshold of comparator 30, Figure 4). The FDG
activation holds on for a short period of time (2 ms). If
during its ac-tivation the other gates (IA, RA, GGA, Figure 3) are
on, the fuze switch 22 turns on and a fuze pulse activa-tes
- 12 -
the squib of the war head.
Figure 5 is a block diagram of an arrangement according
to the inven-tion. Several components are identic~1 to those shown
in Figure 3, but the FDG 14 has been replaced by a Doppler
tracking processor and TCA detec-tor 40. The tracking fil-ter 41
of -this new TCA detector has -the ability to extract the
instantaneous value of the Doppler frequency of a roll-off. The
tunable band-pass filter 41 stays tuned to the instantaneous
Doppler frequency even during roll off. Thus, from this
instantaneous value, it is possible to estimate the value of the
miss distance at TCA or at any time of a roll-off~ This comes
from the fact that any intercept can be described by a model
with normalized parameters as illustrated in Figure 2. This
figure does not illustrate -the general model but illustra-tes a
simplified model and has been briefly discussed above.
Figure 5 illustrates the entire system including the
new TCA Doppler detector according to the invention. In this
arrangement, a Doppler processor and TCA detector 40, processes
the Doppler signal by a narrow band-pass filter 41 (constant Q
or proportional bandwidth) tuned automatically -to be frequenc~--
locked on the main Doppler signal by means of a fast frequency-
locked loop 4~. The loop 44 includes a Fast Automatic Gain
Control (E'AGC) unit 46 to match the band-pass filter 41
output signal to the input of a fast frequency to voltage
converter (F/V) 4~ that allows the tracking fil-ter 41 to
follow the rapid change in frequency of a Doppler rolling-
off signal. This sub-system uses standard low cost ICs like
Operational Amplifier, Operational Transconductance Amplifier
- 13 -
~2~23~
(OTA), Exclusive Or ~EXOR) and Monostable, as illustrated by
the schematic oE Figure 6.
The novel aspect of this invention concerns the dynamic
tracking of the Doppler roll-off as opposed to -the former or
present static detection of a band-pass filter waiting to capture
a transient signal at a fixed specific frac-tional frequency.
This system processes continuously the entire rolling-off
energy of the Doppler from the pre-intercept frequency (pre-
intercept time and space) up to the appropriate fractional
frequency in such a manner as to be more insensitive to the
interferences accompanying the rolling-off Doppler. This is done
by means of the narrow band frequency locked processor.
Response time to a frequency shifk_
The feedback frequency loop time constant of the
tracking filter 41 determines the maximum rate of Doppler frequency
shift of a roll-off that this system can follow before a loss of
lockO In this case, it is approximately equal to the F/V converter
time constant; namely 0.2 ms for the circuit of Figure 5.
Response time to an amplitude variation:
The Automatic Gain Control (AGC) 46 time constant of the
feedback loop 44 determines the maximum rate of an amplitude
variation within its dynamic range that this system can follow
before a loss of regulation at its output. The full-wave amplitude
detector allows b~ decrease~tne time constant of this AGC
amplifier to quite small values (500 ~s from 3.5 kHz to 140 kHz
for the circuit of Figure 5). During a Doppler roll-off, the
pass-band of -the voltage-controlled filter 41 will not be exactly
tuned to the instantaneous frequency to be tracked due to
-- 1~ --
~2~3~g~
the intrinsic time delay of the F/V converter 48. This frequency
offset will cause an amplitude decrease at the ou-tput of the
band-pass filter 41 that has to be corrected by the AGC. It is
de-termined that the AGC dynamic range mus-t be larger than the
sum of the attenuation caused by this frequency offset and the
maximum amplitude variation that could be e~pec-ted during a roll-
off (antenna side lobes...). In addition, the AGC -time cons-tant
must be smaller than the Doppler roll-off time constant (~ u)
Frequency dynamic_range:
The linear dynamic frequency tuning range is
limited by the physical nature of the components. The linear
dynam.ic range of the OTA Model CA 3080A (as indicated by the
manufacturer data sheet, RCA) is limited to 3 decades. The
range is further limited by the nonlinearities of the fast
loop and the stability aspect of the entire tracking unit
for frequency as high as 600 kHz, yielding a ratio of 1 to
40 between the minimum and maximum frequency (namely from
3.5 kHz to 140 kHz for the circuits of Figure 6).
Constant rela-tîve bandwidth
2Q The tunable band-pass filter 41 of this system has been
designed to exhibit a constant Q characteristic (the ra-tio of
the resonant frequency to the bandwidth is kept constant for
any frequency) to insure a proportional rejection of the
interferences for all Doppler frequencies within its -tracking
range. It could be modified to exhibit a constant bandwidth over
the whole frequency range.
- 15 -
Figure 7 is an illustration of the difference between
dynamic detection and static detection of TCA used by the trackiny
system of the invention and the fractional Doppler ga-te, respectively.
Magnitude frequency responses and magnitude of Doppler
spectrums are drawn at four different instants of the roll-
off to form a time history of the responses to the roll-off.
An explana~ion of the guard gate GG and the arming state, GGA
.
The GG channel includes a filter 60 tuned at a
higher frequency than the IF of speedgate 10. The GG (Guard Gate)
channel processes a signal derived from one of the IF stages
of the speedgate 10 (see Flgure 5). Thus, the GG exhibits
an equivalent pass-band center frequency lower than the
speedgate one. This has been illustrated in Figure 7 by
setting its equivalent center frequency at 0.8 fdO.
The GG output has to be enabled by the high level
arm ~HLA) from detector 12 before indicating a GGA.
When the frequency of the Doppler shift approaches -the
GG center frequency and when there is enough energy to ac-tivate
its detector 62, this GG output and the HLA are combined in an AND
gate 64 that triggers a monostable 65. The GG~ holding time
period is set to a value proportional to the larger ~u of
a roll-off that could be expected for any miss distance
within the lethality zone.
The HLA indicates that the signal tracked by the
speedgate prior to the intercept was of an adequately high-level
to reflect the impending intercept (-the signal increases as
the missile-target distance decreases).
- 16 -
23~)8
Also, the arming conditions as shown at Figure 1
(Safety and Range Arming, HLA, IA and GGA) have to be timely
satisfied to allow the activation of the FDG, as well as of -the
tracking system of the presen-t invention, to detonate -the warhead.
Referring to Figures 5 and 6, the voltage-controlled
band-pass tracking filter 41 consists of -two cascaded voltage-
controlled integrators 70, 71 arranged in a biquad structure (a
realisation of a biquadratic voltage transfer function, which
is the ratio of two second - order polynomials) with a
summing amplifier 74 to generate complex poles. In closed
loop, it operates like a very fast discriminator over -the 5-
140 kHz band: it is continuously tunable over -this band by
means of feedback loop 44 that senses its ou-tput and adjusts
its center frequency so as to keep the band-pass center frequency
in coincidence with the component of greatest amplitude in
-the Doppler signal. The loop includes a fast AGC amplifier
46 that matches the output of the band-pass filter 41 to the
input of a fast frequency-to-voltage converter 48 that
allows the frequency-tracking filter 41 to remain locked on
the target return even during the roll-off of the Doppler at
intercept.
The output of the F/V converter 48 is used to detect the
time of closest approach. The speedgate 10 produces a voltage
Vs proportional to the frequency fd that i-t tracks. This voltage
is applied to fractional ratio converter 50 where it is low-pass
filtered (by U9-A and B) ~`~= 220 ms), adapted (scaled) to
the F/V converter 48 output Vcl (14 kHz~V) and attenuated by
a factor of 0.6 in the fractional ratio converter 50 to
- 17 -
finally constitute the reference for firing which is applied
as one input to Doppler frequency comparator 52. The appropriate
time for fuzing is detected when the low pass filtered output
(Vc5 from filter 75) of the F/V converter 4~, applied to another
input of compara-tor 52, becomes smaller than or equal to this
reference, given the occurrence of fuze armingO The output
of comparator 52 drives a monostable circuit 20 to produce a
pulse of predetermined duration for fuze switch 22.
The band-pass fllter 41 has been designed with
operational transconductance amplifiers (OTA's) specifically
RCA CA-3080, although other types of OTA's could be used.
Adding and voltage-following operations are implemented with
operational amplifiers. This configuration allows the
resonant frequency of the filter to be tuned linearly within
the 5-140 kHz band. The quality factor Q and the gain of
the filter remain constant over the band.
The feedback loop 44 senses the output Eo of the
band pass filter 41 and produces the control voltage Vcl that moves
the resonant frequency of the filter 41 toward coincidence with
the frequency that dominates Eo. This is done as follows:
The fast AGC amplifier 46 amplifies Eo to a reference
level compatible with the driving signal of the fre~uency-to-
voltage converter 4~ that follows it. This AGC amplifier
uses an OTA (U5) as a variable gain amplifier (U5, U6-A) to
main-tain the output at a fixed amplitude. The ou-tput of the
OTA is full-wave rectified, fil-tered and compared (U6-D) -to a
reference amplitude to produce an error voltage which, after
- 18 -
23~
ha~ing been filtered (U6-C), controls the gain of the OTA.
The full-wave rectifier allows decreasing the time cons-tant
of the AGC amplifier to as low as l ms while its amplitude
. ~r~
dynamic range is ~ the order of 50 dB.
The F/V converter 4~ transforms -the sa-turated
waveform (output of U6-B) presented to it into a train of 50 ms
duration pulses whose repetition rate is proportional to the
frequency of the input waveform. This is done using a
combination of logical exclusive OR gates (U7~ and delay
operations. The pulses are averaged through a fast-charge
and slow-discharge circuit ~U8-B) with time constants set to
2 ~s and 220 ~s respectively. The output voltage (Vcl) is
adjusted -to give 14.0 kHz/V and an operational amplifier (TJ8-~)
transforms (and filters) it into the voltage Vc2 which
controls the resonant frequency of the band-pass filter 41.
When Vc2 is varied from -15 to 12 ~, -the resonant frequency
of the band pass filter 41 is moved from 5 -to 140 kHz.
The novel aspect in the implementation of the Doppler
fuze processor as a replacement of the fractional Doppler gate
in a missile stems from its capability to dynamically -track the
target return during roll-off as opposed to s-tatic detection
of the rolling-off doppler at a fixed fractional Doppler frequency.
The processor tracks the target return continually after fuze
arming while the speedgate coasts on its pre-intercept frequency.
The accurate detec-tion of the time of closest approach
depends on the following:
1) the tunable bandpass filter ~l mus-t exhibit a cons-tant
frequency-to-bandwidth ratio (quality factor Q) to insure a
-- 19 --
~22~3CI~
proportional rejection of the interferences for all the
frequencies wlthin its tracking range. The value of Q determines
the immunity of the processor to noise and interference signals.
2) the value of Q and the speed oE the feedback loop
determine the capacity of the processor to react rapidly -to
changes in the frequency conten-t of i-ts input: such as those
characterizing the Doppler roll-off. For this matter, the
response time of the feedback loop is dependant on the response
time of the F/V converter 48 and AGC amplifier 46.
3) the time constant of the AGC amplifier of the feedback
loop determines the maximum rate with which a variation in
the amplitude of the input signal can be followed by the processor.
Experimental investigation has demonstrated that the
operation of the processor is optimum when the quality factor
Q is set to about 15, when the time constant of the AGC amplifier
is 1 ms, and when the speed of the feedback loop is around 200 ~s.
This setting of the parameters takes into account that in some
instances the Doppler roll-off may take place in a time interval
as short as 2 ms.
The detailed schema-tic (Figure 6) includes an adaptive
equalizer (tunable high~pass filter) whose frequency response
is controlled by the AFC of the speedga-te. This con-trol keeps
the adjustmen-t (tuning) of the high-pass filter cutoff frequency
to a specific fraction of the Doppler frequency indicated by the
speedgate, i.e. at appro~imately one half the Doppler frequency
20 -
3~
tracked by the speedgate. The cutoff frequency is voltage
controlled by the AFC of the speedgate via the fractional ratio
converter and adapted to the necessary level (slope and offset)
by an operational amplifier (Ul-C of Figure 6). This is
implemented by controlling the transconductance of the three
OTA's (U4, Ull and U13) -that generate the variable poles of
the third-order high-pass filter. Thus, this adaptive equalizer
modifies the video spectrum by the tunable high-pass filter at
frequencies that follow the Doppler frequency -tracked by the
speedgate. The equalizer is inserted be-tween the video amplifier
(not shown) and the input labelled "VIDEO INPUT" at Figures 3
and 5. It reduces the clutter and noise at the input of the
speedgate, the GG and the tracking filter.
In E'igures 6A and 6B, U1, UG, U8, U9 and U12 are
Operational Amplifiers, U2, U3, U4, U5, Ull and U13 are
Operational Transconductance Amplifiers, U7 is an Exclusive
Or Gate and U10 is a Retriggerable/Reset-table Monostable
Multivibrator.
- 21 -
