 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to radar systems generally, and more particularly to
radar equipments which transmit pseudo-random binary coded continuous-wave
signals and which have electronic means for angle determination associated
therewith.
2. Description of the Prior Art
In the prior art the so-called pseudo-random-coded radar concept is, of
itself, well known. The text "Modern Radar"--Analysis, Evaluation and
System Design, by Random S. Berkowitz, published by John Wiley and Sons,
Inc., New York, London and Sydney, (3rd Printing, August 1967, Library of
Congress Catalog No. 65-21446, is one of many references describing this
technique commonly known as PRC radar. Chapter 4 of the aforementioned
text, entitled "Pseudo-random Binary Coded Waveforms," is particularly
pertinent as an elementary reference in respect to PRC radar systems and
concepts. The advantages in respect to high average power-on-target,
favorable signal-to-noise ratio, and accuracy of range determination, as
well as adaptability to Doppler velocity determination have long been well
understood in this art.
The synergistic combination of the present invention also borrows from the
so-called Doppler scanning radar technique, also known per se, in certain
prior art forms. The structural aspects of such simulated Doppler systems
are of interest as background, although the present invention is a new
arrangement and combination of the commutated array technique combined
with pseudo-random coding of the carrier.
The so-called Doppler scanning guidance system has been extensively
described in the patent literature and other technical literature. For
example, U.S. Pat. Nos. 3,626,419; 3,670,337; 3,728,729; and 3,754,261 are
typical of the patent literature relating to this "simulated" Doppler
technology. Also, the technical publication "Electrical Communication,"
published quarterly by International Telephone and Telegraph Corporation,
contains a basic and quite complete description of that particular prior
art as applied to Doppler navigation beacons in Volume 46, No. 4 (1971)
beginning at Page 253.
Still further, U.S. patent application Ser. No. 634,890 filed Nov. 24,
1975, now U.S. Pat. No. 4,042,925 entitled Pseudo-Random Code (PRC)
Surveillance Radar, describes an advanced PRC radar system in which the
inherent range and Doppler determination ambiguities in PRC radar systems
are dealt with to produce an optimum system in which the first range
ambiguity was extended well beyond the useful range of the system, and in
which Doppler ambiguities (i.e., target velocity multiples providing
essentially the same Doppler response) are also placed beyond the velocity
range of interest. That technology may be considered directly applicable
to the present invention. The theory described in the aforementioned
patent application Ser. No. 634,890, which is assigned to the assignee of
the present application, is also helpful in appreciating the advantages to
be obtained from the particular combination of the present invention.
SUMMARY OF THE INVENTION
It may be said to have been the general objective of the present invention
to provide a radar system allowing nearly instantaneous measurement of
range, angle, and Doppler, in respect to multiple targets lying within a
volume being floodlighted by a radar transmitter.
The present invention involves a unique marriage of pseudo-random code
(PRC) range measuring technology and the so-called simulated Doppler angle
measuring concepts as referred to hereabove under the prior art discussion
specifically.
The commonality between the two technologies of the foregoing is that both
utilize continuous waveform (CW) and both are inherently capable of
providing target Doppler (velocity) measurements. The resulting novel
combination provides a low probability of intercept, good clutter
performance, anti-jam and anti-ARM performance. The system has
simultaneous coverage at all angles of interest and a capability for
either surveillance or weapon fire control functions.
Pseudo-random coding is actually by a maximum length sequence, that is, one
which repeats immediately after the end of each predetermined code word
without hiatus. It is important to distinguish at the outset of this
description between the simulated Doppler implied in the so-called Doppler
navigation beacon art hereinbefore identified, on the one hand, and target
Doppler components due to actual target movement with respect to the radar
site at a predetermined ground location. In the so-called Doppler
navigation beacon, a signal is transmitted successively from the elements
of the linear array by commutation, and, at the airborne station the
effect is much the same as if the antenna were physically moving,
producing a "simulated" Doppler effect from which the airborne station can
derive navigation angle information. In the combination of the present
invention on the other hand, the commutated antenna receives only, the
targets having been separately illuminated from a separate transmitting
antenna which covers at least the space sector of interest. This
transmitting antenna radiates the CW signal, bi-phase modulated by the
pseudo-random maximal length sequence. Thus, the commutated receiving
array effectively imposes a phase modulation on received signals and, as
this description proceeds, it will be seen that an individual spectrum
line corresponds to each of the discrete angles in space selected in
accordance with criteria which will be better understood as this
description proceeds. The receiver circuitry connected to the commutated
receiving antenna actually comprises a corresponding plurality of receiver
channels each supplied with a different local oscillator signal and each
having the same predetermined IF bandwidth. Thus, each of those channels
responds only to received signals corresponding to a particular angle in
space. This concept forms the basis for effectively angle-gating received
signals, the range determination being subsequently accomplished by
autocorrelation of the received pseudo-random sequence against the
transmitted code of the same form. Doppler determinations are made
typically by Doppler filter banks responsive to each of these receiver
channels.
The details of instrumentation of the novel concepts and combination of the
invention will be evident as this description proceeds.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1(a) through (e) depict PRC sequence generation with waveforms.
FIGS. 2(a) through 2(c) depict the autocorrelation concept with waveforms.
FIG. 3 is a classical PRC radar ambiguity diagram illustrating the
development of ambiguous range and Doppler responses.
FIG. 4 shows the basic angle-measuring concept which forms a part of the
invention.
FIG. 5 illustrates the development of discrete receivers spectra
corresponding to predetermined space angles.
FIG. 6 illustrates, in graphical form, the amplitudes of spectral
components relating to FIGS. 4 and 5.
FIG 7 illustrates the extension of the concept of FIG. 4 to produce angle
gating of receive signals.
FIG. 8 illustrates the relationship between various angles (beams) in space
as related to FIGS. 4 and 7.
FIG. 9 is a typical system block diagram of an overall device according to
the invention.
FIG. 10 is a timing diagram relating to the PRC code word and antenna scan
timing.
FIG. 11 illustrates the simulation of antenna motion by switching
(commutation).
FIG. 12 illustrates the application of the system according to the
invention to a fire control application.
FIG. 13 illustrates a typical antenna assembly for azimuth, elevation, and
target illumination in each of four quadrants.
DESCRIPTION OF THE PREFERRED EMBODIMENT
Before describing the system arrangement according to the present
invention, it is considered useful to review the techniques of
pseudo-random coding, and to begin that discussion, reference is made to
FIG. 1.
In FIG. 1(a) a CW oscillator-transmitter 101 feeds a bi-phase modulator 102
which codes, or modulates the carrier on lead (d) in accordance with
pseudo-random sequence and feeds this to an antenna 106. The actual coder
104 may be of the well known shift register type with feedback connection
105, arranged to produce the code on lead (c). The code clock 103 produces
a master timing pulses as illustrated in FIG. 1(b). These are spaced
.tau..sub.b which is the bit duration of the code and FIG. 1(c ) is the
code itself. The output of 101 is a steady CW signal represented at FIG.
1(d), and after being modulated in 102 in accordance with the code of FIG.
1(c), the waveform of FIG. 1(e) is produced and applied to the antenna
106. Ordinarily in a PRC system, the antenna 106 (and this is the case in
respect to the present invention also) may be a relatively simple antenna
for producing a relatively broad beam of energy over a sector of space of
interest.
In the code the total time of the maximal length sequence before it repeats
is the word "duration" .tau..sub.w which is equal to the product of L and
.tau..sub.b. L, on the other hand, equals 2.sup.N -1 which is the length
of the code word as a function of the number of states N of the shift
register 104. In the illustrated case, N=5 and L is therefore 31. It will
be understood that the invention is of course not limited to this or any
of the other specific parameters recited. The reasons for the choices made
will be apparent as this description proceeds in connection with the
description of the particular embodiment designed for the achievement of
certain performance characteristics.
Referring now to FIGS. 2a-2c, the PRC autocorrelation function will be seen
at FIG. 2(c). Classically, the code according to FIG. 2(b), which is that
of FIG. 1(c), is multiplied by itself in a mixer 201. The echo signal as
present in lead 202, may be thought of as the received signal and on lead
203 the code is present delayed by an amount D in 204. This signal on 203
is the local or reference code signal. Depending upon the alignment of
these two code signals as they are applied to 201, an autocorrelation
function in accordance with FIG. 2(c) is produced. As these codes "slip"
by each other, due to the changing range of a moving target echo signal,
peaks at a value of +L are produced with spacing .tau..sub.b L spacing as
seen. For a time .tau..sub.b on either side of a correlation peak, the
amplitude of the correlation peak, the amplitude of the correlation
function decreases until it reaches a value of -1 between the correlation
peaks. The signals shown at FIG. 2(b) as present on 202 and 203 may be
thought of as video domain signals having instantaneous amplitudes of
either 1 or 0 and which appear to occur randomly, hence the name
pseudo-random code.
In FIG. 2(a), a Doppler filter 205 is shown through which a discrete value
of target Doppler may be isolated, thus, for a particular value of the
delay 204 and for a particular frequency response of 205, a discrete
target range and velocity is detected.
For maximal length pseudo-random codes (sequences) as employed in the
embodiment being described, the resultant autocorrelation function, FIG.
2(c), always has the same shape. The peak of the autocorrelation function
may otherwise be thought of as occurring when the codes are aligned
bit-for-bit, giving an amplitude of L, i.e., an amplitude value of 31 in
this instance. The range discrimination or resolution time provided is
equivalent to a conventional pulse system with a pulse width equal to
.tau..sub.b. The transmitted signal as PRC modulated is wideband
(approximately twice the code clock rate) and therefore, when received
must be handled by wideband circuits until after the detection function
has been completed. This will be more fully evident at a later point in
this description.
Referring now to FIG. 3, a more or less self-explanatory ambiguity diagram
of a classical type is presented. The term f.sub.d is target Doppler
frequency while .tau. is representative of range.
It is intuitively obvious that, for the correlation described in connection
with FIG. 2, the same response is obtained every .tau..sub.w (relative
slip) between the received and local codes. That is, the response is
ambiguous to the extent that it might be indicative of a range .tau..sub.w
or a multiple thereof. Although not specifically a part of the present
invention, the reader will recall that reference was made to U.S. patent
application Ser. No. 634,890 filed Nov. 24, 1975, entitled Pseudo-Random
Code (PRC) Surveillance Radar, which describes a fully applicable
technique for dealing with this range ambiguity problem. For the sake of
relative simplicity however, that additional disclosure has been omitted
from this specification, since, although it is a highly desirable
addition, it is not germaine to the concepts and specifics of the present
invention, per se.
As is more or less standard, a relatively short code word .tau..sub.w, such
as the 31 bit word assumed in this embodiment, it is required in order
that the reciprocal of .tau..sub.w which is the PRC code repetition
frequency is equal to or greater than twice f.sub.d (max). The value
f.sub.d (max) is, of course, the maximum expected Doppler velocity. The
description in the aforementioned U.S. patent application Ser. No. 634,890
makes the same choice in respect to the elimination of Doppler velocity
ambiguities but employs a multiple clock frequency with associated logic
circuits for recognizing and eliminating false range responses out to a
range much extended beyond the prima facia redundant range of .tau..sub.w.
In FIG. 3, the "0" Doppler lobes are seen to be down by 1/L.sup.2 from the
main response and the Doppler ambiguity sidelobes, although non-existent
for in-range (or zero range) are 1/L times the power in the main lobe for
out-of-range conditions. The width of the Doppler sidelobes, as well as
the main response lobes, is approximately 1/T.sub.i where T.sub.i is the
available integration time, or the time the transmitter dwells on the
target.
Proceeding to FIG. 4, the basic angle measuring technique included in the
invention is described in elementary form.
Assume a substantially omni-directional antenna 401 excited by transmitter
403 at frequency f.sub.t, and a receiving antenna 402 which moves
laterally a distance D in time T.sub.s, then back in zero time (sawtooth
motion) as shown. The antenna 401 radiates a CW signal and 402 receives
reflected energy from the target illuminated by 401. Further assume that
the target is stationary and the discussion is confined to the plane of
the paper. If a stationary target is at angle .alpha., then the motion of
the antenna 402 will phase-modulate the return signal on lead 404 at
f.sub.t linearly with time as shown on FIG. 4. Note that the maximum phase
angle at a given space angle .alpha. measured from a line normal to the
antenna motion is
##EQU1##
It is well known from phase modulation theory that the resultant frequency
spectrum at the antenna output 404 will consist of discrete lines
(assuming infinite integration) around f.sub.t spaced f.sub.s =1/T.sub.s,
where f.sub.s is the frequency of this antenna motion and T.sub.s is its
period. The amplitude of the lines will depend on .DELTA..phi.(.alpha.)
the peak modulation on the return from a target angle .alpha..
Before looking at the general case, it will be instructive to look at the
easily predicted results. Start with the case of .alpha.=0 (i.e., target
on the antenna (normal) boresight). In this case .DELTA..phi.(0)=0, and
the antenna output is a single line spectrum at f.sub.t. After mixing in
mixer 405, this results in a frequency at f.sub.o on lead 406 and the
central filter in the filter bank 407 is excited. A local oscillator 408
at frequency f.sub.o, and mixer 409 operate to provide a mixer reference
on lead 410 at f.sub.t +f.sub.o. FIG. 5(a) depicts this condition, i.e.,
f.sub.o on lead 406.
Now let .alpha. increase in an arbitrarily positive direction. If we let
.alpha. take on a value such that .DELTA..phi.=2.pi., then again a single
line exists at the antenna output at f.sub.t +f.sub.s. This result is
easily seen since the return signal f.sub.t is being phase-shifted exactly
2.pi. degrees on each scan which is equivalent to being modulated
continuously in phase with a phase-time slope 2.pi./T.sub.s =2.pi.f.sub.s.
As is well known, the result of a linear phase modulation on a carrier
frequency f.sub.t is a single frequency at f.sub.t +f.sub.s. This
principal is well known in the technology of single-sideband modulation,
especially in respect to the so-called serradyne modulator to produce a
single spectral line offset from a carrier frequency.
If the i-f amplifier 411 at f.sub.o following the mixer 405 in FIG. 4 has a
bandwidth of f.sub.o .+-.f.sub.s /2, then the resulting line at f.sub.t
+f.sub.s will not get through and the filter bank will not be excited.
This is illustrated in FIG. 5(c). If .alpha. takes an intermediate value
such that 0<.DELTA..phi.(.alpha.)<2.pi., then, in general, energy will
appear at f.sub.t and f.sub.t .+-.n f.sub.s where n is an integer. In this
general case, depicted at FIG. 5(b), only the energy at line f.sub.t will
get through and excite the filter at f.sub.o.
Therefore, in general, as .alpha. moves from .alpha.=0
(.thrfore..DELTA..phi.=0) in a positive direction, the output at f.sub.o
will start at a maximum and drop to zero at an angle .alpha. corresponding
to .DELTA..phi.(.alpha.)=2.pi..
Let the angle .alpha. at which .DELTA..phi.=2.pi. now be determined. Since
##EQU2##
Note that this is just the beamwidth of a rectangular antenna of length D
operating at a wavelength of .lambda., or the first null on its sin X/X
beam pattern.
If the angle .alpha. is extended beyond .lambda./D, the resultant spectrum
out of the antenna will be a single line at f.sub.t .+-.nf.sub.s for
angles of .alpha. which result in .DELTA..phi.=.+-.n 2.pi.. These angles
of .alpha. must correspond to .alpha.=.+-.n (.lambda./D) or at integral
antenna beamwidth angle. (For simplicity, the approximation .alpha.=.+-.n
(.lambda./D) is used although it must be corrected for large angles).
The general case showing the sideband amplitudes of the phase-modulated
spectrum has been worked out by R. C. Cummings and presented in the
Proceedings of the IRE, February 1957, PP 175-186, and the result is shown
in the self-explanatory FIG. 6, interpreted for our example.
This result shows that the output of our receiver of FIG. 4 [which can only
respond to the f.sub.t (n=0) line because of the filtering] will trace out
a sin X/X response as the space angle .alpha. is varied. This again is
what would be expected from an ideal rectangular antenna looking
broadside. In effect, then, the receiver has provided an angle gate around
.alpha.=0.
If we now place in parallel with the receiver of FIG. 4 another receiver
(FIG. 7) with its local oscillator offset from the first receiver by
f.sub.s, then we will have an angle gate (sin X/X antenna response) around
the angle+.lambda./D. This can be seen by looking at the n=1 (f.sub.t
+f.sub.s) component in FIG. 6, which has its maximum at
.alpha.=.lambda./D. Similarly, by placing parallel receivers with their
local oscillators at f.sub.t .+-.nf.sub.s +f.sub.o, we generate a set of
angle gates or simultaneous antenna beams in space. Note that the receiver
channels are all identical, and the components of FIG. 7 are readily
recognized from FIG. 4. FIG. 8 shows still more graphically the situation
hereabove developed.
There is to be understood to be a receiver channel associated with each
antenna beam in space, and each receiver responds (except for the sin X/X
sidelobe response) only to targets in the corresponding beam.
Note again from FIG. 6, that when a target is exactly at .alpha.=n
(.lambda./D) (center of n-th beam in FIG. 8) only receiver n will have a
response. However, when the target moves off the beam center, there will
be coupling between receiver channels per the sin X/X response. For
example, when a target lies at the point where two beams intersect
(.lambda./2D; 3.lambda./2D - - ), there will be equal response in the two
receiver channels associated with the two intersecting beams, as well as
smaller responses in the other receivers.
It will be realized by those skilled in this art that the equivalent of a
scanning beam can alternatively be achieved by using one receiver whose
local oscillator is sequentially switched to the proper offset frequency.
To examine the effect of a moving target, take the simple case where the
target is moving directly (radially) toward the antenna at .alpha.=0. As
stated earlier, it the target is stationary at .alpha.=0, only a single
line at f.sub.t appears at the antenna output. This is translated down to
f.sub.o and rings filter f.sub.o. If the target at .alpha.=0 has a radial
velocity which causes a Doppler shift f.sub.d, then a single line again
will appear at the antenna output at f.sub.t +f.sub.d. After translation,
it will appear at f.sub.o +f.sub.d and will ring one of the filters in the
Doppler filter bank. As long as the maximum Doppler
f.sub.d max.ltoreq.f.sub.s /2
where f.sub.s =antenna scan frequency,
the Doppler shift will not cause an ambiguity in the angle measurement. For
example, if the Doppler shift were equal to f.sub.d =f.sub.s, then the
target would appear to be in the next angle "bin." The significant point,
then, is that if
f.sub.d max.ltoreq.f.sub.s /2
there is no coupling between the angle and Doppler measurement.
It is obvious then that targets at .alpha.=(n.lambda./D) (the center of the
n-th antenna beam) moving radially toward the antenna will produce a
Doppler component in the n-th receiver as described for the .alpha.=0 case
above.
In the general case where the target is off the beam centers, all the
resulting spectral lines out of the receiving antenna are shifted by
f.sub.d, the target Doppler. This will cause the filters at f.sub.o
+f.sub.d for each receiver channel to be energized. The levels into each
filter will be determined by sin X/X response of FIG. 6. Remember that for
the general stationary target, the filters at f.sub.o in each receiver
channel were energized.
Therefore, except for the sin X/X coupling between receivers due to the
angle measurement, the n-th receiver measures the Doppler of the target
associated with the n-th antenna beam. In effect, discrete angle gates are
formed by the first mixer of each receiver channel.
Referring now to FIG. 9, the overall system block diagram of a typical
arrangement according to the invention is shown. Note that that the
transmitter section comprising a X-band oscillator 901, bi-phase modulator
903 and antenna 904, is equivalent to the same components in FIG. 1(a).
Also, in the general case shown, a complete PRC receiver channel (as
previously identified) would be associated with each angle gate (or
beamwidth). This angle gate is formed by the first mixer in each receiver
which is fed by a local oscillator signal at f.sub.t +f.sub.if +nf.sub.s,
where n is equal to the corresponding antenna beam at n (.lambda./D).
These local oscillator signals are generated coherently in synthesizer 908
which is a multi-frequency generator in effect. As shown in FIG. 10, the
antenna scan period T.sub.s is made equal to the PRC word period,
.tau..sub.w. In this manner, both the criteria for proper PRC ranging
without encountering Doppler ambiguities,
f.sub.w .gtoreq.2f.sub.d (max)
and the criteria for angle gating
f.sub.s .gtoreq.2f.sub.d (max)
are both met.
In addition, if f.sub.w =f.sub.s, it is less likely that unwanted beat
frequencies will be created.
Assume first that a target is exactly at .alpha.=(n.lambda./D), a
compensating angle point with a Doppler f.sub.d. In the n-th receiver
channel (or angle gate), the receiver response will be that of a
conventional prior art PRC radar since the antenna scanning modulation is
compensated for as previously described. The filter at f.sub.d will be
excited in the range gate (receiver) channel corresponding to the target
range. The other range gates will not correlate properly, and the
corresponding f.sub.d filter response will be down by the PRC correlation
function 1/L.sup.2 as predicted by PRC radar ambiguity diagram of FIG. 3.
If we place the target at .alpha.=(n.lambda./D) and look at the adjacent
angle receivers (or angle gates), the local oscillator compensation will
move the entire PRC spectrum by f.sub.s =f.sub.w. Here f.sub.s is the scan
frequency (commutation rate) and f.sub.w is 1/.tau..sub.w, i.e., f.sub.w
is the code word repetition frequency. This is equivalent to a PRC
spectrum with a Doppler of f.sub.d +f.sub.w. After PRC correlation in the
range gate in which the target lies, a single line at f.sub.d +f.sub.w
will result. Since the narrow band i-f cuts off at f.sub.w /2, the signal
will not be detected.
If we look at the other range gates in the adjacent angle receiver
channels, the response is that of a PRC radar which is out of range but at
the first Doppler ambiguity reponse of the PRC ambiguity diagram. Thus the
response will be down only 1/L again as predicted by the PRC ambiguity
diagram. The same reasoning can be applied to other than the adjacent
receiver channels.
When the target is between the .alpha.=(n.lambda./D) points, the situation
is more complicated because the scan modulation causes additional lines to
be generated at multiples of f.sub.s (see FIG. 6). The worse case appears
to be when a target is exactly between two compensation angle points, for
example at .lambda./2D; 3.lambda./2D, etc.
As discussed previously in the angle measurement sections, the
corresponding two adjacent receivers will have equal responses and be down
3 dB from maximum response at (n.lambda./D). The out-of-range gate
response in these corresponding two adjacent receivers will be down by
1/L.sup.2 from the response in the in-range gate response (which is down 3
dB). Other angle-gated receiver channels will be affected by the sin x/x
response of the scan modulation as well as the PRC ambiguity diagram
response. The unwanted responses, however, may be expected to be on the
order of 10 dB predicted by the sin x/x response of FIG. 6 since the PRC
unwanted response is on the order of 15 dB.
On FIG. 9 it will be noted that the oscillator 901 is identified as an
X-band device, this being a typical radar frequency. The power divider 902
connected to 901 passes most of the relatively low band power of 901 to
the bi-phase modulator 903. However, a small amount is diverted to mixer
907 where it is mixed with the output of a stable IF oscillator (coho)
906. The frequency of 901 is identified as f.sub.t and that of the coho
906 as f.sub.if, accordingly the output of mixer 907 is f.sub.t +f.sub.if.
The output of 907 drives the angle gate frequency synthesizer 908
directly, this synthesizer being no more than a multiple frequency
generator synchronous with the output of mixer 907. In this way, the
multiple local oscillator signals generated by 908 provide phase coherence
throughout the signal processing functions following the mixer group 912.
The power amplifier 905 picks up the bi-phase modulated output of 903 and
provides wideband amplification thereof, at least sufficient to
accommodate the approximately 10 MHz bandspread at the output of 903.
Subsequently the antenna 904 radiates this energy. The antenna 904 is much
the same as a standard PRC sector illuminating radiator and need be no
more than a relatively straightforward and simple horn radiator. Later, in
connection with FIG. 13 the overall antenna arrangement which would be
typical of the combination of the present invention will be discussed in
more detail.
As is well known in PRC generation, a code clock 909 in this example,
running at 5 MHz, drives the 5 stage PRC coder 910, this arrangement being
substantially the same as that of FIG. 1(a). The output of coder 910 not
only supplies the maximal length sequence to modulator 903 but also drives
the 5 stage binary code delay device 911. This device is ordinarily no
more than a standard device of one code word total delay with 31 taps,
whereby the 31 bits of the code word (the bit duration being typically 200
nanoseconds in this example) are presented at the taps before the word
repeats itself. The total word duration .tau..sub.w being 6200 NS in this
example, the first redundant range is therefore approximately 930 meters,
however, as hereinbefore indicated, the separate technique described in
U.S. patent application Ser. No. 634,890 filed Nov. 24, 1975, entitled
Pseudo-Random Code (PRC) Surveillance Radar, can readily be added to the
total structure to increase this first redundant range greatly.
Referring now to the antenna 913 which is assumed (for simplicity at this
point) to move laterally and linearly with a velocity S.sub.a (see also
FIG. 10) occupies extreme positions 913 and 913a. It is assumed to return
from 913a to the 913 position in zero time for the sake of explanation.
The output of this antenna which receives the reflected signals from
targets illuminated by 904, branches into each of the 2n+1 receiver
channels within block 912. Reference back to FIG. 8 will explain this 2n+1
figure. From the output of block 912, each of the mixers therein
comprising the illustrated zero mixer, the minus n mixer and the plus n
mixer, as well as all the remaining ones of the 2n+1 mixers, includes a
wideband IF amplifier typically 918 or 919 following it. These wideband IF
amplifiers have a bandwidth of at least 10 MHz centered about f.sub.if in
order to accommodate the PRC modulation still present at this point.
The mixers within 912 and these wideband IF amplifiers of which 918 and 919
are typical, comprise the receiver channels variously referred to herein.
Each of the mixers within 912, having its discrete local oscillator
frequency provided by 908 corresponds to a beam (angle gate or angular
bin) depicted in FIG. 8. Although beams corresponding to these discrete
angle "bins" are not actually formed in space, the performance of the
overall equipment has very much the equivalent effect and the term "beam"
is very often used in this connection and in synthetic aperture radar
situations in general.
Two decoder blocks, 914 and 915, are illustrated. It is to be understood
that one such decoder block is provided for each of the 2n+1 receiver
channels aforementioned, and within each of these decoder blocks, for
example 914, the phase coincidence detectors shown, typically 920, 921,
and 922, are only three of the 31 such circuits present in 914 and the
other such decoder circuits. These demodulator blocks may be also
characterized as range channel demodulators, the correlation process being
carried out therein. In addition, the received signal corresponding to
that angle gate (beam), for example as received from 918, is fed in
parallel to all 31 coincidence circuits of 914. The other input of each of
these 31 coincidence circuits of 914 comes from a discrete corresponding
one of the 31 outputs of the delay device 911. Concerning that delay
device 911, it will be obvious to those skilled in this art that suitable
implementations can be obtained from the technologies of delay lines,
shift registers or the like.
The blocks 914 and 915 and their companions are followed by blocks
typically 916 and 917, the latter containing a clutter notch filter,
typically 923 or 925 followed by narrowband IF amplifiers 924 and 926,
respectively. The clutter notch filters 923 and 925 merely eliminate
frequency components corresponding to non-moving signals, i.e., those
whose spectral content does not contain target Doppler (velocity)
modulation components. At this point, the PRC coding is no longer present,
the inputs to blocks such as 916 and 917 being passable through a bandpass
filter having a bandwidth on the order of 75 KHz centered around f.sub.if.
These blocks 916 and 917, come directly from PRC technology and it will be
realized that one clutter notch filter and narrowband IF amplifier follows
for each of the 31 coincidence circuits for each of the 2n+1 correlators
of which 914 and 915 are only typical. Thus, the output of each narrowband
amplifier, such as 924 for example, represents a discrete range within a
discrete beam or angular "bin." Each of these narrowband IF amplifiers is
then capable of indicating at its output in respect to presence or absence
of a signal within that range bin and angle bin. Doppler filter banks,
typically 927 and 928, following each of those narrowband IF amplifiers,
as illustrated at 927 comprising filters 929, 930 and 931 typically. The
number of filters in 927 or 928 or any of the other filter banks provided
is determined by the velocity resolution desired essentially without
regard to other system parameters.
At this point, the significance of FIG. 10 having been already appreciated
(namely that it places T.sub.s =.tau..sub.w, reference is made to FIG. 11.
In a practical system, the antenna 913 which is assumed to move physically
over the distance D in FIG. 9 or in FIGS. 4 and 7 would normally be
replaced by a linear array of radiators separately fed through a
commutating arrangement in accordance with FIG. 11. FIG. 11 is otherwise
self-explanatory.
At this point in the discussion it will be realized by those skilled in
this art that a simpler system can be configured it one were willing to
search with a single angle gate and/or a single range gate per angle
channel. Such a modification can be instrumented once the principles of
the present invention are clearly understood. In that event, the block 912
of FIG. 9 would contain only one mixer and the angle gate frequency
synthesizer 908 would issue a programmed locl oscillator frequency.
Moreover, in such a variation, the correlation process could be carried
out by a searching range gate against the output of a single wideband IF
amplifier following a single mixer of the block 912.
FIG. 12 illustrates a potential fire-control application in which a
transmitter 1201 and surveillance volume illuminating antenna 1202
transmit the CW PRC modulated signals as already described.
In FIG. 12 the commutated receiving antenna 1203 is, as already described,
and an additional receiving antenna 1204 is vertically oriented in order
to provide the same angle gating function in elevation as is provided in
azimuth. The receiver 1205 then provides multiple channels for each of the
1203 and 1204 antennas from which target range, velocity and angle
information issues, to an indicator 1206 and also to a control unit 1207
relating to missile launcher 1208.
From this complete range, angle (in two dimensions) and Doppler information
missiles M.sub.1 and M.sub.2 can be programmed toward the targets T.sub.1
and T.sub.2. That is, the seeker-receiver in the missile can be
"locked-up" in range, angle and velocity. The semi-active missiles can
then utilize reflected energy from targets resulting from the illumination
by 1202 to home on target.
The unique advantage in ground-to-air missile (defense) system in employing
the present invention is the increased fire power provided by the
capability for parallel missile firing. This capability follows from the
contemporanenous range, angle and velocity data for multiple targets which
the system of the invention provides.
Referring now to FIG. 13, a configuration of antennas is illustrated from
which the entire 360.degree. about any given location can (through
switching selection) be subjected to the surveillance and control
capabilities of the present invention. In azimuth, four multi-element
commutatable arrays 1301, 1302, 1303, and 1304 are shown which are
selectively employed contemporaneously with one of the elevation arrays
1305, 1306, 1307, or 1308, respectively. A four quadrant transmitting
(illuminator) antenna arrangement comprising horns 1309, 1310, 1311, and
1312, is shown, it being understood that the appropriate one of these is
also selected for the space quadrant of interest.
The employment of elevation angle determination in accordance with the
present invention of course requires duplication of the angle gating
functions of FIG. 9, but the pseudo-random range and Doppler velocity
determining functions need not be duplicated. The manner of integrating
the elevation operation into the arrangement of FIG. 9 will also be
evident to those skilled in this art once the principles of the present
invention are well understood.
Other modifications and variations and details of arrangements of the
system described may present themselves to those skilled in this art, and
accordingly, it is not intended that the scope of the present in | | |
