U.S. patent application number 11/963077 was filed with the patent office on 2010-12-09 for sonar system and process.
This patent application is currently assigned to AC Capital Management, Inc.. Invention is credited to David Ashworth, Larry Freeman, John Green, Donald Lerro.
Application Number | 20100309751 11/963077 |
Document ID | / |
Family ID | 34915472 |
Filed Date | 2010-12-09 |
United States Patent
Application |
20100309751 |
Kind Code |
A1 |
Lerro; Donald ; et
al. |
December 9, 2010 |
SONAR SYSTEM AND PROCESS
Abstract
A sonar system and method of use capable of discriminating a
direct acoustic signal present at 60 dB or more above the acoustic
echo signal.
Inventors: |
Lerro; Donald; (Colchester,
CT) ; Freeman; Larry; (Potomac, MD) ; Green;
John; (Jewett City, CT) ; Ashworth; David;
(McLean, VA) |
Correspondence
Address: |
KELLEY DRYE & WARREN LLP;STEVEN J. MOORE
400 ALTLANTIC STREET , 13TH FLOOR
STAMFORD
CT
06901
US
|
Assignee: |
AC Capital Management, Inc.
Wilmington
DE
|
Family ID: |
34915472 |
Appl. No.: |
11/963077 |
Filed: |
December 21, 2007 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11514872 |
Sep 1, 2006 |
7330399 |
|
|
11963077 |
|
|
|
|
10953300 |
Sep 29, 2004 |
7106656 |
|
|
11514872 |
|
|
|
|
60506507 |
Sep 29, 2003 |
|
|
|
Current U.S.
Class: |
367/99 ;
367/138 |
Current CPC
Class: |
G01S 15/32 20130101;
G01S 15/003 20130101 |
Class at
Publication: |
367/99 ;
367/138 |
International
Class: |
G01S 15/06 20060101
G01S015/06; H04B 1/02 20060101 H04B001/02 |
Claims
1. An active sonar system for detecting object(s) in water
comprising: a transmitter, with an output connection, that
generates a transmit signal and a conjugate of said transmit
signal; a transmit transducer array, connected to said transmitter
output connection, and configured to produce an acoustic signal in
the water in response to said transmit signal; at least one
hydrophone, with a hydrophone signal output connection, configured
to receive said acoustic signal from the water and produce a
hydrophone signal; a heterodyne stage having transmit signal and
conjugate transmit signal inputs, connected to said transmitter
output connection, and a hydrophone signal input, connected to said
hydrophone signal output connection and a difference frequency
signal output connection, configured to produce a difference
frequency signal in response to the hydrophone signal and said
transmit signal and said conjugate of said transmit signal; a
frequency filter having connected to said difference frequency
output of said heterodyne stage and having an output connection;
and a display having an input connection connected to said
frequency filter output connection and configured to display
detection of said object(s) in water.
2. The active sonar system, in accordance with claim 1, wherein
said transmitter is continuously transmitting said acoustic
signal.
3. The active sonar system, in accordance with claim 1, wherein
said transmitter is transmitting a frequency modulated signal.
4. The active sonar system, in accordance with claim 1, wherein
said transmitter is transmitting a swept frequency modulated
signal.
5. The active sonar system, in accordance with claim 1, wherein
said transmitter is transmitting a linear swept frequency modulated
signal.
6. The active sonar system, in accordance with claim 1, wherein
said transmitter is transmitting an acoustic signal whose frequency
versus time characteristic is a periodic sawtooth.
7. The active sonar system, in accordance with claim 1, wherein
said transmitted waveform bandwidth exceeds the center frequency of
said transmitted waveform bandwidth.
8. The active sonar system, in accordance with claim 1, wherein
said frequency filter is implemented as a Fourier transform with a
shading window that provides tapering for low side-bands.
9. The active sonar system, in accordance with claim 7, wherein
said shading window is selected from a group consisting of
Chebychev, Taylor and Kaiser-Bessel windows.
10. A method of detecting objects in the water comprising the steps
of: generating and transmitting an transmitted acoustical signal in
a desired direction in the water; generating conjugate of said
acoustical signal;receiving reflection signal of said transmitted
acoustical signal in water from said object; heterodyning said
reflection signal with said transmitted acoustical signal and said
conjugate of said transmitted acoustical signal to form difference
frequency signal; frequency filtering of said difference frequency
signal; and displaying said filtered difference frequency signal
whereby object is detected.
11. The method in accordance with claim 10 wherein said transmitted
acoustical signal is continuously transmitting.
12. The method in accordance with claim 10 wherein said transmitted
acoustical signal being frequency modulated.
13. The method in accordance with claim 12 wherein said transmitted
frequency modulated acoustical signal is being swept in
frequency.
14. The method in accordance with claim 12 wherein said transmitted
frequency modulated acoustical signal is being linearly swept in
frequency.
15. The method in accordance with claim 12 wherein said transmitted
frequency modulated acoustical signal's frequency versus time
characteristic is a periodic sawtooth.
16. The method in accordance with claim 12 wherein said transmitted
frequency modulated acoustical signal swept bandwidth exceeds the
center frequency of said swept bandwidth.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This is a Continuation Application of U.S. patent
application Ser. No. 11/514,872, filed Sep. 1, 2006, which is a
continuation of U.S. patent application Ser. No. 10/953,300, filed
Sep. 29, 2004 which claims benefit under 35 U.S.C. 11119(e) of U.S.
Provisional Application Ser. No. 60/506,507, filed on 29 Sep.
2003.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] The present invention is generally related to Sonars. In
particular, it is related to an active sonar system capable of
receiving and transmitting sound signals simultaneously.
[0004] 2. Description of the Related Art
[0005] SONAR (Sound Navigation and Ranging) is a technique that
uses sound propagation under water to navigate or to detect objects
in or on the water. As Is known in the art, there are two types of
sonar: passive and active. Passive sonar seeks to detect an object
target by listening for the sound emanating from the object being
sought. Active sonar creates a pulse of sound, and then listens for
reflections of the pulse from a target object. To determine the
distance to the target, elapsed time from emission of a pulse to
reception is measured. To determine the directional bearing,
several hydrophones are used to measure the relative arrival time
to each in a process called beam forming
[0006] The ability of a system to detect desired sound, or
`signal`, in the presence of interfering sound, (i.e. noise or
reverberation) is generally referred to as its `Recognition
Differential` (RD). RD is defined as the Signal-to-Interference
ratio at which the system can, with some specific probability
(usually taken at 50%), detect the signal while not exceeding a
specified probability of false alert. RD is usually expressed in
Decibels (dB). The lower the RD, the better the system
performs.
[0007] It has been shown that the longer the `look` or processing
time, the lower the RD (see for analysis Burdic, W. S., Underwater
Acoustic System Analysis, Prentice Hall, Englewood Cliffs, N.J.,
1984). Since passive sonar systems can always be in a listening or
`receive` mode, and generally seek target signature components that
are continuously radiating, they are able use many minutes of
processing time and thus to achieve very low RDs. By contrast,
conventional active sonar systems typically use short pulse type
transmissions and their receive processing time is therefore
limited to a very few seconds, or even only fractions of a second.
Consequently, active sonar RDs are generally 10 to 30 dB higher
than those of passive systems; this is equivalent to one to three
orders of magnitude in linear terms.
[0008] Conventional active sonars are typically limited to
relatively short duration pulse type transmissions for a number of
reasons. One reason is that in some environments unwanted
reverberation builds up as transmission times Increases; this is
particularly relevant to limited bandwidth systems in shallow water
environments. A more fundamental reason is that most active sonars
systems cannot receive while they are transmitting. Often, this is
because they use the same device, (called a `transducer`), to both
transmit and receive, and transducers cannot do both at the same
time. Such systems are necessarily `mono-static`, meaning that
their transmissions and receptions take place at one location.
(Note that the converse is not necessarily true; some mono-static
systems use co-located but separate devices for transmission &
reception.) Most generally however, the inability to receive while
transmitting is because active sonar transmission levels are
inevitably so much higher than the levels of the echoes being
sought that the acoustic transmit level overloads, or at least
effectively `jams`, the receiver being used. This is even true for
most so-called bi- or multi-static systems, where transmitting
& receiving are done by separate devices located some distance
apart.
[0009] It would be desirable to have a sonar system that would
permit substantially continuous stream of incoming data that would
not be limited to highly direction transmitters or high
frequencies, particularly in sonar systems dedicated to the
detection of targets, such as submarines, which are seeking to
escape detection.
[0010] In prior art, U.S. Pat. No. 5,150,335 by Hoffman describes a
waveform generation and processing technique that could be used to
resolve Doppler and range ambiguity using interrupted frequency
modulation for continuously transmitting sonar. Hoffman, however,
does not address linearity and rejection criteria and therefore
fails to teach those elements necessary to effectively to detect
echoes while transmitting.
[0011] Teel et al., on the other hand, address in U.S. Pat. No.
4,961,174 the need for acoustic isolation from the transmitter and
receiver and does so with physical vertical separation requiring a
strong acoustic layer not processing rejection as in this approach
and is therefore limited to relatively few specific environments
(e.g., deeper water applications) and can be avoided by an
intelligently operated object such as a submarine. U.S. Pat. No.
6,128,249 by Sullivan discusses a method of continuously
transmitting by using sequences of tones each separated
sufficiently in frequency to avoid interference. Sullivan like
Hoffman does not address linearity and rejection criteria, and the
series of tones used by Sullivan fail to permit effective range
resolution and reverberation rejection.
[0012] In sum, the art fails to show how to effectively
discriminate between signals that sonar receives from its own
transmitter and echoes from the intended target subject to certain
linearity and rejection requirements.
SUMMARY OF THE INVENTION
[0013] The present invention involves an active sonar system
capable of continuously receiving while transmitting. Further
provided is a method of operating the sonar system subject to a
unique set of linearity and rejection requirements. The system's
receiver may discriminate (in bearing, range, and/or frequency)
between the signal it receives from its own transmitter and the
reflected signals, or echoes, it receives from the intended
target(s) even while the system is transmitting. The sonar provides
improved differential gains in reverberation and noise and rapid
target detection.
[0014] One embodiment involves an active sonar system for detecting
object(s) in water. The system comprises a transmitter capable of
continuously transmitting an acoustic signal; a receiver capable of
receiving a reflected acoustic echo of said acoustic signal from
said object(s), and a direct acoustic signal from said transmitter.
The receiver is capable of discriminating between the direct
acoustic signal and the acoustic echo signal it receives from the
object while the transmitter is transmitting. The discrimination
can occur when the direct acoustic signal is present at 60 dB or
more above the acoustic echo signal.
[0015] Another embodiment provides a method for detecting sonar
signals. The method involves transmitting an acoustic signal.
During the transmission of the acoustic signal, a receiver receives
a reflected acoustic echo of the acoustic signal from object(s), as
well as directly receiving the acoustic signal. A signal of the
acoustic echo and the direct acoustic signal are processed in a
manner to discriminate between the acoustic echo signal and the
direct signal when the direct acoustic signal is present at 60 dB
or more above the acoustic echo signal.
[0016] Still another embodiment provides method for detecting sonar
signals comprising: generating a waveform; performing waveform
shaping to avoid signal discontinuities; transmitting the waveform
acoustically; directly receiving the acoustic waveform and
reflected echoes of the acoustic waveform; performing temporal
frequency rejection on the received acoustic waveforms; and
optionally performing the step of spatially filtering the received
acoustic waveform; and optionally tracking the detected reflected
echoes.
BRIEF DESCRIPTIONS OF DRAWINGS
[0017] FIG. 1a is a drawing of a sonar system in the field showing
the relationship between ship-board electronics, transmitter,
target and a receiver sonbuoy, according to the present
invention.
[0018] FIG. 1b is a schematic drawing of an embodiment showing the
various components of the ship-board electronics and its
relationship to a transmitter, a target, and a receiver below the
water-line, according to the present invention.
[0019] FIG. 2a is a drawing showing a more detailed rendition of
the ship-board electronics shown in FIG. 1b in accordance with the
present invention.
[0020] FIG. 2b is a schematic drawing showing the frequency
filtering section of ship-board electronics of the
receive-while-transmit sonar of the present invention.
[0021] FIG. 2c is a graphical drawing showing an output of the
frequency filtering section of FIG. 2b for a linear frequency
modulated transmit sweep with data captured on a single hydrophone,
according to the present invention.
[0022] FIGS. 2d and 2e are graphical drawings showing heterodyne
output from an experiment in a lake where in FIG. 2d shoreline edge
of the lake can be seen at 2 kilometers and 3.5 kilometers, in
accordance with an embodiment of the present invention.
[0023] FIG. 3a is a graphical drawing showing the amplitude of a
signal with and without tapering, according to the present
invention.
[0024] FIGS. 3b and 3c are graphical drawings showing the spectrum
over time without and with the signal (of FIG. 3a) tapered,
respectively, in accordance with the present invention.
[0025] FIGS. 4a and 4b are drawings showing the effect of an
illustrative detection display without and with proper signal
tapering to minimize signal interference, according to the present
invention.
[0026] FIG. 5 is a graphical drawing showing a comparison, between
conventional pulsed sonar and the receive-while-transmit sonar, of
the probability detection of a target as a function of time, and
improvement in detection time in accordance with the present
Invention.
[0027] FIG. 6 is a graphical drawing showing transmit and receive
waveforms, according to the present invention.
[0028] FIG. 7 is a graphical drawing showing heterodyne outputs for
stationary, closing and maneuvering targets, according to the
present invention.
[0029] FIG. 8 is a graphical drawing showing probability of
detection as a function of time, and demonstrating required
detection level, according to the present invention.
[0030] FIG. 9 is a graphical drawing showing root-mean-square (rms)
speed estimation error as a function of time, demonstrating
improved detection time in accordance with the present
invention.
DETAILED DESCRIPTION OF THE INVENTION
[0031] The present invention involves an active
"receive-while-transmit (RWT)" sonar system capable of continuously
receiving while transmitting, and provides a method of operating
the sonar system subject to linearity and rejection requirements
according to the invention. The system's receiver discriminates (in
bearing, range, and/or frequency) between the signal it receives
from its own transmitter and the reflected signals, or echoes, it
receives from the intended target(s) even while the system is
transmitting. This is accomplished by incorporating into the system
a receiver having a large dynamic range (the ratio of minimum to
maximum input levels over which the receiver's output level remains
linearly proportional to the level of the receiver's input), as
explained in more detail below.
[0032] FIG. 1a shows an embodiment sonar system employed on a
surface ship 10 in conjunction with a submerged craft 20 carrying
an array of acoustic sensors or receivers, such as hydrophones 23
towed by surface ship 10. Cable 25 which tows the craft also
includes electronic transmission means, such as conductive wires,
for supplying electronic signals from array 23 to a data processing
center and display 30 on surface ship 10. In active mode, active
transmitter 40 below the water line 45 transmits an acoustic signal
50. The signal Is reflected from target 60 as acoustic echos 70 and
detected by receiver 20. Receiver 20 may comprise one or more of
hydrophones 23. At the same time, the acoustic signal transmitted
by transmitter 40 is directly received by receiver 20 as a direct
signal 80 from the transmitter. The signals are processed on-ship
board electronics 30.
[0033] Ship-board electronics 30 is shown in more detail in FIG.
1b. Active transmitter 40, target 60 and receivers 20 below the
water line are also shown in relation to the electronics of the
same figure. In one embodiment of the present invention, an
excitation signal Is generated to transmit a wave form 80 either
directly to transmitter 40 or, optionally, through transmit
beamformer 110. After the transmitted signal 105 returns to
receiver 20, the signal is either directly passed through a
heterodyne filter 120 or, optionally, through a receive beamformer
110 before the signal is processed 130, and subject to high data
rate tracking 140, as explained further below. The result is
displayed on display 150.
[0034] In an aspect of an embodiment shown in FIGS. 1a and 1b,
transmitter 40 transmits an acoustic signal continuously. The
signal is omnidirectional and arrives at an object, such as 60 in
both Figures, and at its own receiver 20. The acoustic signal may
be less than or equal to 12 kHz, or .ltoreq.10 kHz or even
.ltoreq.5 kHz. In an aspect of the invention, the acoustic signal
travels with a reduced level of energy in the direction 80 of
receiver 20. The energy level may be reduced from about 3 dB to
about 60 dB, or from about 5 dB to about 25 dB, or from about 25 dB
to 55 dB. Furthermore, transmitter 40 can transmit a frequency
modulated signal, a sweeped frequency modulated signal, or a linear
sweeped frequency modulated signal. The modulated signal is
amplitude tapered to avoid frequency discontinuities, which a re
well observed in the art. The transmitter and the receiver may be
separated by 12 acoustic wavelengths or less.
[0035] Receiver 20 receives signal directly 80 from its own
transmitter 40 as well as echo signal 70 reflected from object 60.
The receiver may comprise a vector sensor or an array of vector
sensors. The receiver has a dynamic range greater than about 80 dB
and is capable of discriminating between the direct acoustic signal
and the acoustic eco signal it receives from the object while the
transmitter is transmitting. The object may be still or in motion.
In an aspect of the invention, the discrimination can occur when
the direct acoustic signal is present at 60 dB or more above the
acoustic echo signal. The discrimination capability is achieved by
temporal frequency filtering and spatial filtering which attenuates
the direct signal, as described further below.
[0036] An embodiment of the system shown in FIG. 2a is capable of
detecting an object 60 at a range greater than 5, and 10 kiloyards
from receiver 20. The system is capable of detecting an object at
the same depth as either the transmitter or the receiver. In
another aspect, the system is capable of detecting an object at a
depth other than the depth of either the transmitter or the
receiver as well as detecting an object underneath the ocean floor.
Furthermore, the system is capable of detecting an object at any
bearing in respect of the receiver.
[0037] In one aspect, for the system receiver 20 to be able to
discriminate (in bearing, range, and/or frequency) between the
direct signal 80 it receives from its own transmitter 40 and the
echoes 70 it receives from the intended target(s) 60 even while the
system is transmitting, two conditions must be met: in one aspect,
the acoustic level received at the receiver may not exceed (or at
least may not significantly exceed) the sum of the receiver's
Minimum Detectable Signal dictated by the acceptable sea noise
floor plus its Dynamic Range; that is, the received level may not
overload the receiver. This aspect will be referred to as the
`linearity` requirement (or leveling) of the invention. Second, the
receiver must be capable of sufficiently rejecting, in either or
both the bearing and frequency domains, the transmitted signal so
that it does not raise detection thresholds in the receiver's
search domain. That is, the system must not `jam` itself in the
direction(s) and frequency regimes in which it seeks to obtain an
echo from the target. This other aspect may be referred to as the
`rejection` requirement and it can generally be accomplished
through a combination of beam/filter shading to attain adequate
side lobe control and adaptive/notched filtering to further
discriminate against the levels received directly from the
transmitter.
[0038] The linearity requirement of such embodiment is shown in
FIG. 2a where the numbering scheme of FIGS. 1a and 1b is followed
for ease of relating the various parts of the invention, the primed
numerals referring to similar parts throughout the several views.
One or more elements of a transducer array 90' transmit wave forms
at a source acoustic level L (LS) in the direction of target (60 in
FIG. 1b). Beamformer 100' transmits in the direction of receiver
(20 in FIG. 1b) for null steering which yields Null.sub.1.about.
equivalent to L-Level in direction of receiver. An acoustic baffle
105' may also be used, though not necessarily desirable at times
due to Baffle Loss (Level reaching baffle--level leaving baffle in
direction of receiver). The acoustic signal travels through water
separation 45' suffering transmission loss (TL), as shown in FIG.
2a. The linearity requirement of receiver array 20' comprising one
or more hydrophones is formed by establishing a relationship
between the various parameters of the system as given by the
following equation:
LS-(TN+TL).ltoreq.NL+DR} (Equation I)
[0039] where LS--Source Level in Direction of Target
[0040] DR--Dynamic Range of the system
[0041] TN--Transmit Spatial Null in the Direction of the
Receiver
[0042] TL--Transmission of Loss from Transmitter to Receive
Element(s)
[0043] NL--Receiver Noise Level
[0044] As an exemplary typical system, consider a 2 KHz (2000
Hertz) transmitter array of approximately 20 elements (hydrophones)
designed to provide an approximately 225 dB source level,
functionally comparable to an array operated with a high dynamic
range towed array. Assuming LS=225 dB, DR=120 dB, TN=25 dB, TL=40
dB, NL=50 dB, and the above linearity relationship shows 160
dB.ltoreq.170 dB
[0045] Hence, the linearity requirement is satisfied. It will be
known to those skilled in the art that several of the loss terms
used above is conservative. For example, most bi-static or
multi-static configurations involve transmission losses of 50 dB to
80 dB vs. the 40 dB value used here, thus negating the need for
transmit null steering or baffles. Similarly, as frequency is
reduced, sea noise, and therefore Minimum Detectable signal Level
is increased, again reducing the need for null transmit steering
and/or baffles.
[0046] It will also be noted in the example above that a received
level of 160 dB while well within the dynamic range of such
exemplary system, would also be above the system minimum detectable
level of the noise floor. Since the system remains linear, it is
now possible to provide a combination of spatial and frequency
based "notch filtering" techniques to further reject the transmit
signal from interfering with reception.
[0047] In addition to the linearity requirement defined above,
another aspect of an embodiment of the present invention
establishes a requirement for direct signal (80 in FIG. 1b)
rejection.
[0048] It will be known to those skilled in the art that in any
continuous transmission approach, a strong transmission to the
receiver will have the adverse effect of masking the reception of
target echoes of interest. In a method employing a separate
transmitter and receiver, a bi-static scenario inherently exists
and the transmitted energy will appear at a fixed time delay or at
multiple time delays due to multi-path arrival. With a Frequency
Modulated Continuous Wave (FM-CW) transmission the time delay of
the return corresponds to a frequency shift of the received
waveform. The transmitted energy received directly will be strong
relative to the level of the desired reception and frequency
sideband leakage becomes a dominant source of noise. This problem
can typically be handled by using frequency filtering with
extremely low side band response. This linear frequency filtering
may optionally be augmented by non-linear adaptive frequency
cancellation at the receiver. If the receive sensor contains an
array of hydrophones then receive beam-forming may also be used to
suppress signals arriving from the direction of the source. This
receive beam-forming may use either a conventional linear or an
adaptive approach. If the source has sufficient directionality then
it may also be possible to steer a null of the source transmission
beam in the direction of the receiver. Transmit signal rejection at
the receiver can be performed using spatial or frequency domain
techniques. Depending on the sensor system either one or both
spatial and frequency rejection can be applied.
[0049] In another embodiment, the direct signal rejection
requirement of the present invention involves a spatial (bearing)
component and a frequency (range) component. Thus, defining
parameters,
[0050] RS--Spatial Side lobe Rejection
[0051] NDI--Receiver directionality against noise
[0052] RSR--Receiver Spatial Rejection Against Signal
[0053] RFR--Receiver Frequency Rejection
[0054] then the rejection requirement in terms of an Interference
Level (I) requirement is expressed by the equation:
I=LS-TN-RS-TL-RSR-RFR<NL-NDJ (Eq. II) Spatial and/or frequency
dependent processing 110', 130' are shown in FIG. 2a followed by
the displaying of the detection of the target on display 150'.
[0055] It will be noted that the conventional beam-forming
techniques provide spatial rejection where the level of rejection
typically is determined by the number of hydrophones available and
their spacing. Beyond conventional approaches adaptive techniques
that are used to cancel known signals or "anti-jamming" methods can
be applied to provide additional direct signal rejection.
[0056] It will also be noted that the present invention can readily
take advantage of modern high performance digital electronics in
which the spatial and frequency notch filtering design methods can
readily exceed the rejection requirement. A heterodyned FM approach
may he used to translate frequency into range thus turning spatial
beam forming & frequency filtering into the range and hearing
information provided by conventional active pulsed sonar. This may
be accomplished using adaptive noise cancellation techniques and
appropriate high rejection side lobe methods (see, e.g., Widrow,
B., Steams, S. Adaptive Signal Processing, Prentice Hall, Englewood
Cliffs, N.J., 1985).
[0057] An embodiment of a heterodyned FM approach is shown in FIG.
2b following element 120 of FIG. 1b and as a part of element 110'
in FIG. 2a In FIG. 2b, received signal from a hydrophone 110'' or
array of hydrophones which have been spatially filtered to obtain
beam data is heterodyned 120'' with a local signal that has the
negative of the frequency shifts of the received signal to generate
a fixed frequency signal. Any local signal that has the negative of
the transmitted frequency shifts may be used. A convenient local
reference is the transmitted signal 90''. Heterodyning with a real
transmit signal creates both sum and difference frequencies with
the desired difference frequency out of the heterodyne block
selected by the following frequency filtering. With complex
(analytic) signals using the conjugate 95'' of the transmit replica
yields only the difference frequency out of the heterodyne stage.
Frequency filtering with low side bands is achieved by using a
Fourier transform 125'' with a shading window which provides
tapering selected for low side-bands. Many low side band windows
are available, for instance Chebychev or Taylor or Kaiser-Bessel
windows may be generated with any desired theoretical side lobe
level. The output of the above processing chain (as processed in
element 130'' of FIG. 2b) for a linear FM transmit sweep with data
captured in a lake test on a single hydrophone is shown below in
FIG. 2c. It will be noticed that greater than eighty decibels of
side band rejection 160 is achieved with frequency filtering alone.
The data corresponding to reverberation and direct signal are shown
by reference numerals 170 and 180, respectively.
[0058] An analysis of a direct signal rejection requirement is
provided by applying Equation II to the same analysis used in
Equation I above for determining the linearity requirement. Using
the values of the previous analysis, namely, LS=225 dB, YN=50 dB,
TL=40 dB, NL=50 dB, NDI=25 dB, RS=35 dB, and applying both the
spatial and frequency filtering so that RSR=10 dB and RFR=70 dB,
Equation II yields: 1=20 dB.ltoreq.25 dB thus, meeting the
rejection requirement with the direct signal being 5 dB less than
the noise level of the receiver.
[0059] In another analysis of an application of the present
invention, a bi- and multi-static operation at mid-frequencies is
provided. In an aspect of the invention, there is disclosed
standalone sonar processing capability using off-board sources and
fixed sources, such as a sonobouy. The primary difference from the
analysis above is that spatial rejection at the receiver is
minimized due to the use of a single element sonobuoy although
directionality is obtained by processing orthogonal channels in
dipole or cardioid mode. Similarly a directional transmission to
reduce levels toward the receiver is difficult due to the fact they
operate in a field. Therefore, assuming, for this analysis, LS=200,
TN=0 dB, NL=40 dB, NDI=5 dB, RS=20 dB, TL=55 due to farther spacing
between source and receiver, RSR=10 dB for adaptive spatial
processing, and that a reasonably clean frequency transmission and
frequency filtering can achieve RSR=85 dB, the interference level
from Equation II becomes I=30 db<35 dB Hence the direct signal
would be a resultant 5 dB less than the noise level of the receiver
and this requirement is again attainable.
[0060] In an experiment that was conducted in a lake using the
frequency domain rejection technique described above, it was shown
that echoes could be received while transmitting even without
transmit beamforming or receive array directivity. The experiment
involved analyzing data from a single, non-directional hydrophone
placed so that it received the maximum, rather than the "nulled"
transmission from a low frequency (1 kHz) acoustic source only
about 15 meters away. FIGS. 2c and 2d illustrate the receive
heterodyne output level for 10 consecutive 15 second swept
transmissions stacked up vertically from 850-1150 Hz. The FIG. 2c
is normalized to show defections above the local noise level. The
shoreline edge of lake can be seen clearly at 2 kilometers and 3.5
kilometers. FIG. 2d shows absolute receiver levels and indicates
that minimum detectable level of the ambient noise can be reached
and occurs at about 6.5 km during each transmission.
[0061] In an aspect of another embodiment, in order to allow the
target signal to be separated from the directly received
transmitted signal with frequency filtering, the spectral energy of
the transmitted signal is not allowed to leak into the received
frequency band. A window is used to taper the onset and termination
of transmission segments where abrupt changes in transmission
frequency would normally cause transmission energy to spread over a
wide frequency band. By smoothly tapering the onset and termination
amplitudes of each segment from zero to maximum amplitude the
frequency spreading may be controlled. A typical tapering for four
transmit segments is shown in FIG. 3a where the taper is computed
from the integral of a Chebychev window. The images in the Figure
show the effect of tapering on the signal's spectra-gram and on a
typical final processed display. FIG. 3a illustrates the amplitude
of the signal with 210 and without 200 tapering. FIGS. 3b and 3c
show the spectrum over time without and with the signal tapered,
respectively. It is shown that the tapering reduces the
interference 220 in frequency induced by the signal discontinuity.
In both FIGS. 3b and 3c the spectrum scale is shown on the vertical
tape 230. FIGS. 4a and 4b show the effect on an illustrative
detection display again without and with proper signal tapering,
respectively, to minimize signal interference. Both figures show
heterodyne display.
[0062] An aspect of a high data rate tracking component 140 of the
sonar system shown in FIG. 1b provides information rate detection
gains which are not attainable conventionally. Detection,
localization and tracking entails the processing of acoustic
measurements obtained from a target over time to estimate its
current state which can contain both kinematic and feature
components. The kinematic components may consist of position,
velocity, and acceleration, while analysis of feature component are
radiated signal strength, spectral characteristics, and target
highlight characteristics useful for classification. The sonar
measurements are noise corrupted observations related to the state
of the target or attributes of the target. In both passive and
active sonar applications measurements are received from the sensor
that do not originate from a target of interest due to several
types of interfering sources. Acceptable false alarm rates are
rarely obtained using a single set of measurements In time hence,
track before detect techniques are required. Reduction in the time
between observations greatly reduced uncertainty of target location
and dramatically improves detection performance.
[0063] The results that are shown in FIG. 5 for a target 60 moving
at 3 knots relative to a receive platform 20 shown in FIG. 1a
demonstrate the active sonar detection and localization that may be
achieved in steps 130 and 140 shown in FIG. 1b. In achieving the
results shown below, such system provides reliable and rapid
decisions about the presence of a contact with a target while
attempting to reject a high number of false alerts due to
reverberation backscattering that appear spatially correlated from
ping to ping. Reducing the time between observations greatly
improves reliable track confirmation. Due to the uncertainty of
target motion, multiple observation association requires track
filtering and track gating to spatially locate the correct
measurement at each observation time for use in track updating and
the track decision process. The uncertainty of echo location
increases proportionally to the time between observations.
[0064] Conventional pulsed systems require 12-20 pings (2-5 minutes
depending on the range coverage) to make a reliable decision. As
would be understood, embodiments of the invention provide near
continuous updates of information, with time between measurements
reduced by an order of magnitude from pulsed systems, i.e. 10 times
as many looks at a target than a pulse system in the same fixed
period of time. This information rate gain and the advantages of
reduced uncertainty in echo location from ping to ping (track
gates) may lead to a 12 dB reduction in detection threshold.
[0065] The detection performance gains using continuous
transmission over conventional pulsed sonar operating over the same
transmission band are due to the increased rate of target
information acquired. The improved time to detect for a 10 dB
contact at 10 kiloyards moving at 3 knots relative to the receive
platform is shown in FIG. 5 mentioned above. Curve 240 represents
the performance of the continuous, that is,
"receive-while-transmit" system of the present invention compared
with a curve 250 showing the performance of a conventional, pulsed
system, each operating at 10 dB. The probability of detection,
PD.about. plotted as a function of time shows that a level of
detection P.sub.D=0.5 is achieved in 20 seconds 245 for the
continuous transmission approach 240 whereas the pulsed system 250
requires 300 seconds (5 minutes) 255 to detect. Furthermore, to
achieve an equivalent level of detection a pulsed system would
require >25 dB SNR, which is 15 dB more than that required by
the continuous transmission approach.
[0066] It will be evident to those skilled in the art that the
instant method of active "receive-while-transmit (RWT)" provides
substantial performance improvements over the conventional sonar:
recognition differential gains of 7 dB in reverberation and 20 dB
in noise are achieved along with the advantage of detecting a
target 15 times earlier. As is shown in the analysis (for the
example of FIG. 5) below, a continuous transmission over a
frequency spectrum of 1 KHz provides rapid motion discrimination
for targets moving at less than 1 knot with acoustic detection
improvements in excess of 20 dB for submerged contacts in shallow
water noise limited conditions. The analysis shown below provides a
specific quantitative analysis of how the instant sonar system can
substantially outperform conventional active sonar. Unlike
conventional pulsed active transmission methods, the instant method
herein requires a separate active broadband source and a receiver
designed to process active target reflections in such a manner as
to make the target appear as a passive acoustic source.
[0067] An embodiment for continuous transmission provides a means
of constantly ensonifying a target while simultaneously reducing
reverberation energy by varying frequency continuously as a
function of time. An aspect of this continuous transmission scheme
makes the target look like a constantly-emitting noisy passive
acoustic source. Another aspect utilizes the consistency of the
target scattering strength received over a broad frequency spectrum
and the inconsistencies of backscatter from reverberation as a
function of frequency. Still another aspect of this effectively
large bandwidth approach provides computational simplicity of the
signal processing, flexibility to achieve various levels of time
and frequency diversity, detection gains in shallow water
reverberation, and rapid assessment of target Doppler at any level
for motion discrimination of both the high Doppler target and the
slowly moving submerged threat; all culminating in a substantially
improved performance over conventional active pulsed transmission
sonar systems.
[0068] An aspect of an embodiment of the present
transmission/reception method provides a combined effect of
reverberation suppression obtained by extending the transmit energy
over a wide frequency band and the reduced detection time obtained
by utilizing persistent information received over time. The latter
allows for averaging of the return energy over different frequency
bands. The averaging gain of this approach is difficult to realize
in conventional pulsed active systems. Furthermore, the broadband
approach provides high resolution in range and, due to the
continuous nature of reception, range rate (Doppler) is estimated
rapidly to provide high fidelity target motion discrimination. For
illustrative purposes, this analysis uses a single platform
operation, however, the present method is suitable for multiple
platform multi-static operation with any receiver platform
possessing a passive narrowband detection capability.
[0069] An aspect of an embodiment method as applied to the analysis
of the slow moving target less than, but not limited to 1 knot,
provides modification of the Frequency Modulated Continuous Wave
(FM-CW) principles to work at low frequencies (LF); i.e., the
transmitted waveform bandwidth can exceed the center frequency for
this underwater acoustic detection problem. The transmission and
the receiver processor are designed for a continuous linear
frequency sweep over a large bandwidth. The sweep Es continuously
repeated (due to a practical limit on total available bandwidth) as
illustrated in FIG. 6, effectively producing a saw tooth pattern in
instantaneous frequency. FIG. 6 shows the time delay, .tau., 275
between the low frequency modulated transmissions 260 and the
return echo 260, where the transmit and wave fauns are plotted in
terms of frequency f=r.tau. along the horizontal axis, and time, T,
the vertical axis.
[0070] The received data shown in FIG. 6 is complex-heterodyned
(multiplied) with the transmit waveform and low pass filtered to
obtain the resulting beat frequency (difference frequency). That
is, the heterodyne processing produces beat frequency data
corresponding to time delays. The minimum frequency is zero and the
maximum beat frequency is equivalent to the transmitted bandwidth B
given by r.sub.max.tau..sub.max=B where r is the sweep rate, and
.tau., time Tactical sonar requires removal of own-ship motion
using both Doppler compensation on received beam data as well as a
correction for geographic displacement over time. The own-ship
Doppler correction for each receive beam is accomplished in the
front-end heterodyne processing. The geographic displacement is
performed as a time alignment in the passive narrowband (PNB)
prediction function at the Fast Fourier Transform (FF1) output
level.
[0071] By proper complex processing the saw tooth jump in the real
transmission is transparent in the heterodyned output. For a return
echo then, without a Doppler shift, the heterodyned output is a
narrowband tonal with beat frequency, f.sub.b=r.tau. The parameter,
r, is the known sweep rate and t is the two-way travel time which
provides a measure of the target range. The output of the
heterodyne process is depicted in FIG. 7, which shows the results
obtained for a stationary target 280, closing target 290 and
maneuvering target 300.
[0072] The continuous nature of the transmission reduces the active
sonar detection problem to one of detecting a narrowband tonal in
nonstationary noise. This allows the leveraging of passive
narrowband (PNB) automatic detection and tracking. The
nonstationary nature of the noise background in the active
situation is due to the time-varying reverberation level. This
contrasts with the pure ambient noise limited conditions most often
encountered in passive sonar. However, with proper noise estimation
and normalization techniques the background may be equalized and
PNB techniques applied directly.
[0073] An aspect of an embodiment involves bandwidth availability.
The latter dictates the tactical range coverage possible and also
the amount of reverberation suppression obtainable. The repetition
rate, T, and the maximum sweep rate r, for continuous transmission
are determined by the system bandwidth, B, and the desired maximum
target range coverage. The constraint on the repetition rate is
similar to a pulsed active sonar where T .gtoreq. .tau.
max=2.times. R max c must be satisfied to avoid range ambiguity.
The sweep rate is constrained by the system bandwidth, the maximum
sweep rate occurring when T is equal to the delay at maximum range
coverage .tau. max=B T
[0074] The sweep rate may be maximized to obtain the maximum
reverberation suppression possible in a fixed amount of receive
time and to maximize the time delay (range) resolution. Range
resolution is inversely related to the observation time at the
receiver, T.sub.r, and the sweep rate used. Therefore the maximum
range coverage is determined by the transmitted bandwidth and the
sweep rate R max=c .times. .times. .tau. max 2=cB 2 .times. r max
In another aspect, the processing of the heterodyned output is
designed to obtain the optimal detection performance for the
expected target motion in all tactical scenarios.
[0075] An aspect of an embodiment of the present
"receive-while-transmit (RWT)" invention provides detection SNR
thresholds 7 dB lower In reverberation and 20 dB under Noise
limited conditions than those currently achieved by conventional
pulsed active systems. These performance gains of "RWT"
transmission over a conventional pulsed sonar operating in a
reverberation limited and noise limited environment can be obtained
by examining the required recognition differential (RD) for each
case. RDI accounts for the different processing gain (PG) of each
method by removing the gain from the detection index (DI) required
to achieve similar detection performance N.sub.RD=DT-PG
[0076] The detection performance is obtained using the plots shown
in FIG. 8 where curve 310 is for a conventional 15 dB pulsed system
and curve 320 is for 3 dB continuous system. The probability of
detection of 0.9 after 5 minutes with a false alarm probability,
P.sub.f, of 10.sup.-8 was chosen for each method. The detection
index for the continuous transmission is 10*log(d.sub.ct)=3 dB and
for the pulsed method, 10*log(d.sub.p)=15 dB. The only processing
gain in reverberation for the pulsed system is due to the
semicoherent averaging gain and the reverberation suppression per
pulse N RD p=10 .times. log .function. (dp) -5 .times. log
.function. (N) -10 .times. log .function. (Tp)=10 .times. log
.function. (dp) -5 .times. log .function. (10) -10 .times. log
.function. (0.2)=-17 .times. .times. dB where 10 is the number of
subpulses. The modified detection threshold for the continuous
transmission method is adjusted to account for the continuous
transmission receiver. Here the only adjustment to the detection
threshold is due to bandwidth in each 4 second processed segment. N
RD ct=10 .times. log function. (dct) -10 .times. log function.
(BW)=10 .times. log .function. (dct) -10 .times. log .function.
(100)=-17 .times. .times. dB hence in reverberation the continuous
transmit gain is .DELTA. .times. .times. N RD Raw=7 .times. .times.
dB
[0077] The detection gains against noise are due to the increased
energy in the continuous transmission which is reflected in the
incoherent averaging gain achievable. The recognition differential
in noise for the pulse system reflects the pulse length per
sub-pulse T.sub.p so that N RD p=10 .times. log .function. (dp) -5
.times. log .function. (N) -10 .times. log .function. (BW)=10
.times. log .function. (dp) -5 .times. log .function. (10) -10
.times. log .function. (100) -10 .times. .times. dB where for the
continuous transmission the processing pulse length is 4 second and
the gain in integration over the entire 40 seconds is captured in
the detection threshold so N.sub.RD.sup.et=10 log(d.sub.et)-10
log(4)=-3 dB Therefore a detection threshold gain in noise for the
continuous transmit method is .DELTA..times. .times. N RD noise=20
.times. .times. dB
[0078] Another aspect of significant gain in performance of
continuous transmission over conventional pulsed active sonars
resulting from the increased rate of target information is the
substantially more accurate, and more rapidly obtained target
motion estimates as shown in FIG. 9 .about. As seen in the same
Figure, continuous transmit (curve 330) provides accuracy of target
speed to a root-mean-square (rms) level of 0.5 knots (on the
vertical axis) in less than 2 minutes (on the horizontal axis)
where the pulsed system (curve 340) provides an rms error greater
than 2 knots in the 5 minute receive time (345). Due to the
accuracy of motion estimation, classification decisions based upon
motion clues can be made far more rapidly and with greater accuracy
using the methods of the present invention.
[0079] Though these numerous details of the disclosed methods and
devices are set forth here, such as various parameters, to provide
an understanding of the present invention, it will be obvious,
however, to those skilled in the art that these specific details
need not be employed to practice the present invention. At the same
time, it will be evident that the same methods and devices may be
employed in other similar situations that are too many to cite,
such as the use of the methods cited herein with receivers having
single or more multiple hydrophones.
[0080] Those skilled in the art will also know that although the
continuous transmit concept itself--a subset of the present
"Receive While Transmit (RWT)"--is used in a number commercially
available Continuous Transmit Frequency Modulated (CTFM) Sonar
systems, they cannot operate at certain frequencies (see for
example, Skolnik, M. I., Introduction to Radar Systems, McGraw
Hill, New York, 1962). Whereas many pulsed SONARS can operate at
frequencies below 10 thousand cycles-per-second (10 kHz) current
SONAR CTFM modes are confined to much higher frequencies. Operating
frequencies for existing CTFM systems range from above thirty
thousand (30 kHz) to in excess of three hundred thousand
cycles-per-second (300 kHz). Although these high frequencies offer
spatial resolution suitable to their purpose it is noted that no
low frequency systems exist. This is because existing systems rely
on a combination of highly directional transmitters and high
in-water absorption loss to avoid overloading their receivers while
transmitting, and high frequency sweep rates to facilitate
frequency separation between transmitted signal and returned
echoes. All of these conditions require high operating frequencies.
Since absorption of sound in the water increases (and therefore
operating range decreases) in proportion to the square of the
operating frequency (see for example, R. 3. Urick, Pr/nap/es of
Underwater Sound, McGraw Hill, New York, 1975) CTFM SONARS are
limited to short range applications such as object avoidance or
retrieval, and bottom or overhead under Ice mapping. It will
therefore be evident to those skilled in the art that modern high
dynamic range electronics combined with the disclosed frequency
domain filtering techniques overcomes the limitations of the
present state of the art and enables a broad range of applications
including low frequency long range search sonars such as those used
in anti-submarine warfare.
[0081] While the invention has been particularly shown and
described with reference to particular embodiments, those skilled
in the art will understand that various changes in form and details
may be made without departing form the spirit and scope of the
invention.
* * * * *