Patent Grant
RE45379
1. Field of the Invention
The invention relates to underwater sound technology and, in particular, concerns sonar systems with multiple acoustic beams being formed and steered by frequency division.
2. Description of the Related Art
Nearly all under-water vehicles, whether manned or unmanned, are equipped with an ahead-look sonar (ALS). Common applications include obstacle avoidance, mine detection, and rendezvous and docking capabilities. In general, these sonars use electronically beamformed arrays, although some use simpler systems in which multiple, separate directional hydrophones are used to provide multiple preformed beams. The electronically beamformed systems offer improved performance but are substantially more costly because of the complexities of the beamforming circuitry. A typical multi-beam sonar forms about 100 beams over an angular sector of about 150 degrees. The beamforming circuitry for this sonar requires about 100 receiver channel amplifiers to raise the received signal to a level sufficient for digital beamforming of the 100 beams. Because of this receiver amplifier complexity which is proportional to the beam resolution (beamwidth) and the number of beams formed, the multiple physical beam approach is typically limited to a relatively small number of low resolution beams and may be relatively heavy because of the excess of ceramic material in the multiple hydrophones.
Another method of forming multiple beams without an electronic beamformer is the use of an acoustic lens with a multi-element retina of small hydrophones. Although appealing in principle, lens arrays have proven difficult in practice due to issues such as temperature instability and toxicity of fluid materials and shear wave effects in solid lenses. Lens sonars also have problems in terms of the size and weight of the physical beamformer.
Thus, present beamforming techniques have practical cost, size and weight deficiencies. These are particularly important in applications such as small, low cost unmanned under-water vehicles (UUVs) where size and weight are at a premium and the cost of individual subsystems such as sonars preferably needs be kept low.
A frequency scanning technique has been used in radar for many years. Instead of fixed phase shifts between elements, however, the radar implementations use long delay lines between antenna elements or radiating slots in a dispersive delay line. The typical application is to provide vertical scanning of an array where azimuthal scanning is provided by either mechanical rotation or another electronic phase shifting technique.
Hence, there is a need for a sonar system that permits sequential scanning through multiple beams or forming multiple simultaneous acoustic beams. There is need for such sonar system to be implemented in a simple cost and size/weight effective manner.
In one aspect, the aforementioned needs are satisfied by a sonar system for forming a steerable underwater acoustic beams. The system comprises an array of acoustic transducers and a beamforming system that associates a signal to each of the transducers to form an acoustic beam with a direction. The signal is phase shifted by a selected fixed amount relative to a signal assigned to the adjacent transducer and the direction of the acoustic beam is determined by the frequency of the signals. The beamforming system is adapted to vary the frequency of the signals so as to permit steering of the acoustic beam.
In one embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam. In another embodiment, the beamforming system comprises a receiver that receives signals from the array that results from a received acoustic beam. In another embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam, and a receiver that receives signals from the array that results from a received acoustic beam.
In one embodiment, a formula cos θ=(Δφ/2π)(c/fd) represents a relationship between the direction of the acoustic beam and the frequency, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents a phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, f represents the frequency of the signals, and d represents spacing between the adjacent transducers, wherein the phase shift Δφ is selected to be a substantially constant value andyand the direction angle θ is varied by varying the frequency about a center frequency f0. The phase shift Δφ is selected such that a signal associated with a given acoustic transducer is a simple linear combination of signals proportional to cos ωt and sin ωt, where ω=2πf and t represents time. In one implementation, the phase shift Δφ between the adjacent acoustic transducers is selected to be approximately π/2 radian such that repeating sets of four acoustic transducers can be associated by a sequence of signals proportional to cos ωt, sin ωt, −cos ωt, and −sin ωt. In another implementation, the phase shift Δφ between the adjacent acoustic transducers is selected to be approximately 3π/4 radian such that repeating sets of eight acoustic transducers can be associated by a sequence of signals proportional to cos ωt, −1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −sin ωt, 1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −cos ωt , 1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt, sin ωt, and −1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt. In one implementation, the frequency f of the signals is varied in a range of approximately 0.75f0 to approximately 1.25f0.
In another aspect, the aforementioned needs are satisfied by an underwater sonar system comprising an array of acoustic transducers and a beamforming system that simultaneously associates signals with a range of frequencies to the transducers. A signal to a given transducer is phase shifted by a selected fixed amount relative to a signal assigned to the adjacent transducer. The phase shifted signals with the range of frequencies form an acoustic signal with a range of directions. A given direction of propagation within the range of directions corresponds to a specific frequency of the signals within the range of frequencies.
In one embodiment, the beamforming system comprises a broadband transmitter that simultaneously supplies signals with a range of frequencies to the array so as to form transmitted acoustic signals with a range of directions. In another embodiment, the beamforming system comprises a receiver having a spectrum analyzer that simultaneously processes signals from the array that result from received acoustic signals from a range of directions. In another embodiment, the beamforming system comprises a broadband transmitter and a receiver having a spectrum analyzer. The broadband transmitter simultaneously supplies signals with a range of frequencies to the array so as to form transmitted acoustic signals with a range of directions, and the spectrum analyzer simultaneously processes signals from the array that result from received acoustic signals from a range of directions.
A formula cos θ=(Δφ/2π)(c/fd) represents a relationship between the direction of the acoustic signal and the frequency, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents a phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, f represents the frequency of the signals, and d represents spacing between the adjacent transducers. The phase shift Δφ is selected to be a substantially constant value and the direction angle θ is varied by varying the frequency f. Preferably, the phase shift Δφ is selected such that a signal associated with a given acoustic transducer is a simple linear combination of signals proportional to cos ωt and sin ωt, where ω=2πf and t represents time. In one implementation, the phase-shift Δφ between the adjacent acoustic transducers is selected to be approximately π/2 radian such that repeating sets of four acoustic transducers can be associated by a sequence of signals proportional to cos ωt, sin ωt, −cos ωt, and −sin ωt. In another implementation, the phase shift Δφ between the adjacent acoustic transducers is selected to be approximately 3π/4 radian such that repeating sets of eight acoustic transducers can be associated by a sequence of signals proportional to cos ωt, −1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −sin ωt, 1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −cos ωt, 1/√{square root over (2)} cos ωt −1/√{square root over (2)} sin ωt, sin ωt, and −1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt.
In yet another aspect, the aforementioned needs are satisfied by a method of using an underwater sonar system having an array of acoustic transducers. The method comprises associating signals having a frequency component to the transducers. A signal associated with a given transducer is phase shifted by a selected fixed amount relative to a signal assigned to the adjacent transducer such that the phase shifted signals form an acoustic beam having a direction. The method further comprises controlling the directionality of the acoustic beam by manipulating the frequency component of the signals.
In one implementation, associating the signals to the transducers comprises associating the transducers with signals with a frequency f such that a formula cos θ=(Δφ/2π) (c/fd) represents a relationship between the direction of the acoustic beam and the frequency, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents the selected fixed phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, and d represents spacing between the adjacent transducers. Preferably, associating the signals to the transducers comprises selecting the phase shift Δφ such that a signal associated with a given transducer is a simple linear combination of signals proportional to cos ωt and sin ωt, where ω=2πf and t represents time. The phase shift Δφ between the adjacent acoustic transducers may be selected to be approximately π/2 radian such that repeating sets of four acoustic transducers can be associated by a sequence of signals proportional to cos ωt, sin ωt, −cos ωt, and −sin ωt. Alternatively, the phase shift Δφ between the adjacent acoustic transducers may be selected to be approximately 3π/4 radian such that repeating sets of eight acoustic transducers can be associated by a sequence of signals proportional to cos ωt, −1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −sin ωt, 1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −cos ωt, 1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt, sin ωt, and −1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt.
In one implementation, associating the signals with the transducers comprises associating a narrowband signal with the transducers and varying the frequency of the narrowband signal to change the direction of the acoustic beam. Associating the narrowband signal with the transducers may comprise supplying the narrowband signal to the transducers wherein the signal applied to the transducers results in an outgoing acoustic beam. Alternatively, associating the narrowband signal with the transducers may comprise receiving an echo signal from the transducers wherein the echo signal result from an echo that impinges on the transducers. Alternatively, associating the narrowband signal with the transducers may comprise supplying the narrowband signal to the transducers to yield an outgoing acoustic beam, and receiving an echo signal from the transducers that result from an incoming echo.
In another implementation, associating the signals with the transducers comprises associating a broadband signal having a range of frequencies with the transducers such that corresponding acoustic beams have a range of directions. Associating the broadband signal with the transducers may comprise simultaneously providing a broadband signal to the transducers so as to yield a plurality of outgoing acoustic beams having a range of directions. Alternatively, associating the broadband signal with the transducers may comprise simultaneously receiving a broadband echo signal from the transducers that result from a plurality of incoming echoes. Alternatively, associating the broadband signal with the transducers may comprise simultaneously proving a broadband signal to the transducers to yield a plurality of outgoing acoustic beams having a range of directions, and simultaneously receiving a broadband echo signal from the transducers that result from a plurality of incoming echoes.
In yet another aspect, the aforementioned needs are satisfied by a method of scanning an angular sector underwater using an array of acoustic transducers. The method comprises forming a plurality of acoustic beams wherein each acoustic beam is formed by associating signals to the array of acoustic transducers such that a signal associated a given transducer is phase shifted by a selected fixed amount relative to a signal assigned to the adjacent transducer. The direction of each acoustic beam depends on the frequency of the signals. The method further comprises varying the frequency of signals corresponding to each acoustic beam so as to vary the direction of the acoustic beam, thereby allowing the acoustic beam to sweep a range of direction angles. The frequency is selected for each acoustic beam such that resulting ranges of direction angles cover the angular sector.
In one implementation, a formula cos θ=(Δφ/2π)(c/fd) represents a relationship between the direction of the acoustic beam and the frequency f, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents the selected fixed phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, and d represents spacing between the adjacent transducers. Preferably, the phase shift Δφ is selected such that a signal associated with a given transducer is a simple linear combination of signals proportional to cos ωt and sin ωt, where ω=2πf and t represents time. The phase shift Δφ between the adjacent acoustic transducers may be selected to be approximately π/2 radian such that repeating sets of four acoustic transducers can be associated by a sequence of signals proportional to cos ωt, sin ωt, −cos ωt, and −sin ωt. Alternatively, the phase shift Δφ between the adjacent acoustic transducers may be selected to be approximately 3π/4 radian such that repeating sets of eight acoustic transducers can be associated by a sequence of signals proportional to cos wt, −1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −sin ωt, 1/√{square root over (2)} cos ωt+1/√{square root over (2)} sin ωt, −cos ωt, 1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt, sin ωt, and −1/√{square root over (2)} cos ωt−1/√{square root over (2)} sin ωt.
In one implementation, associating the signals with the transducers comprises associating a narrowband signal with the transducers and varying the frequency of the narrowband signal to sweep the acoustic beam within the range of direction angles. Associating the narrowband signal with the transducers may comprise supplying the narrowband signal to the transducers wherein the signal applied to the transducers results in an outgoing acoustic beam. Alternatively, associating the narrowband signal with the transducers may comprise receiving an echo signal from the transducers wherein the echo signal result from an echo that impinges on the transducers. Alternatively, associating the narrowband signal with the transducers may comprise supplying the narrowband signal to the transducers to yield an outgoing acoustic beam, and receiving an echo signal from the transducers that result from an incoming echo.
In yet another aspect, the aforementioned needs are satisfied by a sonar system for forming a steerable underwater acoustic beams. The system comprises an array of acoustic transducers and a beamforming system that associates a signal to each of the transducers to form an acoustic beam with a direction. The signal is phase-shifted by a selected phase relative to a signal assigned to the adjacent transducer and the direction of the acoustic beam is determined by a combination of the phase and the frequency of the signals. The beamforming system is adapted to vary the frequency of the signals for a given phase so as to permit steering of the acoustic beam.
In one embodiment, a formula cos θ=(Δφ/2π)(c/fd) represents a relationship of the direction of the acoustic beam to phase and frequency, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents a phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, f represents the frequency of the signals, and d represents spacing between the adjacent transducers, wherein the phase Δφ is selected to direct the beam in a general desired first direction, and the frequency f is varied to vary the direction of the beam about the first direction.
In one embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam. In another embodiment, the beamforming system comprises a receiver that receives signals from the array that results from a received acoustic beam. In another embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam, and a receiver that receives signals from the array that results from a received acoustic beam.
In yet another aspect, the aforementioned needs are satisfied by a method of using an underwater sonar system having an array of acoustic transducers. The method comprises associating signals having a frequency component and a phase component to the transducers. A signal associated with a given transducer is phase-shifted by a selected phase relative to a signal assigned to the adjacent transducer. The method further comprises controlling the directionality of the acoustic signal by selecting a first direction of the acoustic signal as determined by the selected phase and varying the direction of the acoustic beam about the first direction by manipulating the frequency component of the signals.
In one implementation, associating the signals to the transducers comprises associating the transducers with signals with a frequency f and a the phase Δφ such that a formula cos θ=(Δφ/2π)(c/fd) represents a relationship of the direction of the acoustic signal to the phase and frequency, where θ represents a direction angle relative to a plane defined by the transducers, Δφ represents the selected phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, and d represents spacing between the adjacent transducers.
In yet another aspect, the aforementioned needs are satisfied by a sonar system for forming a steerable underwater acoustic beams. The system comprises an array of acoustic transducers and a beamforming system having a set of beamformers. The beamformers associate a plurality of signals to the transducers to form an acoustic beam with a direction. Each of the signals is phase-shifted by a selected phase relative to a signal assigned to the adjacent transducer. The direction of the acoustic beam is determined by a combination of the phase and the frequency of the signals. The beamforming system is adapted to vary the frequency of the signals for a given phase so as to permit steering of the acoustic beam. A subset of the beamformers is connected to more than one repeating subsets of the transducers such that each beamformer associates a signal having an assigned phase and frequency to more than one transducer. This allows the total number of beamformers to be less than the number of transducers in the array.
In one embodiment, a formula cos α=(Δφ/2π)(c/fd) represents a relationship of the direction of the acoustic beam to phase and frequency, where α represents a direction angle relative to a normal to a plane defined by the transducers, Δφ represents a phase shift between adjacent acoustic transducers, c represents velocity of the acoustic beam, f represents the frequency of the signals, and d represents spacing between the adjacent transducers. The phase Δφ is selected to direct the beam in a general desired first direction, and the frequency f is varied to vary the direction of the beam about the first direction. The phase Δφ is selected to be an integral fraction of 2π radians to allow repeated duplication of signal assignments of the subset of the beamformers to the more than one subsets of the transducers.
In one embodiment, the array of transducers comprises a first line array. The spacing d is selected to be approximately half of the wavelength, and the phase Δφ is selected as 0, π/8, π/4, and π/8 radians progressively so as to allow progressive scanning about the different first directions as determined by the selected phases. The frequency is varied at each of the selected phases by approximately 67% of bandwidth about a center frequency such that the resulting sweepings of the beam about the first directions yield a generally seamless coverage of scanning that has a range of approximately 0 to 41.8 degrees with respect to the normal.
In one embodiment, the sonar system further comprises a second line array oriented perpendicularly to the first line array so as to form a cross shape to allow scanning in two dimensions.
In one embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam. In another embodiment, the beamforming system comprises a receiver that receives signals from the array that results from a received acoustic beam. In yet another embodiment, the beamforming system comprises a transmitter that supplies signals to the array so as to form a transmitted acoustic beam, and a receiver that receives signals from the array that results from a received acoustic beam.
Reference will now be made to the drawings wherein like numerals refer to like parts throughout. The description below describes the systems and methods of frequency division beamforming and its use to form multiple transmitting and/or receiving beams for sonar applications.
The exemplary transducers 106a-d are supplied with corresponding signals 120a-d from the beamformer such that the signals 120a-d have a common frequency f and selected phases such that neighboring transducers have a phase difference of Δφ. For example, the signal 120b has a phase of φ, while the preceding 120a has a phase of φ−Δφ and the following signal 120c has a phase of φ+Δφ. The transducers 106 are spaced apart by a spacing d. Given such a configuration, the transducers 106a-d, in response to the phased signals 120a-d, emit waves represented by respective wavefronts 126a-d. The wavefronts combine to form a plane wavefront 128 that propagates as an acoustic beam 122 having a direction angle θ with respect to a line that extends through the transducers 106. A relationship between the direction angle θ and the aforementioned signal and array parameters are described below in greater detail. Alternatively, the direction of the propagating beam may be expressed with respect to a line 124 normal to the plane of the transducers.
It will be understood that the term “beam” used herein may refer to a sound wave propagating in a well defined direction, or may also refer to a portion of a sound wave pattern wherein the portion propagates in along a given direction. Furthermore, while the acoustic beam 122 is depicted as being comprised of a plane wave, it will be understood that dispersion of the beam may occur due to various reasons. It will be appreciated, however, that for the purpose of describing the concept of controlling the directionality of the acoustic beam, such simplified illustration suffices without departing from the spirit of the invention.
Conventional beamforming applies time delays to the elements of an acoustic array which compensate for the differences in propagation time of an acoustic wave in the water. Time delay is equivalent to applying a phase shift which is a linear function of frequency to each element. In frequency division transmit beamforming, each element of a uniformly spaced line array is driven by a signal which leads or lags its nearest neighbor by a fixed phase shift, Δφ. For frequency division receive beamforming, the received signal from each element of a uniformly spaced line array is phase shifted so it leads or lags its nearest neighbor by a fixed phase shift, Δφ. This causes the transmitting and/or receiving beam steering direction to become a function of frequency which makes it possible to scan a transmitting and/or receiving beam through a range of angles by changing its frequency. The same principle can be used to form multiple simultaneous transmitting and/or receiving beams. This is accomplished by transmitting a wide bandwidth signal and receiving the echoes through a spectrum analyzer. Each frequency bin of the spectrum analyzer then corresponds to a beam pointing in its own unique direction.
Frequency division beamforming is practical when a bandwidth available in the transducers is substantially larger than what is required to achieve the desired range resolution. This excess bandwidth is then used for beam steering. For an echo ranging sonar, range resolution is related to bandwidth by
ΔR=c/2B. (1)
For a 10 cm. (4 in.) resolution cell which would be appropriate for a typical obstacle avoidance sonar, 7.5 kHz of bandwidth would be required. In one commercially available transducer operating at 300 kHz, available bandwidth is about 150 kHz. Thus, the total bandwidth can be divided into 20 segments, each of which has the required 7.5 kHz bandwidth. Thus such a transducer has a substantial excess bandwidth which can be used for beamforming. One embodiment of the invention is an array in which the beam points in different directions at different frequencies. As long as there is 7.5 kHz of bandwidth available in any given direction, such an array can use the total 150 kHz bandwidth of the transducers to form 20 distinct beams and still achieve adequate range resolution along each beam.
To understand how frequency division transmitter beamforming works, consider first a true time delay beamformer for a uniform line array of elements with spacing, d. Each element is driven with a signal
sn(t)=s(t−τn), (2)
where τn=nτ=(nd/c)cos θs, and θs is the desired beam steering direction. If s(t)=a(t)cos ωt, then sn(t)=a(t−τn)cos ω(t−τn). For relatively narrow band signals, the time delay in the envelope can be ignored. The driving signals are then approximated by sn(t)=a(t) cos(ωt−nΔφ). Ideally,
Δφ=2π(fd/c)cos θs. (3)
Thus, the desired phase shift is a linear function of frequency. However, if the phase shift is selected as a fixed quantity which is independent of frequency, the beam steering direction is given by
cos θhd s=(Δφ/2π)(c/fd). (4)
Thus, as frequency increases, cos θs decreases, and θs increases from endfire toward broadside. Assuming that the transducers have a fractional bandwidth, β, beam steering angle cover the range
(Δφ/2π)(c/f0d)/(1+β/2)≦cos θs≦(Δφ/2π)(c/f0d)/(1−β/2), (5)
where f0 is the center frequency of the transducer. To avoid unwanted grating lobes, a condition can be imposed such that d≦λ/2. Applying this criterion to the center frequency, f0, leads to the inequality
(Δφ/π)/(1+β/2)≦cos θs≦(Δφ/π)/(1−β/2), (6)
It should also noted that for real steering angles, cos θs≦1. Therefore,
Δφ≦π(1−β/2). (7)
In one embodiment, the fractional bandwidth β=0.5, thus yielding Δφ≦3π/4 or 135°.
As described above in reference to
The essence of frequency division beamforming is that if each element of a uniformly spaced line array is driven by a signal which leads or lags its nearest neighbor by a fixed phase shift, Δφ, the beam steering direction becomes a function of frequency according to Equation 4. An example of this effect is shown in
Consider the case of Δφ=π/2 or 90°. If d=λ/2 at the center frequency, f0, then the beam points 60° away from endfire (30° from normal) according to Equation 4. At the low end of the 50% passband, f=0.75f0, and the beam angle is 48.2°, while at the upper end, f=1.25f0, and the beam angle is 66.4°. Thus, as frequency shifts over the available bandwidth of the transducer, the beam pointing angle changes by a total of 18.2°.
One possible set of phases that can produce this effect are 0°, 90°, 180°, and 270°. This can be further simplified to only 0° and 90° and their negatives. The beamformer can be implemented by connecting every fourth element in the array in parallel which results in four distinct signal wires as shown in
As previously described, the acoustic beam geometry illustrated in
Another example of the expected variation of beam angle with frequency is shown in
If instead of a single narrowband signal, a wideband signal covering the entire transducer bandwidth is transmitted, the frequency dependent nature of the beamformer will cause different portions of the spectrum to be radiated in different directions. By the reciprocal nature of the beamformer, the receiving sensitivity will also be maximized in different directions as a function of frequency. Therefore, if the four wires of
The operating principles that have been described above for various embodiments of frequency division transmitting systems also apply to frequency division receiving systems. The frequency division techniques can be applied for the transmitter only, for the receiver only or for both the transmitter and receiver of the sonar system. When applied to the receiver system, a receive beamformer is used which applies the phase shifts to the received signals from each transducer element.
Some Possible Implementation of the Frequency Division Beamforming
Various embodiments of the sonar system utilize a fixed phase difference in signals provided to adjacent transducer, and the direction of the acoustic beam is controlled by varying the frequency of the signals. In principle, any phase difference between the adjacent signals can be used to drive the transducers. However, certain values of the phase difference provide a substantially simpler way of driving the plurality of transducers of the array. Some of such phase difference values are described herein.
Application of the Technique to Forward Scanning Sonar
A forward looking obstacle avoidance sonar for use by swimmers or small vehicles needs to scan a narrow horizontal beam over a wide angular sector or to form multiple preformed beams in order to provide a high resolution search capability. Mechanical rotation of the transducer array is simple but slow and requires electromechanical components which may be unreliable. Multiple preformed beams using time delay and sum techniques are desirable because they provide wideband processing capabilities in many directions simultaneously. However, providing this capability at high frequencies and with large fractional bandwidths remains a difficult task even with the advances of modemmodern electronics. Digital techniques require high sampling rates and a processor capable of high throughput. High frequency analog delay and sum methods are also complex and may experience degradation due to component variations and temperature sensitivity.
Frequency division beamforming can be used if the available bandwidth of the individual transducers in the array is substantially larger than the amount needed for adequate range resolution. This is the case for one embodiment of the transducer wherein useable bandwidth is in excess of 50% of the center frequency.
There are many methods available for using frequency division beamforming to meet the scanning and resolution requirements of typical sonar applications. The following discussion illustrates some of these considerations.
As discussed above, the use of ±90° phase shifts between elements provide one possible frequency division beamforming capability. In general, the frequency division beamforming system can use any arbitrary phase shift, Δφ, between elements. The phase shift applied to the nth element is then nΔφ Modulo 2π.
This could result in a different phase shift for every element in the array which can complicate the driving and receiving circuitry. Among desirable choices are those for which few different phases are sufficient for the entire array. The ±90° case operated using two phases, 0° and 90°, and their negatives. There are numerous other cases which also use relatively few distinct phases for all elements. Consider the example of Δφ=3π/4.
The 0th element is driven by cos ωt.
The 1st element by cos(ωt−3π/4)=−1/√{square root over (2)} cos ωt,+1/√{square root over (2)} sin ωt.
The 2nd element by cos(ωt−3π/2)=−1/√{square root over (2)} sin ωt.
The 3rd element by cos(ωt−9π/4)=1/√{square root over (2)} cos ωt,+1/√{square root over (2)} sin ωt.
The 4th element by cos(ωt−3π)=−cos ωt.
The 5th element by cos(ωt−15π/9)=1/√{square root over (2)} cos ωt,−1/√{square root over (2)} sin ωt.
The 6th element by cos(ωt−9π/2)=sin ωt.
The 7th element by cos(ωt−21π/9)=−1/√{square root over (2)} cos ωt,−1/√{square root over (2)} sin ωt.
The 8th element by cos(ωt−6π)=cos ωt.
The sequence of phases repeats after eight elements. In addition, the phases of the 4th through 7th elements are the negatives of the 0th through 3rd elements. Thus, four phases (0°±45°, and 90°) and their negatives are used to drive the entire array, regardless of the total number of elements.
Now consider the range of steering angles this phase shift can provide. Assuming d=λo/2,
3/5≦cos θs≦1, and 0°≦θs≦53.1°.
Thus, by sweeping over the 50% fractional bandwidth frequency range, the beam can be steered from endfire (0°) to 53.1°. The beamwidth will vary considerably over this range of angles. At endfire, the 3 dB beam width of a uniformly weighted array is given by
Δφ(endfire)≈108°√λ/Nd (8)
whereas well away from endfire, it is given by
Δφ (near broadside)≈50.8°λ/(Nd sin θs). (9)
Consider an array which is 60λ (thus making N=120) at the center frequency. Table 1 lists the beamwidth and steering angle for various frequencies for this array assuming 135° phase shifts between elements.
This design allows the beam to steer over a wide angular range (0°−53.1°), but the beam becomes broad near endfire which reduces cross-range resolution in that direction.
Consider an alternate design in which the elements are spaced slightly apart. Let d=λ0/√{square root over (3)}=0.577λ0. Table 2 shows the steering angles and beam widths case. This approach reduces the scanning range
to only 28.7°, but results in a beamwidth which varies by less than a factor of 3 as opposed to a factor of 18.
The range of angles which can be scanned versus inter-element phase shift and the number of distinct phases used are shown in Table 3. The spacing is assumed to be λ0/2 in all cases. As seen above, changing d can move the sectors slightly.
It will be noted that the principal limitation of this technique is the relatively limited angular scanning region which is achievable with a specific phase shift even given the 50% fractional bandwidth transducers. However, if three sets of phase shifts, say 60°, 90°, and 120° are provided, the scanning could cover a range of 47.2° from 27.3° to 74.5°. In addition, four phases, 0°, 60°, 90°, and 120°, and their negatives are used for all these cases. Different interconnections to the elements can be adapted to implement such a system.
It will also noted that for a line array, a set of beams symmetrical about broadside can also be formed by reversing the sign of the phase shift between elements. Furthermore, the broadside beam can be formed by directly summing all elements with no phase shift. Thus, such arrangement can get coverage over two substantially symmetrical sectors at 15.5° to 62.7° away from broadside plus a single beam at broadside in a relatively simple manner. The location of these angular sectors can be modified slightly by changing the element spacing. However, there remains a problem in filling in the gaps of approximately 15° on either side of broadside. This poses a challenge in using this type of beamforming in a forward looking array. If, on the other hand, a pair of arrays were used in a quasi-side scan configuration but with the axis of each array rotated by 27° relative to the fore-aft axis of the platform, the combination could provide both forward scan and side scan capabilities. There is full coverage of angles from dead ahead (0°) to ±47° which is a reasonable forward scan sector. Then there is coverage from about 79° to 126° on either side which could be used for either single or multiple beam side scan coverage. Thus, there are numerous ways to exploit the capability of frequency division beamforming for various applications.
Some Possible Embodiments of the Transducer Array
Having described the frequency division beamformer above, some possible embodiments of the transducer arrays that can be used in conjunction with the beamformer are now described. As illustrated in
In one embodiment, the frequency division beamformer may also be used to scan the acoustic beam in two dimensions. A two dimensional beamforming technique using a single planar array is disclosed in U.S. Pat. No. 5,808,967 titled “Two-Dimensional Array Transducer and Beamformer”, and is hereby incorporated by reference in it entirety.
Scanning by Combination of Phase and Frequency
In one aspect, the vertical and horizontal scanning is performed by having the beamformer provide signals with a fixed phase value between the adjacent transducers, as described above. The range of scanning is then limited by the range of the frequency of the signals being provided to the transducers.
In another aspect, the beamformer can be adapted to provide signals with different phase values to the adjacent transducers. As seen in the phase array equation
sin α=(Δφ/2π)(c/fd),
the direction angle α can be made to depend on the phase Δφ. Thus by combining the beam direction's dependence of the phase and the frequency, a relatively large angular sector can be scanned in a manner described below.
The bandwidth fraction is defined above in reference to Equation 7.
As seen in Table 4, the beam direction ranges associated with the 50% band with leave slight gaps between the phase set ranges. By utilizing a higher bandwidth of 67%, however, such gaps are removed as the phase set ranges form a continuing coverage.
Although the foregoing description has shown, described and pointed out the fundamental novel features of the invention, it will be understood that various omissions, substitutions, and changes in the form of the detail of the apparatus as illustrated as well as the uses thereof, may be made by those skilled in the art, without departing from the spirit of the invention. Consequently, the scope of the present invention should not be limited to the foregoing discussions, but should be defined by the appended claims.
Notice: More than one reissue application has been filed for the reissue of U.S. Pat. No. 6,678,210. This application is a reissue application of U.S. Pat. No. 6,678,210, which claims priority from U.S. Provisional Application Ser. No. 60/315,579 filed Aug. 28, 2001 titled FREQUENCY DIVISION BEAMFORMING FOR SONAR ARRAYS. This application is also a continuation of U.S. application Ser. No. 11/332,809, filed Jan. 13, 2006, now abandoned, which was a reissue application of U.S. Pat. No. 6,678,210. Another reissue application, U.S. application Ser. No. 11/805,577, filed May 23, 2007, now abandoned, was filed as a continuation of U.S. application Ser. No. 11/332,809. Another reissue application, U.S. application Ser. No. 13/101,835, filed May 5, 2011, now abandoned, was filed as a continuation of U.S. application Ser. No. 11/805,577.
| Number | Name | Date | Kind |
|---|---|---|---|
| 2404391 | Mason | Jul 1946 | A |
| 3419845 | Thiede et al. | Dec 1968 | A |
| 3794964 | Katakura | Feb 1974 | A |
| 4694434 | von Ramm | Sep 1987 | A |
| 4970700 | Gilmour et al. | Nov 1990 | A |
| 5488953 | Vilkomerson | Feb 1996 | A |
| 5512907 | Riza | Apr 1996 | A |
| 5540230 | Vilkomerson | Jul 1996 | A |
| 5640369 | Capell, Sr. | Jun 1997 | A |
| 5675552 | Hicks et al. | Oct 1997 | A |
| 5713362 | Vilkomerson | Feb 1998 | A |
| 5923617 | Thompson et al. | Jul 1999 | A |
| 6084827 | Johnson et al. | Jul 2000 | A |
| 6108275 | Hughes et al. | Aug 2000 | A |
| 6176829 | Vilkomerson | Jan 2001 | B1 |
| 6678210 | Rowe | Jan 2004 | B2 |
| 8811120 | Bachelor et al. | Aug 2014 | B2 |
| 20030214880 | Rowe | Nov 2003 | A1 |
| 20050007882 | Bachelor et al. | Jan 2005 | A1 |
| 20080130413 | Bachelor et al. | Jun 2008 | A1 |
| 20100067330 | Collier et al. | Mar 2010 | A1 |
| 20100074057 | Bachelor et al. | Mar 2010 | A1 |
| 20140230567 | Rowe et al. | Aug 2014 | A1 |
| Entry |
|---|
| Frazier et al., “Analysis of resolution for an amplitude steered array,” IEEE Ultrasonics Symposium Proc., 1999, 1231-1234. |
| Frazier, “A two-dimensional amplitude-steered array for real-time volumetric imaging,” Doctoral Thesis, University of Illinois at Urbana-Champaign, Apr. 2000, 170 pp. |
| Hughes et al., “Tilted directional response patterns formed by amplitude weighting and a single 90° phase shift,” J. Acoust. Soc. Am., May 1976, 59:1040-1045. |
| Mechler, A line hydrophone with frequency dependent beam steering, Masters Thesis, University of Texas, Aug. 1957, 135 pp. |
| Thompson, R.L. et al. “Two Dimensional and Three Dimensional Imaging Results Using Blazed Arrays,” Oceans 2001 MTS/IEEE Conference, Nov. 2001, pp. 985-988. |
| Number | Date | Country | |
|---|---|---|---|
| 60315579 | Aug 2001 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 11332809 | Jan 2006 | US |
| Child | 10232812 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10232812 | Aug 2002 | US |
| Child | 13069321 | US | |
| Parent | 10232812 | Aug 2002 | US |
| Child | 11332809 | US |
