A system of radiating elements arranged for forming one or more beams of radiation having radiation patterns such as a monopole, dipole, quadrupole, other multipoles or combination thereof. The individual radiating elements of the array are interconnected by circuitry providing for the summing and differencing of signals provided by adjacent radiating elements in response to incident radiation. In one embodiment the signal of one radiating element is delayed relative to the signal of an adjacent radiating element. The differencing of the signals provides for the deep nulls found in radiation patterns such as the dipole and quadrupole radiation patterns, while the delay between signals of adjacent radiating elements is adjusted for varying the shape of the directivity pattern. The system provides for varying the direction and shape of beams of the radiation pattern to provide for the detection, classification and/or tracking of a distant source of radiation as well as for illuminating a distant object. The system is responsive to the intensity of radiation received along one or more beams of the radiation pattern, and in response thereto, varies the delay between the signals of adjacent radiating elements and also provides for the selective coupling of specific radiating elements of the system.

Current U.S. Class:	342/375 ; 342/377; 367/100; 367/103; 367/126; 367/97
Current International Class:	G10K 11/34 (20060101); G10K 11/00 (20060101); H04b 007/04 ()
Field of Search:	343/113R,119,16R,12R,1SA 340/6R,16R

Warren; David M. Pannone; Joseph D. Bartlett; Milton D.

---

Parent Case Text

---

This is a division of application Ser. No. 268,416 filed July 3, 1972, now U.S. Pat. No. 3,821,740.

---

Claims

---

What is claimed is:

1. In combination:

an array of radiating elements, said array having dimensions less than a wavelength of energy radiated therefrom;

a plurality of signal carrying channels;

means for coupling individual ones of said channels with respective ones of said radiating elements, each of said channels being coupled to all of said radiating elements to provide combinations of signals of said radiating elements, one of said channels having means for varying the gain of its signal combination relative to that of another of said channels, at least one of said combinations having at least a portion thereof which is the signal of a quadrupole radiation pattern, at least one of said combinations having at least a portion thereof which is the signal of a dipole radiation pattern and at least one of said combinations having at least a portion thereof which is the signal of a monopole radiation pattern;

means for rotating said dipole pattern relative to said quadrupole pattern; and

means for arithmetically combining each of said radiation patterns to form a multiple mode type of radiation pattern, said varying of gain and said rotating serving to shape said multiple mode pattern.

2. A combination according to claim 1 further comprising means coupled between said arithmetic means and said gain varying means for varying the shape of said multiple mode type of radiation pattern in response to signals thereof.

3. A combination according to claim 1 further comprising means coupled between said arithmetic means and said rotating means for varying the shape of said multiple mode type of radiation pattern in response to signals thereof.

4. In combination:

an array of radiating elements;

a plurality of signal carrying channels;

means for coupling individual ones of said channel with respective ones of said radiating elements, said channels being coupled to provide combinations of signals of said radiating elements, one of said combinations being the summation of signals of alternate pairs of said radiating elements subtracted from the summation of the signals of the remaining radiating elements to provide a first quadrupole pattern, a second of said combinations being composed of the summation of signals of pairs of elements subtracted from the sum of the signals of the remaining elements wherein each pair of signals of said pair of elements of said second combination comprises only one signal of the corresponding pair of signals of said first combination to form a second quadrupole pattern having an axis inclined at an angle with respect to the axis of said first quadrupole pattern; and

means for combining at least one of said quadrupole patterns with signals of said channels to provide a polypole pattern.

5. In combination:

an array of radiating elements;

means for algebraically combining signals from individual ones of said radiating elements, such algebraic combination comprising the summing and differencing of paired groups of said signals of said radiating elements, the number of signals in one of said groups of one of said combinations being double the number of signals in one of said groups of a second of said combinations;

means coupled to said combining means for varying the gain of one of said combinations relative to a second of said combinations;

means for determining which of said combinations provide a desired signal; and

means coupled to said determination means for displaying a direction of radiation corresponding to said determination.

6. The combination according to claim 5 further comprising multiplexing means and signal generation means, said array being coupled via said multiplexing means to said algebraic combining means, said array being coupled to said signal generation means via said multiplexing means, said signal generation means providing a signal for transmission by said array to be reflected from a distant object back towards said array to provide information with respect to the direction of said object from said array.

7. The combination according to claim 6 further comprising timing means and gating means, said timing means synchronizing said signal generation means with said gating means, said gating means inhibiting said algebraic combining means during periods of time different from a time delay corresponding to the anticipated range of said object for providing information relative to said range.

8. A system for combining signals of a radiator array comprising:

an array of radiating elements;

first combining means coupled to a first and a second group of said radiating elements for subtracting the sum of the signals of said second group from the sum of the signals of said first group of radiating elements;

second combining means coupled to a third group and a fourth group of said radiating elements, said third group comprising the radiating elements of a portion of said first group and the radiating elements of a portion of said second group, said fourth group comprising the radiating elements of a portion of said second group and a portion of said first group of radiating elements, said second combining means subtracting the sum of the signals of said fourth group from the sum of the signals of said third group of radiating elements;

means coupled to said first and said second combining means for comparing an output signal of said first combining means with an output signal of said second combining means; and

means coupled to said comparing means for varying the gain of an output signal of said first combining means relative to an output signal of said second combining means, said varying of gain providing a variation in the shape of a radiation pattern of said radiator array.

9. A system for combining signals of a radiator capable of vibrating in multiple modes comprising:

a ring of acoustically active material;

a plurality of electrodes positioned along a surface of said ring, each of said electrodes receiving a signal according to its position on said ring and in response to a mode of vibration of said ring, said ring being capable of vibrating in a mode of vibration corresponding to a monopole, a dipole and a quadrupole radiation pattern;

first combining means coupled to a first and a second group of said electrodes for subtracting the sum of the signals of said second group from the sum of the signals of said first group of electrodes;

second combining means coupled to a third group and a fourth group of said electrodes, said third group comprising the electrodes of a portion of said first group and the electrodes of a portion of said second group, said fourth group comprising the electrodes of a portion of said second group and a portion of said first group of radiating elements, said second combining means subtracting the sum of the signals of said fourth group from the sum of the signals of said third group of electrodes; and

means coupled to said first and said second combining means for comparing an output signal of said first combining means with an output signal of said second combining means.

10. A system according to claim 9 further comprising means coupled to said comparing means for varying the gain of an output signal of said first combining means relative to an output signal of said second combining means, said varying of gain providing a variation in the shape of a radiation pattern of said radiator.

---

Description

---

BACKGROUND OF THE INVENTION

This invention relates to super-directive arrays and more particularly to a system including such an array wherein signals from adjacent elements of the array are delayed with respect to each other by preselectable amounts of delay.

Dipole and quadrupole radiators have been well known for many years in the fields of electromagnetic and acoustic radiators. These radiation patterns are obtained at all frequencies for which the interelement spacing and the size of an element are substantially smaller than a wavelength of the radiation transmitted or received by the radiator or array of radiators. For example, in the case of the electromagnetic dipole radiator, the two active regions of the radiator, namely, the positively and negatively charged termini of the radiating element are spaced apart, preferably less than one-quarter wavelength. In the case of the acoustic dipole radiator, a pair of monopole or omnidirectional radiating elements are placed close together, preferably less than one-quarter wavelength spacing, and the effect of a positive and negative pulsation is provided by subtracting the signal received by one of the elements from that of the other element. A directional sonar system employing omnidirectional and dipole radiation patterns is disclosed in U.S. Pat. No. 3,176,262 which issued to S. L. Ehrlich, et al., on Mar. 30, 1965. Since the invention to be described hereinafter is particularly useful for acoustic radiators, the ensuing description will be directed towards acoustic systems while it is understood that these teachings are also applicable to electromagnetic systems.

It is evident that an array of closely spaced radiating elements is essential for the directional transmission and reception of acoustic energies at very low frequencies since a typical phased array antenna in which the radiating aperture is many wavelengths long would be prohibitively large in many applications utilizing frequencies as low as, for example, 10 hertz. However, even with higher frequencies, an array of closely spaced radiating elements may be useful since the directivity pattern of such an array is substantially invariant with the frequency of radiation for all radiations having a wavelength substantially larger than the overall dimensions of the array. At still higher frequencies where the wavelength becomes smaller than the array, the array may still be useful for specific applications such as where a multiple lobed beam pattern is desired, however, such a system tends to be of very narrow band width. It is also apparent that an array of closely spaced radiating elements is inherently useful in situations requiring small size and expense.

A problem arises in that such an array produces beam widths which are excessively wide for the standard techniques utilized in sonar tracking, for example, the tracking of an object submerged in the ocean. In addition, due to the subtraction of signals of adjacent elements as compared with the summation of such signals in the standard phased array antenna, the sensitivity of such an array, namely, the amplitude of the resultant signal produced by the array in response to an incident amplitude of sound pressure, is significantly reduced thereby increasing the unwanted effects of noise.

SUMMARY OF THE INVENTION

In accordance with the invention there is provided a system incorporating an array of radiating elements which overcome the aforementioned problems in the prior art to permit the use of such an array for communicating information by acoustic signals. In one embodiment providing for a polypole radiation pattern, the invention comprises a plurality of radiating elements each of which provides signals in response to incident acoustic energy, the radiating elements being intercoupled via delay lines and circuitry for combining the signals to permit the summing of such signals, the differencing of such signals and the scaling of such signals. In another embodiment utilizing three radiating elements, a simplification is presented which permits the use of a single delay line by connecting the radiating elements via summing resistors having values which provide for a scaling of the signal voltage of one radiating element relative to the signal voltage of the other radiating element; this provides, in the case of small phase angles, (less than approximately 20.degree.), the equivalent of three delay lines. There is also disclosed a multiple element array including a switching circuit for selectively coupling groups of three radiating elements whereby a beam of acoustic radiation can be directed in any one of a plurality of directions by appropriately selecting the group of three elements. In addition there is disclosed the use of an angle tracking circuit interconnected with the switching circuit and the delay lines for operating the switching circuit and the delay lines to rotate a beam of radiation by means of the switching circuit and to alter the shape of the array directivity pattern by varying the amount of delay provided by each of the delay lines. The combination of signals from the radiating elements provides for polypole radiation patterns described mathematically by a power series and polymode radiation patterns described mathematically by a Fourier series. Means are also disclosed for digitizing the signal as by means of a sampling circuit which provides voltage states corresponding to logical 1's for the positive portions of the sinusoidal waveform of the acoustic wave and logical 0's for the negative portions of the wave. A digital delay line in the form of a shift register with multiple outputs is utilized in the digital embodiment as a variable delay line.

BRIEF DESCRIPTION OF THE DRAWINGS

The aforementioned features and other advantages of the invention are explained in the following description and taken in connection with the accompanying drawings wherein:

FIG. 1 is a view, partially isometric and partially diagramatic, of the radiating system in accordance with the invention;

FIG. 2 shows, by way of example, one form of directivity pattern obtainable with the array of radiating elements of FIG. 1;

FIG. 3 shows the underside of a boat in the ocean with a radiating array of the invention mounted thereon, the figure further showing a cardioid radiating pattern with the notch in the direction of the propeller for inhibiting propeller noise;

FIG. 4 shows a block diagram of a beam former of the system of FIG. 1 for received signals;

FIG. 5 shows a block diagram of a delay unit of the beam former of FIG. 4;

FIG. 6 shows a block diagram of an arithmetic unit of the beam former of FIG. 4;

FIG. 7 shows a block diagram of a beam former of FIG. 1 for transmitted signals;

FIG. 8 is a block diagram of an alternative embodiment of the beam former of FIG. 7;

FIG. 9 is a diagrammatic presentation of a generalized beam forming system for a multiple element array;

FIG. 10 is a block diagram, partially schematic, of a simplified embodiment of the beam former of FIG. 4;

FIG. 11 is a diagram of a portion of an array illuminated by a wave of radiant energy for use in a mathematical description of the beam formers of FIGS. 4 and 9;

FIGS. 12, 13 and 14 represent successive mathematical transformations utilized in developing the simplified beam forming system of FIG. 10;

FIG. 15 is a block diagram of an angle tracking unit of the system of FIG. 1;

FIG. 16 is a schematic diagram of a polymode transducer and receiver of the invention;

FIG. 17 is a pictorial equation showing the combination of various receiving directivity patterns provided by the receiver of FIG. 16 to form a polymode directivity pattern;

FIG. 18 is a block diagram of a transmitting and receiving system incorporating the polymode array and receiver of FIG. 16; and

FIG. 19 is a block diagram of an alternate embodiment of the receiver of FIG. 16.

DESCRIPTION OF THE PREFERRED EMBODIMENT

Referring now to FIG. 1, there is shown a system 20 comprising, in accordance with the invention, an array 22 of radiating elements which is coupled to a transceiver 24 via multiplexing circuitry 26. Each of the radiating elements takes the form of a transducer 28 suitable for converting electrical energy signals on wires 30 to sonic energy signals indicated by waves 32, and for converting from sonic energy to electrical energy.

The transducers 28 are of a well-known construction typically comprising an electrostrictive ceramic material such as barium titanate and are suitably mounted upon a base plate 34 to serve as monopole radiators. The surface 34 of each transducer 28 pulsates to provide a directivity pattern having axial symmetry. Six transducers 28 have been arranged, in this embodiment of the invention, in a circle about a seventh transducer 28 at the center of the array 22. As will be described hereinafter, this arrangement of the transducers 28 in the array 22 is particularly useful for providing a variety of radiation patterns as well as a capability for steering beams of radiation about a central axis 36 which is normal to the base plate 34.

The transceiver 24 comprises a beam former 38 for forming the receiving beam for radiation incident upon the array 22 and a beam former 40 for forming the transmitted beam of radiation. In the transceiver 24 a signal generator 42 provides a signal suitable for transmission via the array 22 to a distant object such as a boat 44, and is coupled to the array 22 via the beam former 40 and the multiplexing circuitry 26. Detection of a signal such as an echo from the boat 44, indicated by waves 46, is accomplished by processing the output of the beam former 38 on line 48 by conventional means. Accordingly, the transceiver 24 utilizes a correlator 50 which correlates the received signal on line 48 with a replica of the transmitted signal provided on line 52 from the signal generator 42. Such a correlator may employ recirculating delay lines which may be implemented digitally by shift registers as is disclosed in U.S. Pat. No. 3,594,718 which issued to Charles I. Black, et al., on July 20, 1971.

The transceiver 24 further comprises an angle tracker 54 which, in response to the correlator output on line 56, provides signals on lines 58 and 60 to the beam formers 38 and 40, as well as along line 62 to a beam switcher 64 for providing a desired shape to the directivity pattern of the array 22 and for switching the direction of a beam of radiation provided by the array 22. The beam switcher 64 comprises a well-known switching circuitry which may be either electrical or mechanical, such as a stepping switch for coupling individual ones of the transducers 28 to the beam formers 38 and 40, the specific coupling arrangement being in accordance with the signal provided on line 62. This switching of the beam, analogous to sequential lobing in the radar art, enables the angle tracker 54 to obtain information on the direction of radiation, such as the waves 46, incident upon the array 22. Angle data provided by the angle tracker 54 on line 56 and range data provided by the correlator 50 on line 68 are presented on a display 70 which may be of conventional form such as a cathode ray tube or a graphical display. The signal generator 42, correlator 50, angle tracker 54, display 70 and the beam formers 38 and 40 are coordinated by timing signals from the timing unit 72. The timing signals applied to the beam formers 38 and 40 and the angle tracker 54 are clock pulses indicated by the symbol C.

The multiplexing circuitry 26 comprises in addition to the beam switcher 64, a transmit-receive circuit 74 and amplifiers 76 and 78 for amplifying respectively the received and transmitted signals. The transmit-receive circuit 74 is of conventional form and may comprise, for example, a diode circuit such as is frequently utilized in sonar applications.

Referring now to FIG. 2, there is seen a typical directivity pattern achievable with a linear array of three radiating elements 80 which may be, for example, the transducers 28 of FIG. 1. In FIG. 2 the radiation pattern is that of a quadrupole radiator modified such that one lobe is substantially larger than any other lobe. The particular spacings between the radiating elements 80 and the signal delays to be applied between signals of these elements 80 for producing a specific radiation pattern will be disclosed hereinafter. Comparing the array 22 of FIG. 1 with the arrangement of radiating elements 80 of FIG. 2, it is apparent that any set of three transducers 28 arranged along a diameter of the array 22 correspond to the linear array of three radiating elements 80 of FIG. 2. Accordingly, the radiation pattern of FIG. 2 can be obtained with any three transducers 28 arranged along a diameter of the array 22. Furthermore, by selectively switching in any such set of three transducers 28 by means of the beam switcher 64, the radiation pattern of FIG. 2 can be rotated about the axis 36 of the array 22 of FIG. 1.

Referring now to FIG. 3 there is shown a bottom view of a boat 82 sailing through the ocean waters 84 and having an enclosure 86 mounted on the boat 82 in acoustic contact with the water, the enclosure housing the array 22 of FIG. 1, not seen in FIG. 3. The multiplexing circuitry 26 and the transceiver 24 of the system 20 are not shown in FIG. 3 but are carried within the boat 82. The system 20 is seen providing a directivity pattern 88 having the general cardioid-type form with a notch 90 in the direction of the stern of the boat 82 so that noise caused by the movement of the ocean water 84 past the propeller shaft 92, the propeller 94 and the rudder 96 is greatly attenuated. It is noted that the directivity pattern 88 differs from that of the typical phased array antenna in that there are no reduced level side lobes in the direction of the propeller 94 but rather the notch 90 which can be accurately positioned to exclude the propeller noise. This feature permits the use of the invention in a relatively high noise environment even though the sensitivity of the array 22 of FIG. 1 is substantially decreased when the signal from the transducers 28 are combined in a manner, to be described hereinafter, wherein the signals of adjacent transducers 28 are subtracted from each other to produce cardioid-type or polypole-type radiation patterns.

Referring now to FIG. 4, there is seen a block diagram of the receiving beam former 38. The beam former 38 may utilize conventional analog delay lines for delaying the signals between adjacent radiating elements of the array 22 of FIG. 1 or, alternatively, may employ digital techniques in the form of shift registers for providing this delay. In the preferred embodiment of the invention the beam former 38 utilizes digital techniques, and, accordingly, the beam former 38 comprises samplers 98 for sampling the signals of the transducers 28 indicated diagrammatically in FIG. 4, arithmetic units 100 for combining the signals of the transducers 28 and delay units 102 for delaying the signals between the transducers 28. The delay unit 102 and the arithmetic unit 100 will be more fully described with reference to FIGS. 5 and 6.

The beam former 38 operates as follows. Signals received by the transducers 28 pass via the multiplexing circuitry 26 to respective samplers 98. Each sampler is of conventional design and is driven by clock pulses received from the timing unit 72 of FIG. 2. Each sampler 98 preferably comprises a switch or gate driven by the clock pulses for obtaining samples of the incoming signal, and a comparator (not seen in the figures) which produces an output pulse on line 104 when the sample of the incoming signal is above a preset threshold. To simplify this explanation, it is assumed that the incoming acoustic signal is a sinusoid and, accordingly, the value of this threshold is selected so that for each positive half cycle of the incoming sinusoidal signal, a succession of pulses having a logic state of 1 is provided along line 104, and that during each negative portion of the sinusoid there is an absence of the pulses on line 104 or, equivalently, the 0 logic state is present on line 104. The clock pulse frequency is greater than the frequency of the incoming signal by a factor of, for example, 1,024 so that a total of 1,024 samples can be taken during each cycle of the incoming signal. The 1,024 samples permit a 10-bit resolution of the phase of the incoming signal, namely, increments of approximately 1/3 degree.

Referring now to the block diagram of FIG. 5, the delay unit 102 comprises a shift register 106 and a digital selector 108 (also known as a decoder or multiplexer switch). The shift register 106 has a single terminal in which the input signal is applied, and a plurality of output terminals on lines 110, each of the lines 110 corresponding to one stage of the shift register 106. The shift register 106 is driven by clock pulses, indicated by the letter C in the figure, with the result that each pulse of the input signal appears after successive increments of delay at successive ones of the output terminals on lines 110, with each of these delay increments being equal to the interval between clock pulses. The digital selector 108 has multiple input terminals corresponding to each of the lines 110 and a single output terminal. A specific connection is made between one of the input terminals and the output terminal in accordance with a binary number applied at terminal D and provided by the angle tracker 54 as seen in FIG. 4. Thus, the angle tracker 54 is able to select, via the digital selector 108, a specific output terminal of the shift register 106 so that the delay unit 102 provides a desired amount of delay between the input and output signals of the delay unit 102.

Referring now to the block diagram of FIG. 6, the arithmetic unit 100 comprises digital selectors 112 and 114, each of which is similar to and functions in a fashion analogous to the digital selector 108 of FIG. 5. The arithmetic unit 100 further comprises a scaling unit 116, an operational amplifier 118, a band-pass filter 120 and a sampler 122 similar to the sampler 98 of FIG. 4. The arithmetic unit 100 has two inputs, one providing a signal from the sampler 98 and the other providing a signal from the delay unit 102, and combines these two signals to provide a single output signal. The two input signals may be combined by summing the two together by taking their difference or by first scaling the signals to combine a fraction of one with the other. This is readily accomplished by summing resistors (not seen in the figures) applied to the input terminals of the operational amplifier 118; in this case, a multiplicity of such summing resistors of differing values are provided by the scaling unit 116, with the digital selectors 112 and 114 interconnecting each input signal with any desired summing resistor. The digital selectors 112 and 114 provide the inverse function of the digital selector 108, that is, each digital selector 112 and 114 has a single input and a multiplicity of outputs on lines 124 with a specific interconnection of one of the outputs to the input being provided in accordance with a binary number appearing at terminal A and provided by the angle tracker 54 as seen in FIG. 4. The two input signals to the arithmetic unit 100 are in binary digital format and, after being combined by the operational amplifier 118, are converted by the band-pass filter 120 to an analog signal on line 126. The pass band of the filter 120 is selected to pass the frequency of the acoustic radiation incident upon the array 22 of FIG. 1 while excluding the sampling frequencies introduced by the samplers 122. The analog signal on line 126 is then converted by the sampler 122 to a binary digital format to appear at the output terminal of the arithmetic unit 100.

Referring now to FIG. 7, there is shown a block diagram of the transmitting beam former 40 of FIG. 1. The beam former 40 accepts a signal in sampled format on line 128 from the signal generator 42 of FIG. 1 and converts this signal into analog signals suitable for being transmitted via the transducers 28. The beam former 40 comprises inverters 130, delay units 102 previously described with reference to the receiving beam former 38, a summer 132 and band-pass filters 134. The beam former 40 may utilize analog signals and analog delay lines, however, the preferred embodiment of the invention, as shown in FIG. 7, utilizes digital delay units 102 with the aforementioned sampled or digital format of the signal on line 128. Each of the three transducers 28, namely, the left, center and the right transducers 28, receive separate signals in order to radiate energy having a desired directivity pattern. The amplitude of the signals for each of the transducers 28 is controlled by amplifiers 136 and in accordance with a signal provided at terminal A from the angle tracker 54 of FIG. 1. To provide the signal for the left transducer 28, the digital format signal on line 128 is passed through the band-pass filter 134 having a pass band centered at the frequency of the acoustic signal radiated by the transducers 28; the pass band is sufficiently narrow to attenuate components of the sampling frequency so that only the baseband signal passes via the amplifier 136 and the multiplexing circuitry 26 to the transducer 28. To provide the signal for the right transducer 28, the digital format signal on line 128 is first applied to inverter 130, then delayed by delay unit 102, again inverted by a second inverter 130 and again delayed by a delay unit 102, and is finally filtered by a band-pass filter 134 to remove all of the sampling frequency components thereby converting it to the baseband signal suitable for radiation from the right transducer 28. Each of the inverters 130 is a digital inverter and converts a logic state of 1 to a logic state of 0 and vice versa. The reason for using two inverters as well as two delays will become apparent in a mathematical description of the operation to be provided hereinafter. Each of the delays 102 provides a suitable phase shift between the signals emanating from the right and left transducers 28 as is required for producing the desired directivity pattern. The center transducer 28 obtains its signal via the summer 132 which sums together the delayed signal provided by a pair of delay units 102 and the inverters 130. It is evident that the diagram of FIG. 7 can be simplified, as shown in FIG. 8, however, the diagram of FIG. 7 parallels that of FIG. 4 and thus shows an arrangement suitable for forming a polypole radiation pattern just as the diagram of FIG. 4 is suitable for recieving a polypole radiation pattern.

Referring now to FIG. 8 there is shown an alternative transmitting beam former 137 which may be substituted in place of the transmitting beam former 40. The beam former 137 is simpler than the beam former 40 in that the three delay units 102 of FIG. 7 have been replaced in FIG. 8 with two delay units 138 similar in operation to the delay units 102 but having different delays. In addition two selector switches 140 are provided to select either the signal on line 128 or the inverted form of the signal and thereby provide a more general synthesis of directivity patterns.

Referring now to FIG. 9 there is shown a schematic diagram of the arithmetic units and delay units of a beam former 142 which is similar to the receiving beam former 38 of FIG. 1 except that the beam former 142 is shown in a generalized form which is suitable for any number of radiating elements such as the transducers 28. Arithmetic units 144, some of which are further identified by the letters A-C, are shown providing the differences between signals of neighboring transducers 28 while delay units 146 are shown providing the appropriate delays. In the beam former 38 of FIG. 4, the arithmetic units 100 are shown as being capable of forming either the sum or the difference of signals from neighboring transducers 28, however, in FIG. 9 the arithmetic units are presumed to be forming only the difference and this is adequate for demonstrating one form of higher order system in which any number of elements may be utilized for multiple radiation patterns. It is also noted that the delay units 146 in the first row of the diagram of FIG. 9 are of equal value as is indicated by the symbol .tau..sub.1 ; similarly the values of the delays provided by the delay units 146 in the second row of the diagram are of equal value as is indicated by the symbol .tau..sub.2, and that furthermore, there is one less delay in the second row than in the first row of the diagram. Similar comments apply to the third and subsequent rows of the diagram. Control lines 148 select the desired amount of delay, each control line 148 being energized from a suitable source similar to the angle tracker 54 of FIG. 1.

Referring now to FIG. 10 there is shown a simplified embodiment of a beam former 150 which may be used in lieu of the receiving beam former 38 of FIG. 1 in the situation wherein a polypole radiation pattern is to be received. The beam former 150 comprises four resistors 151-154, a delay unit 156 and summer 158. The appropriate values for the resistors and the value of delay of the delay unit 156 will be derived in terms of the values of delay utilized in the beam former 38 of FIG. 4 in the ensuing mathematical analysis.

The directivity patterns obtainable with a pair of radiating elements, such as the elements A and B seen diagrammatically in FIG. 11, may readily be expressed mathematically. A plane wave of radiation having an angular frequency .omega. is impinging upon the array of the two elements A and B at an angle .theta. to an axis of the array. Each radiating element is an omnidirectional hydrophone and has s sensitivity of K volts/microbar. If a pressure of 1 microbar is assumed, the voltages provided by each element in response to the impinging radiation is, in exponential notation,

v.sub.A =Ke.sup.j.sup..omega.t e.sup.j.sup..pi.(D/.sup..lambda.)cos.sup..theta. (1)

and

v.sub.B =Ke.sup.j.sup..omega.t E.sup.-.sup.j.sup..pi.(d/.sup..lambda.)cos.sup..theta. (2)

respectively for the elements A and B where the point of zero phase is taken for convenience at the midpoint between the two elements and where d is the interelement spacing. The sum of these two voltages forms a monopole v.sub.M given by

v.sub.M =2Ke.sup.j.sup..omega.t cos[.pi.(d/.lambda.)cos.theta.].apprxeq.2Ke.sup.j.sup..omega.t (3)

the approximation being valid for small values of d/.lambda. such that

.pi.(D/.lambda.)cos.theta..ltoreq.20.degree.

approximately. The difference of these two voltages is proportional to the pressure gradient and forms a dipole v.sub.D given by

v.sub.D =2jKe.sup.j.sup..omega.t sin[.pi.(d/.lambda.)cos.theta.].apprxeq.2jK.pi.(d/.lambda.)e.sup.j.sup..om ega.t cos.theta. (4)

the approximation being valid similarly for small values of the argument .pi.(d/.lambda.)cos.theta.. A cardioid response is obtained by combining the monopole and dipole patterns (j.pi.d/.lambda.)v.sub.M +v.sub.D wherein the factor (j.pi.d/.lambda.) is the same as that seen in the expression for v.sub.D in Equation 4. The cardioid expression contains a term invariant with .theta. and a term dependent on .theta.. A quadrupole radiation pattern has a term dependent on cos.sup.2 .theta. or cos2.theta.. The general representation of the voltage P.sub.N (.theta.) provided by the beam pattern of a polypole line array of (N+1) elements may be expressed as ##SPC1##

where .theta.=0.degree. and 180.degree. along the endfire directions of the line, and where the a.sub.n are normalized such that ##SPC2##

Since reciprocity applies, the same expression may be utilized for a transmitting array.

Considering the case of the three elements A, B and C in the array of FIG. 11 and, for convenience, setting the point of zero phase at the radiating element nearest the impinging wave, namely, element A, the output voltages are respectively

v.sub.A =Ke.sup.j.sup..omega.t (7)

v.sub.B =Ke.sup.j.sup..omega.t e.sup.-.sup.j2.sup..pi.(d/.sup..lambda.)cos.sup..theta. (8)

v.sub.C =Ke.sup.j.sup..omega.t e.sup.-.sup.j4.sup..pi.(d/.sup..lambda.)cos.sup..theta. (9)

These voltages are readily combined in accordance with the invention by reference to FIG. 9 where the array is understood to contain only the first three transducers 28 further identified by the letters A, B and C. Three arithmetic units 144A-C and three delay units 146A-C are utilized with the three element array and the combined output voltage appears on line 160. It is presumed for the purposes of this calculation that each arithmetic unit 144 forms the difference of two voltages without performing any scaling. Each of the delay units 146 are presumed for the purposes of this calculation to be analog delays (for ease of mathematical representation) with the delays and resultant phase shifts expressed in terms of the interelement spacing qd and rd where q and r are constants, typically fractions. Thus, letting .tau..sub.1 and .tau..sub.2 be delays equivalent to the incident wave propagating through distances qd and rd respectively, the voltage present on line 162 is given by

V.sub.162 =2jKe.sup.j.sup..omega.t sin[.pi.(d/.lambda.)(q+cos.theta.)]e.sup.-.sup.j.sup..pi.(D/.sup..lambda.) (Q.sup.+cos.sup..theta.) (10)

the voltage on line 164 is given by

V.sub.164 =2jKe.sup.j.sup.[.sup..omega.t.sup.-2.sup..pi.(d/.sup..lambda.)cos.sup..th eta..sup.] sin[.pi.(d/.lambda.)(q+cos.theta.)]e.sup.-.sup.j.sup..pi.(d/.sup..lambda.) (q.sup.+cos.sup..theta.) (11)

and the voltage on line 160 is given by

V.sub.160 =- 4Ke.sup.j.sup.[.sup..omega.t.sup.-.sup..pi.(d/.sup..lambda.)(r.sup.+Q.sup. +2cos.sup..theta.).sup.] sin[(.pi.d/.lambda.)(q+cos.theta.)]sin[(.pi.d/.lambda.)(r+cos.theta.)]. (12)

For small values of the arguments .pi.(d/.lambda.)(q+cos.theta.) and .pi.(d/.lambda.)(r+cos.theta.) in the expression for the voltage on line 160, the peak amplitude of this voltage may be approximated by

V.sub.p =4K(.pi.d/.lambda.).sup.2 (q+cos.theta.)(r+cos.theta.) (13)

or

V.sub.p =4K(.pi.d/.lambda.).sup.2 (1/p)[pqr+p(q+r)cos.theta.+pcos.sup.2 .theta.] (14)

where (1/p) serves a normalization factor (1/p=qr+q+r+1) in Equation 14, and in which the bracketed term is recognized as a power series such as the polypole representation of Equation 5. It is also interesting to note that for small values of these arguments the voltages on the lines 162 and 164 are approximated by the sum of a monopole and a dipole radiation pattern, the monopole being provided by the factor q and the dipole being provided by the factor cos.theta..

The three constants pqr, p(q+r) and p may be given any desired values for providing a desired directivity pattern to the three element array. In particular, a maximum directivity, in the sense of a maximum fraction of the overall radiant energy being found in the main lobe, is obtained in the three element case for the following values, namely,

pqr=1/6

p(q+r)=1/3 (15) p=5/6

for which

q=0.29

r=0.69 (16)

The minus sign in front of q simply means that the delay units 146A-B should provide a time advance or equivalently, the delay units 146A-B need be placed on the opposite input ports of the arithmetic units 144A-B respectively so that a delayed voltage from element A is combined with an undelayed voltage from the element B, and similarly, with respect to elements B and C.

The preceding mathematical development also applies to the beam former 38 of FIG. 4 in the case where the arithmetic units 100 perform simply a subtraction operation as did the arithmetic unit 144 of FIG. 9. In such a case, the simplified beam former 150 of FIG. 10 closely approximates the beam former 38 and may be substituted in its place as will now be seen.

Referring now to the signal flow diagrams of FIGS. 12, 13 and 14, the simplified configuration of the beam former 150 of FIG. 10 is now derived. There are provided in FIGS. 12-14 delay units 166-170, and in each delay unit the exponential phase factor of the preceding mathematics is shown. The radiating elements are identified as in FIG. 9 by the letters A, B and C. Inverters 174A-C are provided in FIG. 12, and each applies a 180.degree. phase shift to a sinusoidal signal waveform applied thereto from delay units respectively 166-168. These inverters 174A-C will be replaced by inverter 176 in FIGS. 13 and 14. Summers 178A-C in FIG. 12 combine the various signals by adding them together, the subtraction of FIG. 9 being accomplished by the use of the inverters 174A-C. In FIGS. 13 and 14, these summers 178A-C will be replaced by summer 180.

The beam former of FIG. 12 is functionally equivalent to that portion of the generalized beam former of FIG. 9 comprising the first three radiating elements A-C and the delay units 146A-C described above with reference to FIG. 9; and the preceding mathematics applies also to FIG. 12. In FIG. 13 the delays presented to a signal from radiating element C by the delay units 167 and 168 have been combined into a single delay provided by the delay unit 169, and the delayed and inverted signals from radiating element B are now shown by an equivalent representation wherein inverter 176 replaces the inverters 174A and 174C. Since, in the FIG. 13 a single delay unit, namely, delay unit 169, provides all of the delay for the radiating element C, the summing together of the various signals can now be accomplished with a single summer, the summer 180.

The two delays provided by delay units 166 and 168 in FIG. 13 can be combined into the single delay of delay unit 170 of FIG. 14 in the following manner. In summing together the signals of the delay units 166 and 168, it is seen that the exponential delay terms add (since the two signals are otherwise equal and may be factored out) with the result

e.sup.-.sup.j2.sup..pi.qd/.sup..lambda. +e.sup.-.sup.j2.sup..pi.rd/.sup..lambda. = 2cos[.pi.(r-q)d/.lambda.]e.sup.-.sup.j.sup..pi.(Q.sup.+r)d/.sup..lambda. .apprxeq.2e.sup.-.sup.j.sup..pi.(q.sup.+r)d/.sup..lambda. (17)

where .pi.(r-q)d/.lambda. is small. The factor of 2 is provided for in the diagram of FIG. 14 by scaling the signals of the elements A and C by a factor of 1/2.

FIG. 10 is readily seen to be equivalent to FIG. 14 since the resistors 151-154 provide for the scaling of 1/2, these resistors being of equal value, and the inverter 176 as well as the amplifiers of the multiplexing circuitry 26 function as voltage sources. With respect to the signal from element A, the signal splits between the resistors 152 and 153 so that one-half goes to the summer 158 and one-half goes to the delay unit 156. Thus, at the output of the summer 158 there appears the following combination of delay factors

1/2e.sup.-.sup.j2.sup..pi.(q.sup.+r)d/.sup..lambda. +1/2=cos[.pi.(q+r)d/.lambda.]e.sup.-.sup.j.sup..pi.(q.sup.+r)d/.sup..lambd a. .apprxeq.e.sup.-.sup.j.sup..pi.(q.sup.+r)d/.sup..lambda. (18)

where .pi.(q+r)d/.lambda. is small. The final expression in the above equation is seen to be the delay factor in the delay unit 170 of FIG. 14.

Referring now to the block diagram of FIG. 15, the angle tracker 54 comprises four programming units, respectively, a search selector programmer 182, a search pattern programmer 184, a track selector programmer 186, and a track pattern programmer 188, a switch 190 for selectively coupling the programming units 182 and 186 to a memory 192, and a switch 194 for selectively coupling the programming units 184 and 188 to a memory 196. The angle tracker 54 further comprises a detector 198 and a predictor 200 both of which are responsive to signals on line 56 from the correlator of FIG. 1, the detector 198 actuating the switches 190 and 194, and the predictor adjusting the programs of the track selector programmer 186 and the track pattern programmer 188.

In operation, the detector 198 senses the magnitude of the correlator signal on line 56 to determine the presence of a distant source of sound such as the reflected waves 46 from the boat 44 in FIG. 1. The detector 198 has a preset threshold for determining that such a distant source of sound is or is not present. When no such source of sound is present, the switch 190 couples the search selector programmer 182 to the memory 192, and the switch 194 couples the search pattern programmer 184 to the memory 186. The memory 192 stores data with respect to specific ones of the transducers 28 of FIG. 1 which are to be coupled by the beam switcher 64 for orienting the radiation directivity pattern of the array 22 in a specific direction. The search selector programmer 182 provides a succession of directions, in response to the clock pulses at terminal C, for redirecting the directivity pattern of the array 22 for searching for a distant source of acoustic radiation. These directions are sent from the search selector programmer 182 via the switch 190 to the memory 192 which, in response thereto, provides signals along line 62 for operating the beam switcher 64 to couple the appropriate ones of the transducers 28. In a similar manner, the search pattern programmer 184 sequentially selects specific shapes for the directivity pattern during each orientation of the directivity pattern as provided by the beam switcher 64 to aid in the searching. Accordingly, the search pattern programmer 184 designates a specific shape via the switch 194 to the memory 196 which, in response thereto, commands the appropriate set of delays and scaling factors for the beam formers 38 and 40 of FIG. 1, these commands being transmitted in the form of digital signals on the lines 58 and 60.

When the detector 198 determines that a suuitable source of acoustic radiation is present, it actuates the switches 190 and 194 to couple the track selector programmer 186 to the memory 192, and the track pattern programmer 188 to the memory 196. The track selector programmer 186 continually reorients the directivity pattern of the array 22 of FIG. 1 so that the source of acoustic radiation is present on alternating sides of a null in the directivity pattern, this providing effectively a tracking error signal analogous to the tracking error signal developed by a monopulse feed in a radar antenna. Similarly, the track pattern programmer 188 commands the bean formers 38 and 40 via the switch 194 and the memory 196 to adjust the directivity pattern to provide a shape more amenable to tracking the source of radiation.

The predictor 200 monitors the angle coordinates of successive positions of the source of radiation and instructs the track selector programmer 186 to reorient the directivity pattern when the source of radiation tends to move away from the null in the directivity pattern. When the source of radiation moves an angular distance smaller than an angular increment that can be provided by the beam switcher 64, the predictor 200 commands the track pattern programmer 188 to move the null of the directivity pattern to accommodate the angular movement of the source of radiation rather than to command the track selector programmer 186 to update the orientation of the directivity pattern. The average angular orientation of the null in the directivity pattern corresponds to the angular coordinate of the source of radiation, and this information is transmitted from the track selector programmer 186 along line 66 to the display 70 of FIG. 1. To facilitate discrimination betweeen targets of differing ranges from the array 22 of FIG. 1, the correlator 50 of FIG. 1 is provided with a range gate, not seen in the figures, which, in response to a signal from the timing unit 72, passes only those pulses on line 56 corresponding to a radiation source at the desired range, this range being selected by a manual control provided on line 201 in FIG. 1.

It is also apparent that while the preceding discussion has described the implementation of a directivity pattern formed by a linear array of three of the transducers 28 of FIG. 1, it is apparent that four or more of the transducers 28 can be switched into the circuit to further tailor the directivity pattern to a desired shape so that it is possible to keep a null of the directivity pattern oriented towards the stern of the boat 82 to reduce the effects of propeller noise. It is also apparent that a cylindrical array of the transducer elements 28 can be provided by coupling all of the transducers 28, except for the central transducer 28, to the beam formers 38 and 40. Such an interconnection of the transducers 28 with the beam formers 38 and 40 can be accomplished, for example, by coupling together the pair of the transducers 28 labeled A and D, the pair of transducers labeled F and E, and the pair of transducers labeled E and C with these three pairs of transducers 28 being connected to the beam formers 38 and 40 in lieu of the previous connections of the transducers indicated earlier in FIG. 9 with reference to the transducers A, B and C.

The description of the preferred embodiment of the invention, hereinbefore presented, provided for an implementation of the expression for a polypole radiation pattern as given by Equation 5. An analogous embodiment of the invention will now be presented with reference to FIGS. 16-18 for providing a polymode radiation pattern given by Equation 19 ##SPC3##

which is readily seen to be of the same form as Equation 5 except that the summation is seen to be a Fourier series rather than a power series. For the three element array considered in FIG. 2, or the eight element array to be described in FIG. 16, the Fourier series of Equation 19 can readily be converted by trigonometric identities to the power series of Equation 5. Accordingly, the terms polypole and polymode are seen to be convenient terms for describing the mathematical series representation of multiple mode types of radiation patterns.

Referring now to FIG. 16, there is seen a schematic diagram of an alternative embodiment of the invention showing an array 202 of radiating elements which are in the form of a cylindrical radiator having inner and outer concentric metallic electrodes located on the inner and outer surfaces of a ceramic cylinder 204. The outer electrode is identified by the numeral 206 while the inner electrode is segmented into eight segments, each of which are insulated from each other and are identified generally by the numeral 208 with specific segments being further identified by the letters A-H. In the most general form the array is of spherical or spheroidal shape; however, the preferred practical configuration is cylindrical when, for example, elevation angles are secondary to azimuthal angles.

The array 202 is shown coupled to a receiver 210 having eight preamplifiers which will be identified generally by the numeral 212 with individual ones of these preamplifiers 212 being further identified by the letters A-H which correspond respectively to the eight inner electrode segments 208A-H. The inner electrode segments 208 are coupled to their respective preamplifiers 212 via wires 214. Each of the preamplifiers 212 are also identified in the figures by legends such as NNE meaning north-by-northeast and WSW meaning west-by-southwest. A coordinate axis indicating the directions north, east, south and west is also seen overlaid upon the array 202.

The receiver 210 further comprises a first set of operational amplifiers 216 and a second set of operational amplifiers 218 with individual ones of these amplifiers being identified respectively by 216A-G and 218A-H. A filter 220 is provided at the output of the operational amplifier 216A and filters 222A-B are provided at the outputs of the operational amplifiers respectively 216F-G. Adjustable attenuators 223 are coupled to the outputs of the filters 220, 222A and 222B as well as to the outputs of the amplifiers 216B-E for scaling the respective signals, these attenuators being further identified by the letters A-G corresponding respectively to the amplifiers 216A-G. The operational amplifiers 216 are further identified by legends such as the legend M representing a monopole or an omnidirectional radiation pattern, NW/SE representing a dipole radiation pattern oriented with its axis running from northwest to southeast, 000 which identify a quadrupole radiation pattern having its reference axis at an angle of 0.degree. (along the north-south axis) with respect to the array 202, and 045 which represents a quadrupole radiation pattern having its reference axis inclined at an angle of 45.degree. with respect to the array 202. Each of the preamplifiers 212 is connected to all of the operational amplifiers 216, however, for purposes of simplifying the drawing, these interconnections are indicated by means of the numerals 1-8 identifying the output terminals of the preamplifiers 212 and the corresponding numerals 1-8 at each of the eight input terminals to each of the seven operational amplifiers 216. As is well known, operational amplifiers are typically provided with a positive input and a negative input with summing resistors connecting with both the positive and the negative inputs to provide an algebraic combination in which a plurality of signals may be summed together and subtracted from a second plurality of signals which are summed together.

The filters 220 and 222A-B serve as equalizing networks to equalize the frequency response of the omnidirectional pattern and the frequency response of the quadrupole pattern with that of the dipole pattern. As is well known from the mathematics of dipole patterns in acoustic systems, the receiving frequency response of ceramic transducers for frequencies below resonance tends to rise at a rate of 6 dB/oct (decibels per octave) change in frequency. Accordingly, the omnidirectional pattern which has a flat frequency response below resonance is passed through the filter 220 which may be, for example, a lead network, introducing a phase shift of +90.degree. and having a frequency response of +6 dB/oct. Since quadrupole radiating patterns have a frequency response for frequencies below the second order mode resonance frequency of the array 202, wherein the amplitude of signals provided by the amplifiers 216F and 216G rises at a rate of +12 dB/oct, each of the filters 222A-B which may be, for example, a lag network, provide a phase shift of -90.degree. and a frequency responsivity of -6 dB/oct. Thus the signals on each of the lines 224 are characterized by a frequency responsivity which varies at the rate of +6 dB/oct for all frequencies below the resonance region of the array 202. In the situation where a source of sound 226 transmits sound waves 228 towards the array 202, the sound waves 228 having become substantially planar waves in the vicinity of the array 202, the array 202 is built with a diameter which is sufficiently smaller than the wavelength of the waves 228 so that the frequency of these waves is below the resonance region of the array 202.

Each of the lines 224 is connected to at least some of the operational amplifiers 218 but not necessarily all of these amplifiers. Here again, in order to simplify the drawings, the interconnection of individual ones of the lines 224 are indicated by legends affixed to each of the inputs of the respective operational amplifiers 218A-H. Thus, with respect to the operational amplifier 218A, the omnidirectional signal provided by the filter 220 is supplied to one input, the NE/SW dipole signal provided by the operational amplifier 216D is applied to a second input, and the quadrupole signal provided by the filter 222A is applied to the third input of the operational amplifier 218A. As is indicated by the plus signs adjacent each input of the operational amplifier 218A, each of these inputs are summed together to give a voltage V1 which is the voltage associated with one polymode directivity pattern for receiving the waves 228, as will now be described with reference to FIG. 17.

Referring now to FIG. 17, there is shown a schematic representation of the combination of an omnidirectional radiation pattern 230, a dipole radiation pattern 232 and a quadrupole radiation pattern 234 to give a polymode pattern 236 represented by the voltage V1 of the operational amplifier 218A of FIG. 16. The omnidirectional pattern 230 is a graph of the voltage appearing at the output of the filter 220 of FIG. 16 as the source 226 is moved in a large circle around the array 202. Similarly, the dipole radiation pattern 232 and the quadrupole radiation pattern 234 represent the voltages provided by respectively the operational amplifier 216C and the filter 222A as the source of sound 226 is moved in a large circle around the array 202. The summation of these voltages by the operational amplitude 218A is represented in FIG. 17 by the pictorial equation having two plus signs and an equal sign. It is noted that the polymode pattern 236 has one main lobe in the north direction corresponding to the contributions of the northerly portions of the three radiation patterns 230, 232 and 234, each of which have a positive voltage in the northerly direction. On the east and west axis the negative portions of the lobes of the quadrupole radiation pattern 234 are portrayed as having equal and opposite amplitudes to the amplitude of the omnidirectional radiation pattern 230 at the east and west directions and therefore cancel each other so that zero amplitude appears at the east and west directions of the polymode pattern 236. Similar comments apply to the construction of the polymode pattern 236 and other directions. Mathematically this construction is indicated by the algebraic expressions beneath each of the radiation patterns 230, 232 and 234 in which the maximum amplitudes of these radiation patterns have all been set equal to one-third so that the peak amplutude of the polymode pattern 236 is equal to unity.

It is frequently desirable to scale the respective amplitudes of the respective radiation patterns 230, 232 and 234 to provide a maximum directivity for the polymode pattern 236 with the maximum directivity being understood to be a maximization of the ratio of energy received from the northerly direction of the radiation pattern 236 to the total isotropic incident energy received by all of the lobes of the pattern 236. This can be readily accomplished mathematically by integrating the energy received in each direction and forming the standard ratio for the directivity index, with the result that the maximum directivity is obtained for the scaling factors indicated in the attenuators 223A-G of FIG. 16. These scaling factors are, respectively, a value of 3 for the attenuator 223A a value of 4 for each of the attenuators 223B-E, and a value of 5 for each of the attenuators 223F-G. In FIG. 16, the arrow adjacent the scaling factor in each of the attenuators 223 indicates that the attenuation of these attenuators 223 may be varied to provide selective scaling of the monopole, dipole and quadrupole radiation patterns to produce a polymode pattern in which the positions of the nulls can be varied as well as for providing a selective tailoring of the shapes of the various lobes in the polymode radiation pattern.

Returning momentarily to FIG. 16, the operational amplifier 218B is seen to provide a polymode pattern having the same shape as the polymode pattern 236 of FIG. 17 but being oriented at an angle of 45.degree. with respect to the coordinate system of the array 202. This results because of the summation of the monopole signal of the filter 220 with the signal of the operational amplifier 216B and the quadrupole signal of the filter 222B, for which the dipole radiation pattern and the quadrupole radiation patterns have their main axis in the 45.degree. or NW/SE direction. With respect to the voltage V3, provided by the operational amplifier 218C, a polymode radiation pattern perpendicular to that of the polymode pattern 236 of FIG. 17 is provided since here the quadrupole pattern provided by the filter 222A is subtracted from the sum of the monopole pattern of the filter 220 and the east-west dipole pattern of the operational amplifier 216E; this subtraction is indicated by the minus sign adjacent the third input of the operational amplifier 218C. Similar comments apply to the remaining voltages V4-V8 wherein it is seen that a family of eight polymode radiation patterns are simultaneously produced by the receiver 210.

Referring now to FIG. 18, there is seen a system 238 comprising the array 202 and the receiver 210 of FIG. 16, here shown coupled via multiplexing circuitry 240 similar to the multiplexing circuitry 26 of FIG. 1. A signal generator 242 similar to the signal generator 42 of FIG. 1 provides signals which are coupled via the multiplexing circuitry 240 to the array 202 for transmission therefrom as in an active sonar system. The receiver 210 receives either echoes, which are reflected from an object such as the boat 44 of FIG. 1, or signals generated by a distant source of sound such as the source of sound 226 of FIG. 16. The direction of such a distant source of sound can readily be determined by means of logic circuitry 244 which performs the function of noting which of the voltages V1-V8 has a maximum value. A display 246, such as a cathode ray tube display, is coupled to the logic circuitry 244 for displaying the direction of the distant source of sound. Additionally, a timing unit 248, similar to the timing unit 72 of FIG. 1, and a range gating unit 250 may be utilized in an active sonar mode for gating out all received signals except those corresponding to echoes emanating from a preselected range from the array 202. To implement this range gating, the timing unit 248 supplies timing signals to both the signal generator 242 and the range gating unit 250 so that a range gate can be set corresponding to a preselected time interval subsequent to the transmission of a signal generated by the signal generator 242. The range gating unit 250 is coupled to the receiver 210 at the "gate" terminals of the preamplifiers 212 (FIG. 16) to gate out all the polymode voltages of the receiver 210 except for such voltages falling within the designated range gate.

Referring now to FIG. 19 there is shown a block diagram of a receiver 250 which is similar to the receiver 210 of FIG. 16 in that it utilizes the eight preamplifiers 212, five of the operational amplifiers 216, namely, the operational amplifiers 216A, 216C, and 216E-G, and one of the operational amplifiers 218. Each of the preamplifiers 212 connect via the wires 214 to the array 202 as in FIG. 16. The receiver 250 differs from the receiver 210 in that the receiver 250 combines the output from a pair of dipole operational amplifiers, namely, the north-south and east-west dipole operational amplifiers 216C and 216E via a coordinate converter 252 to obtain a dipole, represented by a voltage on line 254, with an axis having any desired orientation with respect to the coordinate system of the array 202. The omnidirectional channel comprises the operational amplifier 216A and the filter 220 of FIG. 16 but no coordinate converter is utilized since the omnidirectional pattern is symmetric about the axis of the array 202. A quadrupole radiation pattern oriented in any direction about the axis of the array 202 is provided by the operational amplifiers 216F-G and the filters 222A-B of FIG. 16 in combination with a coordinate converter 256 providing a voltage on line 258 representing the magnitude of signals received by the polypole radiation pattern.

The coordinate converters 252 and 256 comprise multipliers 260A-D, trigonometric units 262A-B and amplifiers 264A-B each of which has a variable gain control as indicated by the legend G.sub.1 and G.sub.2. The output signals of the multipliers 260A and 260B are summed together by the amplifier 264A, and the output signals of the multipliers 260C and 260D are summed together by the amplifier 264B. The multipliers 260A-D may operate in an analog or in a digital fashion. Thus, the multiplier 260A multiplies the north-south dipole signal by cos.theta..sub.1 and the multiplier 260B multiplies the east-west dipole signal by sin.theta..sub.1 whereupon the two products are summed together to give a resultant dipole having the same amplitude as the north-south dipole but having a dipole axis inclined at an angle .theta..sub.1 relative to the axis of the array 202. Subscripts 1 and 2 are utilized with the angle .theta. to indicate that the dipole pattern may be rotated at different amounts from the quadrupole pattern. Similar comments apply to the operation in the coordinate converter 256 except that here the multiplying factors are cos(2.theta..sub.2) and sin(2.theta..sub.2) since the expression for the quadrupole radiation pattern is in terms of cos(2.theta.), while that of the dipole is in terms of cos.theta. where .theta. is the angle relative to the coordinate system of the array 202.

With respect to the capability of rotating the dipole patterns relative to the quadrupole pattern, as seen with the embodiment of FIG. 19, it is noted that such a capability exists also in the embodiments of FIGS. 1 and 16 but to a limited extent. Thus, in FIG. 1 a rotation of 180.degree. imparted by the beam switcher 64 rotates the dipole pattern 180.degree. while rotating the quadrupole pattern 0.degree., .+-. 90.degree. or 180.degree. since each of these rotations are fully equivalent with the quadrupole pattern which is periodic with each rotation of 180.degree.. With respect to the embodiment of FIG. 16, a dipole pattern can be rotated with respect to a quadrupole radiation pattern by simply selecting the outputs from a different operational amplifier 216 such as the combination of the outputs of the operational amplifier 216C and 216F as is provided by the operational amplifier 218A or the combination of the outputs of the operational amplifiers 216E and 216F as is provided by the operational amplifier 218C.

To implement the coordinate conversion digitally, the multipliers 260A-D would comprise an analog-to-digital converter, a digital multiplier, and a digital-to-analog converter (not seen in the drawings). The trigonometric units 262A-B in digital form would comprise, by way of example, a read-only memory in which the value of cos.theta..sub.1 and of sin.theta..sub.1 would be presented in response to an address supplied on the line labeled .theta..sub.1 ; and with respect to the trigonometric unit 262B the values of cos(2.theta. .sub.2) and sin(2.theta..sub.2) would appear in response to an address supplied along the line labeled .theta..sub.2.

To implement the multiplication of the multipliers 260A-D in analog fashion, the multipliers would comprise, for example, gain control amplifiers in which the gain would be proportional to cos.theta..sub.1 in the case of the multiplier 260A, and similarly for the other multipliers 260B-D. With respect to the trigonometric units 262A-B, in analog form they would comprise, for example, a nonlinear diode-resistor circuit in which the output voltage approximates cos.theta..sub.1 in response to input voltage of .theta..sub.1 with a second such circuit being provided to produce sin.theta..sub.1 in response to .theta..sub.1.

The omnidirectional channel also comprises an amplifier 266 coupled to the output of the filter 220 and having a variable gain in accordance with a voltage applied at terminal G.sub.0 for varying the gain of the omnidirectional pattern relative to the gains of the dipole and quadrupole patterns. The relative gains for the three channels, namely, G.sub.0, G.sub.1 and G.sub.2, as well as the angles of rotation of the dipole, .theta..sub.1, and of the quadrupole .theta..sub.2, are provided by a beam former 268 in response to signals provided by a logic circuit 270. The logic circuit 270 is similar to the logic circuit 244 of FIG. 18 and is useful for tracking a source of sound such as the source 226 of FIG. 16. Thus, the logic circuit 270 in response to the polymode voltage, V, at the output of the operational amplifier 218 provides signals to the beam former 268 for rotating the polymode directivity pattern, and furthermore, for varying the shape of the directivity pattern by varying the gain G.sub.0, G.sub.1, and G.sub.2. The shape of the directivity pattern can also be varied by making .theta..sub.1 different from .theta..sub.2, this providing a flexibility not found in the receiver 210 of FIG. 16. The receiver 250 and the logic circuit 270 may be substituted in the system 238 of FIG. 18 in place of the receiver 210 and the logic circuit 244 to provide a tracking system with improved tracking capability.

It is also evident with respect to the receivers 210 and 250 of FIGS. 16 and 19 that an octopole radiation pattern could be formed by summing and differencing alternate outputs of the amplifiers 212. In this respect it is noted that alternate pairs of inputs to the operational amplifiers 216F-G are summed and differenced to provide the quadrupole radiation pattern, there being two such patterns spatially offset by 45.degree.. In the case of the octopole pattern, only one fixed pattern could be produced with the eight inner electrode segments 208, it being necessary to employ 16 such segments for providing a second octopole radiation pattern, and similarly for providing a combination of these patterns to give a resultant octopole radiation pattern in any arbitrary direction. Via the same reasoning, it is also evident that if only four inner electrode segments 208 were employed in the array 202, then one fixed quadrupole radiation pattern could also be produced; however, there would be no facilities for altering the directional quality of such a pattern.

it is also apparent from a comparison of the array 202 of FIG. 16 and the array 22 of FIG. 1 that the two arrays may be combined into a single composite array (not seen in the figures) in which one or more transducers 28 of the array 22 are replaced with an array having the form of the array 202. In view of the fact that the diameter of the array 202 is substantially smaller than a wavelength, such a radiator may readily be substituted for a transducer 28. With respect to connecting this composite or hybrid array to the multiplexing circuitry 26 and the transceiver 24, the multiplexing circuitry 26 must be enlarged to accommodate a transmit-receive circuit for each of the wires 214 of FIG. 16 for each one of the arrays 202 utilized in the composite array. A plurality of the receivers 210 of FIG. 16 are utilized, one for each of the arrays 202 with the omnidirectional channel in each of these receivers being connected to the receiving beam former 38 of FIG. 1. This provides a composite receiving system having both features of the system 20 of FIG. 1 and features of the system 238 of FIG. 18, and may be referred to as a polypole, polymode receiving system; the composite array may similarly be referred to as a polypole-polymode array. Transmission of signals generated by the signal generator 42 of FIG. 1 may still be accomplished in the polypole fashion by energizing each of the inner electrode segments 208 of an array 202 in parallel circuit from a single output of the transmitting beam former 40 of FIG. 1.

A polypole or polymode or hybrid array may also be constructed in three dimensions, such as a four element array in which each of the radiating elements are located at the vertices of a tetrahedron. Such an array is readily implemented for polypole operation with radiating elements having a spherical radiating surface so as to radiate an omnidirectional pattern in three dimensions. In the case of the polymode operation, a radiating element must be segmented in three dimensions as by utilizing sectors of a sphere. The resultant radiating patterns of the tetrahedronal array are significantly more complex than those presented hereinbefore for the planar array, as may be understood by noting that the projections of the interelement spacing upon an axis of the radiation pattern is dependent on the orientation of the axis relative to the tetrahedron configuration. In particular, there are six major axes each of which passes through the center of the tetrahedron and is parallel to an edge of the tetrahedron. It may also be desirable to place a radiating element at the center of the tetrahedron to provide greater flexibility in forming the radiation pattern in any desired direction. The same form of circuitry as has been disclosed hereinbefore, may be utilized for the tetrahedronal array, however, the beam forming circuitry and the logic control for the beam forming circuitry is extended to encompass all three dimensions in a manner which may be seen by extending the preceding mathematics to three dimensions.

It is understood that the above described embodiments of the invention are illustrative only and that modifications thereof will occur to those skilled in the art. Accordingly, it is desired that this invention is not to be limited to the embodiments disclosed herein but is to be limited only as defined by the appended claims.

* * * * *

---