US3192530A - Electronically scanned array with diode controlled delay network - Google Patents

Description

June 29, 1965 l. SMALL ELECTRONICALLY SCANNED ARRAY WITH DIODE 2 Sheets-Sheet 2 Filed Oct. 24, 1962 Nbm RH.. w OA mM w E5 R tm V 1 m @2.1355 E ,N10 H l R A .Q .w A M \\l\|| ll/ E B Y s; B mmm am Bm Rm m. .WS v Q Q MG l. Tl! @SWS mmmH United States Patent O 3,192,530 ELECTRONICALLY SCANNED ARRAY WTH DIODE CONTROLLED DELAY NETWORK Bernard I. Small, San Diego, Calif., assignor to United tszatilsI of America as represented by the Secretary of e avy Filed Oct. 24, 1962, Ser. No. 232,915

2 Claims. (Cl. 343-854) (Granted under Title 35, U.S. Code i(1952), sec. 266) The invention described herein may be manufactured and used by or for the Government of the United States of America for governmental purposes without the payment of any royalties thereon or therefor.

The present invention relates to scanning antenna systems and more particularly, to a scanning antenna system where the phase front may be shifted in order to direct the beam specifically, to a scanning antenna system which incorporates diode controlled radio frequency switches for switching in various physical path lengths over which the radio frequency energy is coupled to the radiating elements to shift the direction of the phase front of the array.

It has long been thought desirable to have highly directive antennas that move the beam over wide angles without rotation of the main mass of the antenna. Such a technique is sometimes classified as electronic scanning, but would be better known as low-inertia scanning, to include methods in which mechanical motion of a relatively light-weight part of the antenna lis utilized.

For the application at hand, the obvious antenna type is the array of radiating elements scanned by control of phase of the individual elements. In such an array a linear or flat array is the first that comes to mind but curved arrays are not excluded.

There are two basic modes of scanning possible for the stationary aperture antenna with the beam moved by lowinertia electronic or electric-mechanical phase control. One of these is continuous sweep of the beam from one extreme angular position to the other, not unlike scan by the usual physical rotation of the aperture. The other method is step or jump scan, in which the beam position is changed in discreet jumps, either step-by-step through the complete set of angular positions from one extreme to the other, or through any set of selected positions in any selected order.

In the present application continuous scan is not used in favor of step scan in that the application intended is for IFF purposes. However, it is apparent that the system to be disclosed may be used with a continuous scan as well as for step scan. Continuous scan is not used in that the purpose of the present system is to interrogate and identify unknown targets whose positions, within limits, have been determined by the search radar. A more eicient use of the IFF antenna would be to have it aim on command, on only those bearings where there are unidentified targets. Beyond that, efliciency would be further improved by holding the beam on a single bearing only long enough for a positive identification to be made, then jumping the beam to the next position called for by the controller.

In this context efficiency means high data rate, or the ability to handle high traffic density. Basically, the

objective is for the IFF system to be interrogating un.

knowns for the largest possible proportion of the time, that is, to waste as little time as possible in pointing at empty space or vehicles whose identity is not in question. The aim is, therefore, to move the beam as rapidly as possible from one interrogate angle to the next, with as little loss of data in the process as possible.

A reasonable demand on the phase switches which will 3,192,53@ Patented Jurre 29, 1965 be used to scan the antenna is that all changes in the scanning system, mechanical or electrical, necessary to Jump the beam from any position to any other be cornpleted within one period of the pulse-train repetition rate. It is assumed that the changes would be initiated by the trigger that starts some selected interrogation chain and would be completed before the next. Thus for each jump of the beam, only one interrogation would be lost to the scanning mechanism.

If very fast electronic switching elements are used a different principle may be followed. The parameters would then be selected to allow the few microseconds required for the phase switching without loss of any interrogations. For example in the case of a 400 mile range, the repetition period might be as much as 5.360 milliseconds, allowing 16 microseconds for switching. Comparing this with a slower switching for the requirement of, say, 8 interrogations per look, leads to a factor of 12% improvement in over-all data read for the high-speed case.

There are several basic ways in which electronic components can be used to control the relative phase of excitation of array elements. One that has received considerable attention in recent years utilizes a heterodyning technique in which shift of a lower control frequency, when mixed with an R-F signal near the carrier frequency, results in a carrier of the desired phase. Another technique uses any phase control method at lower frequency, then multiplies to final frequency at each element. Such elements, however, require electronic components at the array elements having stable phase characteristics with life and with environmental change.

A second basic technique is to place in series with each array element a component or components that can be made to shift phase, or stretch electrical path length, by electronic means. Two classes are considered here. One is the class of components that use the variable characteristics of ferrite material with variation in magnetic biasing field. Such components in wave guide configurations for higher frequencies have been developed to the point of application, although they have not yet been proved in service.

Problems and limitations in all configurations include relatively high loss for the required phase shift, hysteresis effects and lack of linearity, requiring some complication of control circuits; and susceptibility to temperature and power variations which must be met by control of environment and/or by compensating circuitry. Initial cost is high but the elements themselves should have indefinite life.

The second class of passive and stationary phase Shifters utilizes ferroelectric material in somewhat similar conligurations. Where the ferrite properties that are varied to provide phase shifting may be either Faraday rotation or magnetic permeability, depending on the frequency and configuration, attempts to utilize ferroelectrics are based on the variation of dielectric constant with change in electric biasing field. Thus, the control is by voltage rather than current, which may have some advantage, but so far the materials utilized have even more sensitivity to temperature and to power and development of practical devices is not progressing at a satisfactory rate. A similar approach is the use of discreet variable capacitance elements, i.e., the semi-conductor diodes known as Varactors, but again sensitivity to signal magnitude and temperature, and variations of characteristics from element-to-element in production have precluded the use of these elements` An object of the present invention is to provide a scanning antenna array which is of low cost and high speed.

An additional object of the invention is to provide a scanning antenna system which utilizes passive scanning techniques.

A further object or" the invention is to provide a delay line which may be used in conjunction with a scanning antenna array for shifting direction of the phase front of the beam from the array.

An additional object of the invention is to provide a radio frequency switch which may be used in conjunction with a scanning antenna system for switching in and out diterent physical lengths of transmission over which the radio frequency energy travels from the input to the antenna elements.

A further object of the invention is to provide a scanning antenna system where the phase front of the beam may be controlled through selection of actual transmission line lengths which can be precisely made and/or adjusted to have the required electrical length and will thereafter maintain that length independent of signal strength and temperature.

Various other objects and advantages will appear from the following description of the invention, and the novel features will be particularly pointed out hereinafter in connection with the appended claims.

The invention will be better understood when taken in connection with the following detailed description and drawings, in which:

FIG. 1 is a schematic diagram of a diode-controlled single-pole double-throw radio-frequency switch;

PEG. 2 is an embodiment of a two step delay line;

FIG. 3 illustrates an embodiment of the invention comprising a four step delay line;

FG. 4 is a schematic diagram of the phase shift scan of the linear array;

FIG. 5 is a schematic diagram of an array incorporating the delay line used for phase shift scan;

FG. 6 is a block diagram of one embodiment of the beam steering system.

The circuit that is utilized as the single-pole doublethrow radio-frequency switch is set forth in the schematic of FlG. 1 and is inherently a narrow-band device, utilizing the properties of quarter-wave-line transformers in conjunction with possible wide variations in effective irnpedance of semi-conductor diodes to furnish, alternately, conducting and isolating paths. Thus, in FIG. l there are shown signal sources or terminations 100, 102 and 103 and junctions 104, 105 and 10e. nput 100 is connected to junction 104; input 102 is connected to junction 105; input 103 is connected to junction 106; and junction 104 is connected to both junctions 105 and 106 through lengths of transmission line. Junction 105 is connected to a cathode of a semi-conductor diode 107, the anode of which is connected through a capacitor 10S to ground. The anode side of the semi-conductor 107 which is connected to one side of capacitor 103 is also connected to one side of an inductance 109 and then through a resistor 110 to a source of bias voltage, not shown.

Junction 106 is connected to the cathode element of a semi-conductor 111 and the anode of semi-conductor 111 is connected through a capacitor 112 to ground and also through an inductance 113 and resistor 114 connected in series to a source of bias voltage, also not shown.

In the configurations shown in FEG. l it is assumed that all of the transmission line shown has characteristic impedance Re, that signal sources or terminations at 100, 102 and 105 are matched to Re, and that the three junctions 104, 105 and 105 are shunt. In operation let it be considered that a matched connection is to be made between inputs 100 and 102, and that input 103 is to be isolated. It is then necessary that impedance from junction 104, looking toward junction 105, appears as Rc and looking toward junction 106 appear as infinite impedance, i.e., open circuit. This requirement will be met if the stub at 105 introduces negligible shunting, i.e., appears as infinite impedance or open circuit, and the stub at 106 appears as zero impedance, i.e., short circuit, which would transform through the length of line to open circuit at junction 104. The converse will, of course, be required when input is to be connected to input 103 with input 102 isolated.

The required conditions for the stubs will then be met by terminating each with a semi-conductor diode to ground, and the alternative requirements of short circuit and open circuit (in the practical case, low impedance and high impedance) at the input to a stub are to be achieved by alternative states of D.C. bias, with appropriate lengths of line transforming the effective impedance of the diode combinations. Since, in the example, the closed path 100 to 102 is to carry signal, which is to pass through the switch at minimum loss, the negative-bias condition is selected for the diode 107 associated with junction 105. Since a very high speed diode of the type under consideration appears as a very small capacitance when biased in this manner, the diode 107 with its mounting will appear t primarily as a capacitive-reactance which would then be transformed, through a length of line greater than and less than to appear as a very high impedance. Conversely, the diode terminating the same length of stub at junction 106 would be biased sufficiently positive to make it appear as a relatively low impedance, which would transfer to a still lower impedance in shunt at junction 106.

Alternatively, forward (positive) bias may be used, causing the diode to appear as a low resistance plus a small series inductance in which case the length of line from junction to diode 107 would be transforming to a high impedance as seen from junction 105.

FIG. 2 illustrates a two step delay line consisting of two single-pole double-throw radio-frequency switches and 121 respectively. The switches 120 and 121 are shown as blocks, however, the respective switches consist of the elements shown in FIG. 1. The arrangement of switch 121 corresponds to the switch shown in FlG. 1 wherein the terminals 100, 102 and 103 correspond to the terminals 100, 102 and 103 of FIG. l. The terminals 100', 102 and 103 of switch 120 correspond to the corresponding terminals 100, 102 and 103 of FIG. 1. Between terminals 103 and 103 is inserted a length of transmission line 122, and between terminals 102 and 102 is another length of transmission line 123. The physical lengths of transmission lines 122 and 123 are different so that the path of energy from terminal 100 to 100 is different through the two sides of the configuration.

Thus, a step delay line is achieved by the combination of two single-pole double-throw switches interconnected with appropriate lengths of transmission line. The variable delay is given by the difference between the lengths of transmission line connecting 102.', 102 and that connecting 103, 103. It is to be noted that the variation is in physical path length and not directly in electrical length or phase, which will be discussed later as advantages for the scanning function.

FIG. 3 illustrates a four step delay line comprising switches 124, 125, 126, 12'7, 128 and 129. Associated with switch 124 are terminals 100, 102 and 103; associated with switch 125 are terminals 1001, 1021, 1031; associated with switch 126 are terminals 1003, 1023 tween the driving source and the driven element a delay 4is acquired which is frequency independent. In order to illustrate the use of the delay line in an array reference is made to FIG. 5 which shows an end-fed array. To scan the beam by introducing step-delay-line elements, such as suggested in FIG. 3, into the network of FIG. 5, consider the relationship demonstrated in FIG. 4. The endfed `configuration allows the most straightforward application. To scan the beam at any angle, I say clockwise in FIG. 5, it is necessary to introduce relatively delay D between the input 501 and the element 502, 2D to element 503, and so on to 23D in the total path from input to element 524. In general terms, the maximum required delay is (N-l)D Where N is the number of elements and ,D equals s sin 1i. Since the delay between each element and its neighbor must be just D, the simplest possible arrangement for the delay structure is to insert one delay element between each tap on the feed line as illustrated in FIG. 5.

Actually, this procedure is not as straightforward as it looks in that if a network is used which is designed to provide a broadside beam, adding delay corresponding to a small electrical length of line serves only to scan the beam through positive values of I clockwise in FIG. 5. For scan counter-clockwise from broadside, it would be necessary to add delay suflicient to generate positive o approaching 360 degrees, that is; the electrical length of D would have to be in the fourth quadrant, approaching one wavelength. The simple geometrical relationship of FIG. 4 no longer holds in that case; instead, it is necessary to use the electric-lengths relationship:

D s of) Sm i and modify it to include negative values of phi by setting D L L' (MU-1+ sin o1. D (M) (Us sm I or, where TEM air line is used \j= D=?\{s sin '1A This solution, however, throws away the bandwidth advantage of the purely physical-length relationship, i.e., will not be invariant when l) is fixed because .\j lis -a function of frequency. Y

The preferred way to approach the problem would be to recall that a phase reversal at every other element is usually employed in a spaced array to achieve broadside operation, for example, by placing wave guides slots on either side of center line, or by reversing dipole connections to a balanced transmission line. If that phase reversal were eliminated, there would be an inherent phase delay between adjacent elements of so that the necessary added delay to obtain counter-clockwise D would be reduced by that amount, resulting in for Aj=k (lambda): D=)\-s,(1-sin d1). However, this serves only to reduce the amount of delay to be added, and does not reduce the frequency dependence of I now sin (1):@

as against the previous s A reduction in frequency dependence is achieved only by using a larger spacing, s, but this avenue is limited, by pattern considerations, to a small amount over when absolute b is-to be large.

The block diagram of FIG. 6 depicts in a more practical manner how the various components are assembled in a working embodiment of the system. Delay No. l and all inclusive delay sections up to Delay No. (N -1) are identical to the four step delay line of FIG. 3 and hence are numbered in an identical manner. Control signal lines 601 through 612 have been added to functionally represent how switching is accomplished. In actuality lines 601-612 each comprise two leads, one for switching each diode of the two diode SPDT switches. The array of FIG. 6 is shown as end fed, as described with respect to FIG. 5. Terminal 613 is a radio frequency terminal for connecting the antenna system to the desired transmitter or receiver.

It should be understood that the delay sections of FIG. 6 only have 4 switchable delay paths and thusly only 4 beam directions are possible.

In operation with a transmitter, the switching circuits, not shown, would program all the delay units to have a certain delay time. A signal from terminal 613 would therefore appear at the Various antennas delayed by an amount equal to the number of delay sections between the particular antenna and point 6M.

The use of the step delay line to obtain scan control is not limited to 'an end-fed configuration as disclosed in FIG. 5 and FIG. 6 in that it is fairly obvious that a center fed structure may be utilized if a power spread is required or desired. In addition various types of corporate feeds or corporate structures may be utilized as required or desired.

The parameters that govern the design of the basic delay unit of the type specified are: total scan angle; value of smallest mean step angle; array-element spacing s; the transmission wave length, M, of the line used, the selection of which must depend on available space; ruggedness required; and similar mechanical considerations.

The required total scan angle of plus or minus 45 from broadside, has been taken as that desired. The basic beam step, or angular resolution, has been selected to be 3. These two figures together set the minimum number of steps in each delay unit at 30. Thus, in theexample of FIG. 5 each delay unit would consist of 30 steps. This leads to `a fairly high number of diodes unless some refinement is done. y

It should be noted that, in view of the sinusoidal relationship between cI and D, for t'he large total range of r to be covered, the steps in fr will not be equal for equal steps in D, or vice versa. Bypassing, for the moment, the matter of this non-linearity and the errors that may result from ignoring it, a substantial reduction in the number of switches can be shown to result from the use in tandem of two or more delay units having different size steps. For example,.a -three division unit, having, say four coarse `steps followed by four medium steps then three fine steps, would result in a 36 step unit with only 6-i-44-4=14 switches required. This process may be extended to a Variety of combinations, the ultimate being a binary system, i.e., 2 2 2 2=32 steps, requiring 2|2l2+2l2=10 switches. This reduces the requirement of the number of diodes to a reasonable figure.

As noted in the discussion with respect t-o FIG. 5 the feature that was objectionable about this configuration is the frequency-dependent delay unit that is required for the series-feedlarr-ay. This can be obviated if variable `delay lines accomplish the scanning. The basic delay unit may vary in .length between two Values, the mean between these becoming part of the com-mon length of path :between the input terminal :and elements. This mean length, D0, yis then .the delay setting for the broadside beam position, `and we now have Di=D0zts sin d. D0 is not related to Aj; it should be as lsmall as practicable, however governed by the limit; Dos sin @mam with whatever additional length is needed by the physical design of the elernent.

This .regained frequency independence results in an increased cost through the use of additional length of terand 1033; associated with switch 127 are terminals 1004, 1024, 1034; associated with switch 128 are terminals 1005, 1025, 1035, and associated with switch 129 are terminals 1005, 1026 and 1036. Connected between terminals 1025 and 1021 is a length of transmission line 130; connected between terminals 1023 and 1024 is a length of transmission line 131; connected between terminals 1026 and 1004 is a length of transmission line 132; connected between terminals 103,, and 1005 is a length of transmission line 133; connected between terminals 1035 and 1031 is a length of transmission line 134; connected between terminals 1034 and 1033 is a length of transmission line 13S; connected between terminals 103 and 1001 is a length of transmission line 136; and connected between terminals 102 and 1003 is a length of transmission line 137.

As can be seen from FIG. 3 the length of transmission line 132 difers from the length of transmission line 133; the length of transmission line 131 differs from that of 135; and the length of transmission line 130 dilfers from that of 134. FIG. 3 shows that between terminals 1006 and 100 there are four possible energy paths depending upon the position of the switches 124-129. The rule for forming delay lines is quickly evident: for M equal number of steps and N equal number of switches,

Thus, for example, a 30 step delay line requires 58 switches, or 116 diodes.

The formation of the pattern of an array antenna, and the control of the direction of its principal radiation, or beam, is merely a matter of addition of vectors. Patterns of high-gain Vantennas are most often plotted as relative power on a logarithmic power, linear db, scale, as a function of angle measured from some axis of the antenna structure. In the example of FIG. 4 the azimuth angles are measured from the normal to the array. When the antenna is considered as transmitting, the plot is of relative power density at any fixed distance from the antenna beyond a certain distance which sets the inner limit of the far Zone. When the antenna is considered as receiving, the plot is of relative power at the terminals when a distant power source remains at a xed radius. The situations are reciprocal and the two plots are identical. The vectors that add to form the pattern are, for the transmitting case, the field strength vectors that can be attributed to the currents on the radiating elements, one vector for each element; assuming identically shaped elements, with identical forms of current distributions. The amplitudes of the vectors may be taken to be directly proportional to the amplitudes of the corresponding element currents and the phases will be in the same relation as the phases of the respective elementl currents but with the addition to each of the phase angle equal to the difference in the electrical path length from the element to the point in the far eld.

The relative amplitude of the vector sum, when squared, is the quantity that is plotted for the familiar power pattern. It should be evident thatl this sum will obtain maximum possible value whenever all vectors line up in phase. It is the objective of directive antenna design to achieve this condition in the direction selected for beam peak. The direction of the maximum preferred to the normal, defines the scan angle In For the receiving case, a source substituted for a receiver at the far field point, the reciprocal relationship results in currents being induced on the radiating elements with amplitudes and phases in the same relation as those that were responsible for the farlield signal in the transmitting case.

Since the point in space is taken to be at a great distance from the antenna relative to the antenna dimensions, the paths of radiation from each element on the antenna to the far-field point are taken to be parallel. Then, if there is a direction in which all element contributions will add in phase, a line drawn perpendicular to that direction will intersect the radiation paths at points of equal phase. Such an equal-phase contour is called the phase front. Hence, the phase relationships introduced by differences in path length follow a simple geometrical relationship, which can be demonstrated by consideration of an array of a very few elements.

In FIG. 4 there is shown a linear array of four elements backed by a ground plane, the elements being equally phased from each other and from the plane. In the phase plane of the figure, taken normal to the ground plane, the radiation pattern of the individual elements will be assumed to be broad, i.e., essentially circular in the region of space to be examined. Then if there is a direction angle qb equals 12 in which the radiations are to add in phase, there will be linear phase fronts perpendicular to that direction. Consider such a phase front drawn through element 140. It is at the angle I to the line of the array, and taking the element spacing as s, simple trigonometric relationships relate the phases of the element currents as follows: The current in element 141 must be leading the current in element 140 by a phase angle -360 Similarly current in element 142 must lead that in element 140 by sin I degrees Likewise,

and so 0n for any number of elements in the array. The difference in phase between any two adjacent elements is always s 360 S111 I so assuming linear phase relationships throughout, the relative phase of any pair of adjacent elements, is the fundamental unit that, with spacing s and wave length (A), sets the scan angle dx The requirement for a broadside beam, tI =0, is met when all element currents are in phase. This is taken as the starting point, or reference condition, of the scanningarray design. In order to scan to angle fr, phase delay of current to element 140 with respect to that of element 141 must be introduced. If this is done with a phase shifter that introduces a delay assuming the transmission line to have free space propagation velocity, and the simple frequency-independent rela` tionship holds: D=s sin 1, a purely geometric relationship that is evident in FIG. 4. If the delay line used has dielectric other than air, or is line of other than tem mode configuration, the difference between line and free-space wave length must be taken into account thus:

equals 360 D=s sin (I) Thus, through the use of a longer physical path be- 9 minal-element path. That length must inclu-de the physical distance Ito the furthest element, taken over .the shortest practical path, plus D multiplied by the number of unit delay elements in that path, i.e., for the array with an odd number of elements n, add

yIt will be understood that various changes in the de-tails, materials, steps and arrangements of parts, which have been herein described and illustrated in order to explain the nature of the invention, may be made by those skilled in the art within the principle and scope of the invention as expressed -in the appended claims.

What is claimed is: 1. An electronically-sterable-beam comprising;

Ian array of N antenna elements where N is a whole number greater than one, (N-l) delay means connected in series with each other, each of said elements of said array being connected to Irespective points along said series of delay means, each of said delay means compising a plurality of `different lengths of transmission line paths operatively connected to a plurality of semiconductor diode single-pole-double-throw switches, said switches bein-g operable to selectively choose one of said plurality of different lengths of transmission line paths whereby the delay introduced by said delay means is varied in accordance with the line length selected, and radio-frequency terminal means connected to 'a tirst element of said array whereby energy appearing at said first element will appear at .other array elements antenna system subject to increased delay with increased distance Ifrom said rst element.

2. The .system of claim `1 wherein said switches comprise;

rst and second terminals interconnected by a .transmission line having a length of one-half wave at a dresired operating frequency,

another transmission line,

said another line having one of i-ts ends connected midway on said half-wave line and the other of its ends connected to a third terminal,

two semiconductor diodes,

each of said diodes being connected between a respective one of said Iirst and second terminals and a grounding point, and

lmeans for selectively biasing said diodes in their conducting or non-conducting states whereby said `first and second terminals may be selectively grounded and whereby radio frequency energy may selectively pass between one of said first and second terminals and said third terminal.

References Cited by the Examiner UNITED STATES PATENTS 2,160,857 6/39 Gothe 343-854 2,415,242 2/47 Hershberger i-- 333-7 `2,418,124 4/47 Kandoian v343-813 X i3,038,086 6/62 Sterzer 307-885 3,056,961 10/62 Mitchell 343-854 3,069,629 12/62 Wolf 343-854 3,076,155 1/63 Parker 333-7 3,131,367 4/64 Pitts et al. S33-31 ELI LIEBERMAN, Primary Examiner.

Claims (1)

1. AN ELECTRONICALLY-STEERABLE-BEAM ANTENNA SYSTEM COMPRISING; AN ARRAY OF N ANTENNA ELEMENTS WHERE N IS A WHOLE NUMBER GREATER THAN ONE, (N-1) DELAY MEANS CONNECTED IN SERIES WITH EACH OTHER, EACH OF SAID ELEMENTS OF SAID ARRAY BEING CONNECTED TO RESPECTIVE POINTS ALONG SAID SERIES OF DELAY MEANS, EACH OF SAID DELAY MEANS COMPRISING A PLURALITY OF DIFFERENT LENGTHS OF TRANSMISSION LINE PATHS OPERATIVELY CONNECTED TO A PLURALITY OF SEMICONDUCTOR DIODE SINGLE-POLE-DOUBLE-THROW SWITCHES, SAID SWITCHES BEING OPERABLE TO SELECTIVELY CHOOSE ONE OF SAID PLURALITY OF DIFFERENT LENGTHS OF TRANSMISSION LINE PATHS WHEREBY THE DELAY INTRODUCED BY SAID DELAY MEANS IS VARIED IN ACCORDANCE WITH THE LINE LENGTH SELECTED, AND RADIO-FREQUENCY TERMINAL MEANS CONNECTED TO A FIRST ELEMENT OF SAID ARRAY WHEREBY ENERGY APPEARING AT SAID FIRST ELEMENT WILL APPEAR AT OTHER ARRAY ELEMENTS SUBJECT TO INCREASED DELAY WITH INCREASED DISTANCE FROM SAID FIRST ELEMENT.

Images