GB2258949A - A transmission delay device for beam steering or selection - Google Patents

Abstract

An array of transceivers (3) each comprised of a pair of crossed orthogonal planar metal dipoles (20 & 21) lying adjacent a common body (5) of high dielectric constant material has control switches (32 to 35) connected between adjacent pairs of dipole arms (22 & 23, 25 & 22). Different or like switch states at adjacent transceivers (3) are used to select relative phase differences between transceivers (3), the totality of switch states being chosen according to each of several predetermined patterns, each pattern corresponding with a different select beam direction. The array may be combined with a transmitter and used to steer a transmitted beam into any one of several select directions. Alternatively the array may be combined with a receiver and used to select input radiation from any one of several select directions. Transceiver-to-transceiver phase differences may be modified using variable length dipoles (20), dipoles having detached parts (22, 24) connected by further control switches (62, 64). As further alternative, the array (3) may be combined with a transmitter (TX), a receiver (RX), and a polarisation sensitive mirror (17) and used for switching between a receive mode (all switches on) and a transmit mode (all switches off). <IMAGE>

Description

A TRANSMISSION RELAY DEVICE FOR BEAM STEERING ON SELECTION TECHNICAL FIELD The present invention concerns a transmission relay device for beam steering or selection; and, in particular, a relay device of the type comprising an array of transceiver elements, each element having a receiver antenna, a transmitter antenna and a phase shift control component providing feed from one antenna to the other. As a beam steering device, such a relay will receive radiation from a fixed transmitter and re-transmit radiation in a direction determined by the phase shifts set for each of the array transceivers. The radiation beam may be steered to an alternative direction by resetting the phase-shifts for each array transceiver.Whereas, a relay arranged for beam selection, will direct radiation received from a selected direction, one of several different directions, to a fixed receiver, the direction selected being determined by the phaseshifts set for each of the array transceivers.

BACKGROUND A space-fed relay device used at radar frequencies 3 to 8 GHz is known comprising an array of horn-fed antennae, and backing this array another array of antennae each having a horn output. Typical hornto-horn centre spacing across each array is between 5 and 10 em.

Each horn-fed antenna is coupled to a corresponding antenna in the other array by a controlled phase-shift network. The device is costly and bulky.

There is a need for relay devices operable for higher frequencies 10 to 100 GHz.

DISCLOSURE OF THE INVENTION The invention is intended to provide compact relay devices, and in particular devices operable for radiation of a frequency in the 10 to 100 CHz range or thereabouts.

According to the invention there is provided a transmission relay device comprising an array of transceivers, wherein the transceivers are mounted adjacent and in close proximity to the surface of a common body of high dielectric constant material, each transceiver comprising a pair of planar metal antenna dipoles crossed and orthogonal one to the other with a network of control switches providing connection between the pair of dipoles, one switch connected across each pair of adjacent orthogonal dipole arms, each switch having, in its conducting state, a high-frequency impedance that is low compared with radiation resistance.

The construction is thus that when incoming radiation is coupled to one of the antenna dipoles of a transceiver and the switches are set so that one pair of opposite switches conduct whilst the other pair do not, power is efficiently transferred from the one dipole to the orthogonal dipole and re-radiated. On reversal of the switches, a signal of reversed polarity is re-radiated. A phase-shift of either 00 or 1800 may thus be introduced, depending upon which of the switch pairs is conducting.Thus when all the array switches are set and radiation is directed onto the array, the power re-radiated from the relay device will be concentrated in directions where the contributions from the radiating transceivers reinforce, in manner similar to the diffraction of X-rays from a crystal or of light reflected from a grating: These directions are determined by the set phase-shifts.

Thus, with appropriate choice of switch combinations for the transceiver array, the direction of the re-radiated radiation may be controlled.

The device may thus be used in conjunction with a control logic circuit, a circuit for controlling the selection and combination of the switch settings, each combination of switch settings corresponding to a different beam direction.

For lower frequency, long wavelength applicatIons, - eg for X-band, the device may be of hybrid construction - ie the switches may be discrete semiconductor components. The dipoles may thus be fabricated from thin or thick metal fiim on alumina ceramic.

For higher frequency, short wavelength applications - eg for Q-band, the device may be of integrated construction - ie the switches may be incorporated in semiconductor material and integrated with the antenna metal.

In an integrated construction the common body may be of high dielectric constant semiconductor material - eg high resistivity silicon Si (e r 10) or gallium arsenide-GaAs ( - 12). Alternatively the common body may be a composite of semiconductor material and insulating dielectric material, the semiconductor material being located between the array and the dielectric material. It is preferable that in the composite, the insulating dielectric material exhibits low dielectric loss, high thermal conductivity and a dielectric constant fairly close to that of the semiconductor material.

Examples of composites are silicon or gallium arsenide bonded on high purity alumina ceramic (c r 10) or on beryllium oxide ceramic (e ~ 7).

In all these cases the semiconductor material serves to incorporate the switches as also to propagate both incoming and re--radiated radiation.

In an alternative and preferred integrated construction the array may be mounted between semiconductor material and the common body. It is preferable in this case, that the dielectric constant of the common body material is significantly higher than that of the semiconductor material, and such, therefore, that the common body serves as principal propagating medium for both incoming and re-radiated radiation. An example is silicon on barium titanate (e r 39).

BRIEF INTRODUCTION OF THE DRAWINGS In the drawings accompanying this specification: FIGURE 1: is a cross-section view of a transmission relay device, a device including an array of integrated transceiver elements; FIGURE 2: is a plan view of a part of the array as viewed from plane X-X normal to the cross-section of figure 1; FIGURE 3: is a schematic enlarged plan view of one of the transceiver elements shown in the preceding figures; FIGURE 4: is a cross-section view of a PIN diode switch, a switch included in the transceiver element of figure 3 above; FIGURE 5: is a schematic diagram illustrating transceiver elements connected by a network of bias control conductors; FIGURE 6: is a simplified perspective view of an array of transceiver elements, showing array lattice axes; FIGURES 7 & 8: illustrate two arrangements of reflectors used with an array of transceiver elements; FIGURE 9: is a graph showing the dependence of a beam amplitude parameter |F | on a direction parameter "z" for different transceiver switch combinations B B2, and B FIGURE 10: is a plan-view of a modified transceiver element; and, FIGURE 11: is a cross-section view of a multi-mode transmission relay device, a modification of the device shown in figure 1 above.

DESCRIPTION OF EMBODIMENTS Embodiments of the invention will now be described, by way of example only, and with particular reference to the accompanying drawings.

There is shown in figures 1 and 2 a transmission relay device 1 arranged and constructed for electromagnetic beam steering. In the preferred construction shown, an array of transceiver elements 3 is mounted upon the surface of a semiconductor chip 5 and is disposed between the surface of this chip 5 and the surface of a composite body 7 of insulating dielectric material, a body 7 formed of two optic components 9 and 11.The material of the composite body, at least the material of the body component 9 adjacent the array, has a dielectric constant (eel) considerably greater in value than the dielectric constant (2) of the material of the semiconductor chip, ie and > e2; andthe composite body 7 is either bonded in intimate contact with the array of transceiver elements 3, or it is held in fixed position very close to the array. This being so, the transceiver elements 3 will couple most predominantly to radiation on the composite body 7 side of the array. The composite body 7 thus serves not only as the propagating medium for incoming radiation, but also as the propagating medium for radiation re-radiated from the array of transceiver element 3. [See Electronics Letters Vol 17 No 20 pages 729-731 (October 1981)1.

The semiconductor chip 5 is bonded in good thermal contact with a metal block 13, to enable the effective extraction of heat dissipated in the array of transceiver elements 3. The thickness of the semiconductor chip 5, should exceed a minimum value set by the condition that the proximity of the metal block 13 to the transceiver elements 3 should not strongly affect the properties of the dipole antennae of the transceivers 3.Calculations show that a chip thickness of 400 micron is acceptable for a semiconductor dielectric constant close to 12 (eg for Si or GaAs) and a frequency of 35 GHz.. This minimum thickness varies as lifrequency. For lower frequencies where a much greater semiconductor thickness may be inconvenient, a dielectric layer may be interposed between the metal block 13 and the semiconductor chip 5 to increase the metal to transceiver spacing. For a silicon chip, a spacer of high purity alumina ceramic is suitable The array of transceiver elements 3 is coupled to the radiation field of a transmitter antenna 15.In the compact arrangement shown, this antenna 15, a dipole, is mounted on a surface of the composite body 7, and the radiation R from this antenna 15 is deflected onto the array t by means of a curved mirror 17 which is incorporated in the composite body 7 between the optic components 9 and 11. This mirror 17, which is comprised of a grating of parallel metal stripes, is polarisation selective. It is arranged with these stripes disposed parallel with the plane of polarisation of the transmitted radiation Rt emanating from the transmitter dipole antenna 15, so that for this polarisation the radiation is reflected onto the array. The position of the mirror 17 relative to the transmitter antenna 15, its inclination and its curvature, are all such that radiation is collected from the transmitter 15 and deflected as a collimated beam onto the array.As will be explained below, when the transmitted radiation Rt is incident upon the transceiver elements 3, radiation is absorbed and re-radiated.

The plane of polarisation of the re-radiated radiation Rr is however orthogonal to that of the incident radiation. The re-radiated radiation Rr is not reflected by the polarisation selective mirror 17, for the plane of polarisation is across the metal stripes and not parallel to them. The re-radiated radiation Rr instead is refracted at the interface between the components 9and 11 of the composite body 7 and thereafter it is coupled to the radiation field of the ambient medium 19 - air.

A typical example of this preferred construction is one designed for Q-band ( 35 GHz) operation and 200 Watts maximum reflected power, using a chip 5 of silicon semiconductor material (C2 ~ 10), 400 pm thick and 1 cm square bonded to a copper sink 13. A composite body 7 of barium titanate (Ba2Tig020) ceramic material (e - 39.5) is used and the clearance between this body 7 and the array of elements 3 is of between 0 and 5 pm maximum. Typical centre-to-centre spacing of the transceiver elements 3 for the design frequency (35 GHz) is 800 pm.

One of the transceiver elements 3 of the relay device 1 is shown in detail in the enlarged plan view, figure 3. Each element 3 is comprised of two crossed dipoles 20 and 21, orthogonal one to the other. One of these dipoles, dipole 20, has two spaced arms 22 and 24 of rectangular pattern sheet metal, metal deposited on the surface of the semiconductor chip 5. To prevent formation of intermetallic compounds, there is a barrier layer 6 of insulating material included between the semiconductor material of the chip 5 and the metal of the transceiver elements 3 - eg thermal grown silicon oxide typically 0.5 pm thick, on the silicon semiconductor material. It is thick enough to provide chemical isolation, but thin enough to have no appreciable effect on the electrical properties of the antenna dipoles 20 and 21 in the relay device 1.The other dipole 21 has two arms 23 and 25. The length of each dipole 20 or 21 is chosen according to the design frequency. Ideally, these two lengths are chosen so that both dipoles 20 and 21 are resonant at the design frequency, each to act as a half-wave resonant current dipole. The half-wave resonant length 1Xeff is given by the simple relation below:

; - where e1 > E2 In these expressions the symbols Avac and A represent wavelength in vacuum and in the dielectric body 7 (component 9), respectively.The resonant frequency, frs of a dipole, defined as the frequency at which the dipole impedance is purely resistive, is slightly below the frequency, fO, for wllich 2Reff equals the dipole length. This reduction depends on the width of the dipole element, eg, for a dipole with a width of 10% of its length, the resonant frequency, for is about 85% of fO. For the maximum efficiency of transfer of power from one dipole to the other, the impedances of the dipoles should be complementary, ie the resistive parts should be equal and the reactive parts equal and opposite.Where the dipoles 20, 21 are of equal dimensions, their impedances are equal, hence for the best power transfer efficiency both should have purely resistive impedances, ie the operating frequency should be fr or close to it. An alternative arrangement is to make the dipoles 20 and 21 of unequal length, the longer one presenting an inductive and the shorter one a capacitative reactance. Their resistances should be approximately equal, and this will obtain as long as the dipole lengths are not too different, typically the fractional length difference should be less than 20%.

The dipoles 20 and 21 are connected by a network 30 of controlled switch elements, switches 32 to 35. These switches 32, 33, 34, and 35 are connected one between each pair of adjacent arms, pairs 22 and 23, 23 and 24, 24 and 25, and 25 and 22 respectively; arranged in a ring at the centre of the two crossed dipoles 20 and 21. The network 30 of control switches 32 to 35 serves to transfer power from one of the two dipoles 20 and 21, dipole 20, to the other dipole 21. So that this transfer is effecient, it is also requisite that each of the switches 32 to 35, when ON - ie when it is in its conducting state, has an hf impedance, ie at the design frequency, that is low compared with the radiation impedance of the transceiver element 3. The switch impedance, when OFF - ie when each is in its non-conducting state, is as high as possible. The switches 32 to 35 are arranged to be controlled and set wit one opposite pair 32 and 34 ON, with the other opposite pair 33 and 35 OFF, as also vice-versa - 32 and 34 OFF with 33 and 35 0N, when addressed by appropriate control signals - dc or low-frequency switch control signals.

The construction of a suitable switch 32, a PIN diode is illustrated in figure 4. The PIN diode switch 32 is formed in the semiconductor chip 5 and has metal contacts, arms 22 and 23. This switch 32 includes a buried dopant enriched Cn+) layer 37 and a diffused dopant enriched (n ) connective region 39 extending between the buried layer 37 and the surface of the semiconductor chip 5. Ohmic contact is formed between this connective region 39 and one of the metal contacts, arm 23, through a window in the barrier layer 6, to form one pole of the switch 32. The other pole of the switch is provided by another diffused dopant enriched region 41 - of different polarity type (p+) - which is in ohmic contact with the other switch contact, arm 22 through a second window in the barrier layer 6.A region of low doped semiconductor material, region 43, sandwiched between the + p region 41 and the buried layer 37 - typically 1 to 2 microns thick and 10 to 20 microns diameter - serves as the I-region of the diode.

This region 43 may be made conductive by applying a forward bias current to the switch contacts 22 and 23. The non-conductive state (ie a state of high impedance) is obtained by applying a reverse bias voltage. In this condition the reverse bias current is very small. Switches of alternative construction, of suitable high/low high-frequency impedance, include: + i) Bipolar Transistors - eg n upon - having a thick, highly doped, base region.Control signal, a control voltage, is applied + between thep and one, or both, n regions; ii) Schottky Barrier Diodes - eg metal-nn - which presents a significantly lower hf impedance with increase in forward bias, and, iii) Field-Effect Transistors - in which the source and drain regions are connected to the antenna dipole arms, and bias voltage is applied to the gate to control the source-driven impedance.

Bias control for the PIN-iodes, switches 32 to 35, is provided by a logic circuit integrated in the semiconductor material at the periphery of the chip 5. The arrangement of connections, between this logic circuit and the switches 32 to 35 of the transceiver elements 3, is shown in figure 5. To simplify the design of the logic circuit 45, and to isolate the diode-switches 32 to 35, each of the arms 22 to 25 has been subdivided into two dc isolated strips 22A and 22B to 25A and 25B. Provided that the arm portions are close-spaced, there is significant reactive coupling between the two strip portions A and B at high frequency and each arm behaves in manner similar to an undivided arm.If desired, the reactive coupling can be enhanced, and impedances trimmed, by overlaying each divided arm with an isolated strip metal electrode (not shown). The isolated arm portions 22A, 22B, 24A and 24B of dipoles 20 are connected in pairs to common ground rails, metal conductors 45, through parallel high resistivity 1 KQ sheet resistors 47. The isolated arm portions 23A, 23B, 25A and 25B of the other dipoles 21 are connected in pairs each to a voltage rail 49, one rail corresponding to each transceiver element 3. Voltage signals + V, are applied to each of these rails 49 to apply an appropriate bias.The diode switches 32 to 35 are arranged head-to-tail in a ring configuration, and so when a voltage of + V is applied, one pair of opposite switches is ON (pair 32 and 34 when voltage - V is applied or, pair 33 and 35 when voltage + V) whilst the other pair of opposite switches is OFF (pair 33 and 35 when - V, or, pair 32 and 34 when voltage + V). A different switch combination can thus be set for each transceiver element, according to whether a i V voltage is applied. The pattern of different switch combinations set is determined by the outputs of the logic circuit. The logic circuit is designed to give different combination patterns to effect different deflections of the re-radiated beam.

The lattice constant of the array, the centre-to-centre spacing "d" between the transceiver elements 3, is chosen to be of value close to the wavelength X of radiation in the composite dielectric body 7 (component 9). For spacings "d" less than X, the array of transceiver elements 3 presents a very effective capture cross-section to incident radiation. This capture cross-section, however, would fall off very rapidly if "d" were greater than A. To leave space for peripheral control circuit connection, and to keep transceiver-to-transceiver coupling low, the spacing 'td't needs to exceed the dipole length X significantly. Thus the chosen spacing d h X arises as a natural constraint, taking all these requirements together.

During operation, incident radiation I/P is directed onto the array with its plane of polarisation parallel to dipoles 20, the receiving dipole-antenae of the array. Opposite switches 32 and 34 are set ON and the other opposite switches 33 and 35 are set OFF. For this first combination of the switches, arms 22 and 23 are tied, and arms 24 and 25 are tied. Thus a current in dipole 20 in the positive y direction as indicated gives rise to a current in dipole 21 in the negative x direction. The plane of polarisation of the re-radiated signal is orthogonal to that of the incident radiation, and corresponds to a rotation of + 900. These switch states are now reversed, opposite switches 32 and 34 are set OFF and the other opposite switches 33 and 35 are set ON. For this second combination of the switches, arms 22 and 25 are tied, and arms 23 and 24 are tied. In this case a current in dipole 20 in the positive y direction gives rise to a current in dipole 21 in the positive x direction. As before, the plane of polarisation of the re-radiated signal is orthogonal to that of the incident radiation, but it corresponds to a reverse rotation, a rotation of - 900. There is thus a 1800 phase-shift difference between the two alternative switch combinations.

This property is utilised in the reconstruction of a directional beam. The shape and direction of the re-radiated beam will depend on the pattern of switch combinations selected, (ie it will depend upon the various phase shifts Oo or 1800 inserted at each transceiver element 3 of the array), and also upon the direction of incident radiation, and on the size and spacing of the transceiver array.

These dependencies will now be illustrated by example.

Consider the 4 x 4 transceiver array shown in figure 6. Reference axes z- and y- are chosen parallel to the array axes, and each transceiver 3 is arranged with the axis of its radiating dipole 21 inclined at an angle fI to the lattice axis x. The transceivers 3 are arranged in a square lattice, a distance "d" apart from centreto-centre. The direction of the incident radiation is chosen to lie in the plane containing the re-radiating dipoles 21 and the z-axis (the direction normal to the array plane). The incident direction is inclined at an angle 0' to the z-axis. The incident radiation is polarised parallel to the receiving dipoles 20.

To simplify illustration, a few basic assumptions are introduced:a) The incident radiation is a plane wave of uniform amplitude over the area of the array; b) The coupling between adjacent transceivers 3 is negligible; c) The polar diagram amplitude f(o,) of each re-radiated wave is similar to that from an isolated dipole; and, d) The amplitude of each re-radiated wave is a factor + 1 times the amplitude of the incident wave received. The phase-shift set is 0 or 1800.

The various phase-shifts set may be represented by a 4 x 4 matrix A, the coefficients an,m Cn = 1 to 4, m = 1 to 4) of this matrix being of value either + 1. A further simplification is introduced, by restricting this matrix to a matrix given by the product of two line vectors, a column vector B (coefficients b , n = 1 to 4) and a row vector C (coefficients cm, m = 1 to 4), where these coefficients En, c are of value + 1. In other words a .= b c .The polar diagram m n,m n m R(O,) for the re-radiated power is then given by the following expression: R ((#,#) Fx(#,#,#,#) F Fy(0,ss,6' > +') where:-

exp (jkdn(sin e cos + sin O' cos ')} exp {jkdm(sin 6 sin # + sin 6' sin #')} and k is the wave-vector 2ir/X for radiation in the dielectric body 9.

If the incident radiation is normal to the array, ie O' = 0, then the expressions for F and F are such that the modulus of F is the same x y x for (p = #1, -#1, # - #1 and s + 1 and the modulus of F is also the y same for these values of f. This means that however the coefficients b and c are chosen, there will be four re-radiated beams of equal n m power.This will not be the case for other incident beam directions for which 6' > o. However a 2-fold degeneracy will always occur if the incident beam direction lies in the xz or the yz planes, ie if +t = O or #/2 Eg if +' = 0, then the moduli of Fx and of F are the y same for = 1 and = For single beam applications, the re-radiated beams may be re-directed and combined by appropriate arrangement of one or more reflectors.

An example is shown in figure 7. The incident radiation direction lies in a plane containing one of the array axes and the normal to the plane of the array, but is inclined to the normal to the plane of the array. In this case there are two re-radiated beams symmetrically disposed about the plane containing the incident beam and the normal to the array. A mirror 51 is formed on a side face of the dielectric body 9. This face is parallel to the plane containing the incident beam and the normal to the array. The mirror 51 reflects one of the re-radiated beams and this reflected beam is parallel to the other re-radiated beam.

The mirror 51 can be made by metallising the surface of the dielectric body 9. Alternatively the surface may be unmetallised and total internal reflection used at the dielectric: air interface. This is straightforward because the re-radiated beam will be very weak for beams inclined more than 400 from the normal to t'ne plane of the transceiver array. For a mirror surface normal to the array plane, the angles of incidence for the radiation to be reflected will there fore exceed 50 , well in excess of the critical angle.

A variant of this method for a 4-beam radiator, is to use two mirrors 53, 55 to recombine the beams R1 to R4. In the preferred arrangement the incident radiation beam is normal to the plane of the array see figure 8. The advantage of mirror arrangements is that the reradiated beams are combined into a single concentrated main beam.

This is weighed against the disadvantages of restricted angular coverage and the presence of interference fringes within the resulting beam.

An alternative to the use of mirrors, is to align the incident beam away from the normal to the array and in a plane normal to the array that is parallel to neither one of the array axes. This has the effect of giving greater weight to one of the four beams.

This is illustrated in figure 9, a graph of the function Fx(z) IFx(Z) 12 computed for an angle of incidence O' = 200 and array angle ' = 450, for three sets of different coefficient vectors B - namely B1: (1,1,1,1), B2: (1,1,-1,-1); and, B3: (1,-1,1.-1). Thevectors C namely C1 to C are chosen from the same three sets, giving a total of nine different switch combinations A1 to A described bythe products B1C1, B2C1, B3C1,B1C2 etc to B3C3. For the first set B1, maxima occur for z = 0, + 1 ie sin 6 cos f = - a or 1 - a (where a = sin 20 cos 45 ). It is noted that only the - a value gives a large radiation peak. The other corresponds to a value of 6 of at least 500 where the dipole polar amplitude function f(6,(p) is always weak. In this case, as also found for the other sets B2 and B3, there is dne dominant beam direction. The same is found for the sets C1 to C3. Thus a total of nine different beams in all can be selected, by applying one of the nine different switch combinations.

The control of the re-radiated beam is not ideal because the modulation factor that is introduced at each array point is restricted to one of two values + 1 corresponding to discrete phase shifts of Oo or 1800.

Additional versatility can be added if other values of phase shift are introduced. Thus in figure 10 the electrical length of one of the transceiver dipoles 20 is stepwise variable. Extra switches 62 and 64 are included, one in each dipole arm 22 and 24. These switches 62 and 64 are switched together independently of the four central switches 32 to 35.

A change of length alters the dipole impedance, the main effect near resonance being to change the reactance (a shorter dipole being more capacitative) and this shifts the phase of the re-radiated signal at the expense of some impedance mismatch and consequent reflection of some radiation from the receiving antenna 20.

Further improvement of re-radiated beam shape and resolution may also be achieved by modifying the intensity of the incident radiation across the array and/or varying the transceiver-to-transceiver spacing across the array. The transceiver inclination angles T may also be varied, to vary the degree of coupling to the incident radiation from site to site.

Another way of using the transceiver array 3 is to connect the switching elements 32, 33, 34, 35 to a bias control circuit arranged so that all four elements may be switched to the ON state at the same time or all four may be OFF at the same time. There are then four choices of switch state, (a) all OFF, (b) one pair ON, (c) the other pair ON, (d) all ON. When either the all-ON or all-OFF case is selected the symmetry of the crossed antennae and switches is such that there is no re-radiation from the transceiver element with polarisation orthogonal to the incident polarisation. There will be re-radiated components with the same polarisation as the incident wave.For resonant half-wave dipoles, the effective cross-section of the transceiver will be large when the four switches are in the ON state because the transceiver is then a resonant dipole loaded by the switch elements, which present a load much smaller than the dipole impedance. The transceiver element reflects most of the incident radiation back into the dielectric block. If therefore all the transceiver array elements are set to the all-ON state, the array as a whole acts as a plane mirror. When all the switches are set to the OFF state, the coupling to the incident radiation is weak because the transceivers approximate to isolated sections of metal of about half the resonant length. If all the transceiver array elements are set to the all-OFF state the array becomes transparent to the incident wave.

Use may be made of this switching arrangement in a radar in which the array is, during the transmit mode, a polarisation-changing beam deflector as described previously. During the receive mode the switches in the transceiver elements are set to the all-ON state and the array becomes a mirror with unaltered polarisation. As shown in figure 11 there is a dielectric body 7, a transceiver array 3, a curved polarisation selective mirror 17, and a dipole transmitter 15.

In the transmit mode, the radiation from the transmitter 15 is reflected by the polarisation selective mirror 17 onto the transceiver array 3, and the direction of re-radiated waves of orthogonal polarisation is controlled as described above. In the receive mode, the array transceiver switches are set to the all-ON state and incoming radiation of the same polarisation as that transmitted, is reflected at the array 3 with unchanged polarisation. It propagates through the polarisation selective mirror 17 since its polarisation has not been changed by the transceiver array 3. The radiation is gathered by a separate receiver unit, Rx.

Claims (13)

1. What I claim is:1. A transmission relay device comprising an array of transceivers, wherein the transceivers are mounted adjacent and in close proximity to the surface of a common body of high dielectric constant material, each transceiver comprising a pair of planar metal antenna dipoles crossed and orthogonal one to the other with a network of control switches providing connection between the pair of dipoles, one switch connected across each pair of adjacent orthogonal dipole arms, each switch having, in its conducting state, a high-frequency impedance that is low compared with radiation resistance.
2. 2. A device as claimed in claim 1 wherein the common body is or semiconductor material, and the switches are incorporated as components integral therewith.
3. 3. A device as claimed in claim 1 wherein the common body is a composite of semiconductor and insulating dielectric materials, the semiconductor material being located between the array and the dielectric material and incorporating the switches as components integral therewith.
4. 4. A device as claimed in claim 1 wherein the array is mounted upon a substrate of semiconductor material and is located between the substrate and the common body; the common body being of insulating material of higher dielectric constant than that of the semi conductor material; the switches being incorporated as components integral with the substrate.
5. 5. A device, as claimed in any one of the preceding claims, including a transmitter antenna and a polarisation selective mirror, the mirror being arranged to direct transmitted radiation onto the array.
6. 6. A device as claimed in any one of the preceding claims, and including a control circuit, the circuit being constructed and arranged to switch on either one of two pairs of opposite switches for each transceiver, and capable of switching different switch pairs to introduce a relative phase shift of either 0 or 1800 between adjacent transceivers in accordance with any one of a plurality of different selectable patterns
7. 7. A device as claimed in any one of the preceding claims 1 to 5 wherein the electrical length of at least one dipole, for each of several of the transceivers, is stepwise variable, each dipole of variable length having a detached part at each of its extremities and a corresponding pair of further control switches for providing selectable connection between the detached parts and the dipole proper.
8. 8. A device, as claimed in claim 7, including a control circuit, the circuit being constructed and arranged to switch on different pairs of opposite control switches and different pairs of the further control switches in accordance with any one of a plurality of different selectable patterns.
9. 9. A device as claimed in either claims 6 or 7, including at least one reflector arranged relative to the array to redirect one re radiated beam into the direction of another simultaneously re radiated beam.
10. 10. A device as claimed in claim 9 including a pair of orthogonal reflectors arranged relative to the array to direct all re radiated beams into a single direction.
11. 11. A device as claimed in any one of the preceding claims 1 to 5 including a control circuit, the circuit being constructed and arranged to switch the control switches between an all-on switch state and an all-off switch state.
12. 12. A device as claimed in claim 11 including a transmitter, a receiver and a polarisation selective mirror, arranged such that for operation with switches in the all-on state the device is in a receive mode, and with switches in the all-off state the device is in a transmit mode.
13. 13. A transmission relay device constructed, adapted and arranged to operate substantially as described hereinbefore with reference to and as shown in the accompanying drawings.

    Amendments to the claims have been filed as follows 1. A transmission relay device comPrising an array of transceivers.

    wherein the transceivers are mounted

    close proximity to the surface of a common body of high dielectric constant material, each transceiver comprising a pair of planar metal antenna dipoles crossed and orthogonal one to the other with a network of control switches providing connection between the pair of dipoles, one switch connected across each pair of adjacent orthogonal dipole arms, each switch having, in its conducting state, a high-frequency impedance that is low compared with radiation resistance.