GB2330714A

GB2330714A - A matrix amplifier with unwanted outputs radiated away

Info

Publication number: GB2330714A
Application number: GB9828629A
Authority: GB
Inventors: Paul W Dent; Ross W Lampe
Original assignee: Ericsson GE Mobile Communications Inc
Current assignee: Ericsson Inc
Priority date: 1994-01-11
Filing date: 1995-01-11
Publication date: 1999-04-28
Anticipated expiration: 2015-01-11
Also published as: GB9828629D0; GB2330715B; GB2330466A; GB2330468B; GB9828635D0; GB2330468A; GB9828625D0; GB9828632D0; GB9828621D0; GB2330715A; GB2330466B; GB2330467A; GB2330467B; GB2330714B

Abstract

A "matrix" amplifierlaerial is arranged to flexibly amplify a plurality of signals with a plurality of amplifiers such that, for example, a group of amplifiers may be devoted to one signal, so increasing the possible output. Third order intermodulation signals produced in this process are radiated in harmless directions. Various other arrangements are described including: deriving constant amplitude signals from varying amplitude signals (amplitude information contained in phases) for application to non-linear amplifiers to increase efficiency, and the recovery of energy of intermodulation products.

Description

1 OC WASTE ENERGY CQNTROL NI-A-N,AGEfEN-T FOR POWER AEPLUU-R 2330714

Field of the Di-,;closure

The pre-sent invention relarts to ciass C mawx power amplifiers for 5 use in cellular radio communication systems, and particularly to a satellite cellular radio communication. L1 addition, the pre-se-ic invention relates to methods for recovering enercry that would otherwise be wasted energy in wr.,2!n power amplifi=i'antenna combLinations.

Back,crround of the Disclosure

In conventional satellite wmrnunic--doris sys:z.-ns, =s-xnde.-s formerly consisted of a number of separate powc... amplifiers each can-y- ing multiple signals. The operating point of ezch amplifier was normally set to produce an average output level substandally below the sari=ted output level ofthe amplifle.r in order to maintain linear operation.

Elowever, U.S. Patent No. 3,917,998 to Welti entitled 'Butler Matrix Transponder" deschbes an arrariaernent of N coupled power amplifiers for amplifying N signal padis. The N signal paths envisaged comprise the relaying of signals from at least one ground station to N locations on the carth using an orbiting satellite. The benefit of using coupled amplifiers over usin. a set of N non-coupled ampliflers is that the set of noncoupled amplifiers is Limited to generating a power which docs not c-xc--; d the peak power capability of a single amplifier in any signal path, wher-cas the technique which = coupled amplffi=s allows the generation of a power equal to the sum of thr- powers of 0 the amplifiers in any signal path, provided that all signal paths do not require higher than the mean pow= at the same tirne. As a result, signals that vary both above and below a mean power level are accommodated more efficiently due to a be= statistical averaging of the power demanded by the 'N signal paths. The matrix power C - C7 b ^ k4.."

amplifier of the Weld patent is for use in frequency division multiple access (FIDMA) applications, and provides the faciliry to vary the number of FDMA carrier frequencies used in each signal path and thus wrresMndingly the power needed in each signal path over a wide nange.

A matrix power amplifier =rding to the We!ti patent contains a butler matrix for cembining a riumbe,r N of input signals to be ampLified to produce N diffmit combinations of the input signals. In addition, a set of N powex amplifiers are pro-,-.ded se that each arr.udfie7 amplifies one of the combinations to produce N ampLified signals. The marrix power ampLifiers also wntain a butler mat:a for combining the amplifled signals to produce N outputs that are =plified versions of the ori,-,-i=] N Lriput signals. The benefit as compared with simply amplifying the onigi-riall N inputs in indepe.ndent ampliflen is the ability, if instantaneously needed, to devote more than the power of a single amplifier to one of the N signal paths. In principle, the matrix power ampLifier can deLiver the sum of the power output of an ampliflen to a single output.

7he characteristics of intermodulation gener-ated by non-Linearities in a matrix power amplifier are diff=nt than in a singe amplifier. It can be shown that third order in=odulation between signals input respectively to inputs I and 1 of the input butler matrix appears on the output numben (2i j)?,, and (2j - i)N of the output butler matrix. As a fl= step to reducing interrnodulation in a matrix power amplifier, one embodiment of the prewnt invention provides an excess number of amplifying paths so that outputs (2i j) or (12j - i) or their conding inputs are not used for desired signal outputs, but are t=linated in dum.my loads. Tbus, third order modulation between sign21s i and j will not be =srriitted. This requires that the number of butler matrix input and output porTs M be grcater d= the number of signals to be ampILfied N, wh =in the rernaining M - N signals arrt=inated in dummy loads.

1 W It is easy to see- that il only two signals are to be amplified, then usLflz ports 1 and 2 as inputs and outputs will rwult in dijid order inrz. -,rc,.'-ulation apoe--ung on perts 0 and 3, which we t=iinared. It is not so obvious how to achieve this when many signals are present. This problem is however solved by Babcock in another writext. B,--ocock wanted to 5.rid a method of allocating frequency ch=-les on an equaEY mac--J g- rid to 51-mJ-s arnpLifed by the same non-Linew arnplifiers such tha: third order int-c-modulation berve= any two or d= signals would not fall in a chamnel used by a signal. 'ne mathernatical formulation of he problem. is the same as for he inventive matrix. power amplifier, wherein a set of integers is found 11, 112, 13... such that L + Ik - Ij is not in the Tle scludon is called "B.-.bcock spacino'. Babc,-)r-k applied these integers to choosmig azion Nf ft-equency channels for the =srrdssion of signals. owever, the present invention applies Rabcocks integer sets to choosing among 'Nf physical output channels wILch are used for 2NT desired signals.

Ccn-,e:c:ue-,itly, the first improvement over the prior art matrix power ampUrier according to the present invention is to employ a larger matrix than the number of signals to be arnplified, and to allocate input and ourputs to si a-Is or not according to the Babcock spacing or other optimum allocation, 9-1 t> c thus insurinc, that intermodulation c.,ne.,gc-s principally from ourouts that are not allocated to si gn21s.

Furthermore, In the present cellular wmmunications market, there is an emphasis on rnaking mobile telephones small, handheld uni:s which operatefrom a=hed, rechargeable batteries. A parameter of great interest to the user of such phones is the length of time that can be spent in conversa.tion without needing to change or recharge the batt=7. This pa=etp-r is simply known as the "talk-time" offered by different tyTes of mobile phones. Of w=, it is possible to offer a Ion-er talk-tim e by using bigger batteries, but the bigge.r bantries increase the size and weight of the mobile telephone. Tlierp-fore, dwigners and inventors strive to construct 1 1 0 ^ devices which aelhieve Ion-er talk-times for a given battery capaciry. During a czrlve.-sadon, the radio transmitter power ampLi:-ler is the dominant power consume:. '-re eff-lciency of the amplifier's conversion of barx,-v energy into radio energy thus has a direct be&.n. on the length of available talk5 time for a celliflar mobe telephone.

Ce2TuLar telephone systenns were first introduced using analog frequency modulation to impress a voice on a radio signal. Analog f=ue-icy modulation has the advantage of producing a consm.,it amplitude signal whose phase angle chances. T1he most efficient transmit power amolifiers can be con=Qc.,ed for cons= amplirade signals which operate in a sa-Li=ed output mode.

0,ricy---.al ceLlul2.r systern s using analog frequency m od uLa ti o n were also duplex, m=ming that they received a signal in the reverse direction at the sam. e Erne as =srni-,ang a signal. A device knowm as a duplexer was therefore needed to couple both the tr-ansrnitter and the receiver to the =e antenna so m to avoid inL-r-,e.--,lcz. As shown in Figure 1(a), the antenna 12 is connected to the duplexe.r 16, which is formed by the filters 11 and 13. The duplexer 16 then controls the signals to and from the power amplffier 10 and the receiver 14 so as to avoid interference. Figure 1(b) illustrate-s the addition of an isolator 15 in the transmit path which is used in some cases to protect the a-msmitter apinst antenna mismatches and/or to prot=t the transmitter ft= other signals picked up on the antenna and fed back to the trarisrnitter causing an undesired phenomenon known as backintermodulation- 7rle isolator 15 diverts signals reflected frorn the anterma 2-5 rnismatches or received from other sources into a dummy load 18. The prior art does net disclose recycling of the energy diverted to the dummy load in order to increase the taik-time of handheld radios. A =srru"tte. power ampl,,,11= illustrated in Figure 1(a) can be a single power ampL!flr-.-.

A =sn3itr= power amplifier can also be constructed by combining two similax, srnaller sized ampLfiers. If the amplifier devices are driven in 1 1 0 ^ W 1 andphase and their outputs are ccrrbined wir-ti a 180 degree relarive phase so that their au,,uuts add constructively, the arnplifier is known as a push-puLl ain,,jU.ier. Sornedmes, two similar amplifiers 2.0 and 21 = be driven 90 deg.-m-s out of phase and their outputs combined using a 90 degr-- or 5 quadrarure couple-- as illustrated in Figure 2. The quadracure couple.- 23 can be fornned by running two strip =srnission Lines in para:lel prolirnity to wch other. 7fie ene,-,y is =nsi^e.-red between d,,e Lines Lil such a M=er that a sianal ilowL-2a from left to right on one Line induczs a signal flowing from ri-ht to left an the other line but with a 90 degret phase shift. Tbus, two ampLifiers connected respec'-"ve-lv to the 'e,' hand end of a first Line and the hand end of a second Line will produw signal's travelling from 'Left to righ: on the fir-st Iine and from, d-ht to left on the =nd line.

When the amplifiers are approximately driven 90 degrws cut of phase, Le net signal flowing on the first Line becomes a sum signal and the ne, siganal flowing on the =nd line is a difference sigriall, which can be arranged to be zerro. The output of the diE=r.c-- line is usually terrniriated in a dummy load 24 which nonn,-,LL]y dissipates no powe... Practical telerw,= of the matching between the amplifiers, the accuracy of the phase shift, energy at h=onic frequencies or antenna rrusmatch at the sum line output can however result in significant energy dissipation in this dummy load. The prior art does not disclose recycling this oth=-wise wasted energy in order to increase the talk-time of a handheld radio.

Yet another comguration of a power ampLifier, known as a feedforward amplifier, may be employed in some circumstances, in which linear power amplificadon rather thari saturated power amplification of cl= C amplifiers is desired. In feedfcrward power amplifier configurations, a more or less non-linear amplifier 30 produces an output signal that is then corrected by adding an c.-,or signal, which is producted by an error ampLifier 31, to the output signal using a directional coupler 32 as described above and as is shown in Figure 3. A waste energy signal nonnally produced in 0 ^ 1 9 --- - dummy load 33 corresponds to the unwanted diFie-.-.nce signal. T-he unwanted difference signal is always c)r(y-luced when two dissir,-. signals to havin. ove.-laDo' . ing spec=.% are added together. Aaain, the prior art does not disclose the recycling of the waste energy produced by the dLEe-re-,ice signal 5 in order to increase talk-time.

Su=ar.v of the Disclosure

L is an objective of the present invention to overcome the deficiencies of the prior ar: and to provide an effective rneans to amplfy multiple signals to a transmitted power level using power amplifiers ada-pted for efficiently ampLfvl-ia signals of constant ami)L,7.ide, such as Class C amplifiers, while a'dinc, hi-h levels of interm odulacion products.

vol z 0 One embodiment of the prewnt invention discloses an amplifying app=rus for linearly amplifying a desired signal using a pair of coupled non- linear ampU-i=s. n, e ampEfyir. apparatus comprises limiting means for separating amplitude variations from the desired signal and producing a constant ainplitude signal bearing the phase of the de-sired signal and an amplitude related signal. In addition, a drive signal generation means produces two drive signals each dependent on the- constant amplitude signal and the amplitude related signal such that each drive signal depends on the phase of the desired signal and such that the = of the squares of the amplitudes of the drive signals is constant. Finally, a coupling means couples the two drive signals to produce two constant amplitude signals for driving the pair of non-linew powe.. amplffiers and for coupling the outputs of the power ampUen to produce two amplified output signals, one of which is the linearly amplified desired signal and the other of which is a waste energy signal.

Another embodiment of the present invention relams to a system for communicating using a 5= station with a first plurality of second stations using a phased array antenna- The system comprises signal generating v means for generating a Erst plurahry of signals for =sriiIss':ive 0 0 ion to respec:

second stations usina radio wave modulation. Combining means then product a =nd plurahry of combined signals which are complex weighted sums of the f=t plurality of signals, wherein the second plurality carresMndincy to the number of antenna arrav elements disposed along a first dirnension of the phased array. In addition, drive signal generation means produce from each of the second plurdEry of 5'ar 1.Pals a third plurality of signals, the number of wilch corresponding to diff=nt Poups of elements disposed along a second dimension of the phased =v. AmpLifing means are provided for amplivina the s=ond pluraLity of si-c-nals with the third plurality of signals using respective =smit power amplifiers adapted to transmit constant amplirude signals. Finally, an antenna means is connezzed to the ainpEf.g me= such that wanted signals are radiated in desired directions and unwanted signals are dissipated in other directions.

In another embodiment of the present invention, the present invention relates to methods of a-ansponding a multiplicity of TDMA signals from a satellite in a satellite communications system, and Lin particular to an arrangement of the satellite power amplifiers so as to provide some flexibility for raising the power level in certain TDMA time-slots.

It is another object of the pres.ent invention to overcome the problems cited above by recovering the waste energy produced by the power ampLLer in order to increaw the amount available talk time for a handheld mobile telephone. One embodiment of the prewnt invention relates to a device for increasing the energy effidency of an arnplifier using fc--d forvard linearization. The devict- comprises a direct current power source which provides power to the amplific.. which produces a main output signal. A se-cond amplifier amplifits an =r signal. A combining network then combines the error signal with the main signal to produce a corrected sum signal and a wasted en=,7, signal. Finally, a recdfier converts the wasted 0M 1 Ln dir,_z current that is ceupled back to ihe dirt---t current energy signal ' to a power sourc-- in order to redue-e the net power consumpdon.

Detailed De:szription of the Drawinzs These and other features and advantages of the invention win be 5 rezdily apparent to one of ordinary skill in the art from the ','ollowing description, used Lin conjunction widi the drawings, in which:

Figures 1(a)-(b) illustrart. a conventional power a,-,iDiffle.- coupling to an antenna; Figure 2 illus=t--s a conventional quadrature coupled ampU.Ler; Fic-ure 3 ihustrates a conventional feed forward linearization ampLifier a=ched to an ant=.a; Fl-uxe 4 illustrates an overdirnensicned matrix power amplifier according to one embodiment of the present invention; Figure 5 illustrates a Class C linear power =plifker according to one embodiment of the present invention; Figure 6 Illustrates an N channel matrix power amplifle., using two N Class C amplifiers a=ding to one embodiment of the present invention; Fi-ure 7 illustrates a ge-neral Class C matrix power amplifier ac.-ording to one embodirnent of the present invention Figure 8 Wustrates a signal generator for prc>duc:Li. phase modulated drive signals according to one embodiment of the presera invention; Figure 9 illustrates a cylindrical array of slat ante-inn and a phase or a base station; Figure 10 Elus=tes a =smit array proczsser =ording to one embodiment of the present invention; Ficyure 11 illust-ates a block diagram of a heat reduction method acl- ordin. to one embodiment of the present invention; Figure 12 illustra= a block diagram of the recovery of reflected power according to one embodiment of the present invention; W n Figure 13 a g=ph showira the =nslrAlrxr tp 0 effriciency venus the antenna msriarch with and 4:hout one ernbodiment of the present invention; Figure 14 illus=t--s in a gruph the In talk time using one e-,ibod.,'ment of the present invention to recycle eneW; Figure 15 illus--at--s the re-covery of was:z criergy in hybrid coupled ampUrie.-s acwrding to one embodiment of the pre-sent invention; Figure 16 illus=tes a feed forwaid with a Linproved effic,.-lncv accerdina to one embodiment of the presen., invention; j CY Figure 17 illus=tes a graph of the multi-carner =smner e M., ciency according to one embodiment of the present invention; Figure 18 illuscates a wide dynamic =,cc rwdfielCircult a=ding to one ernbodiment of the present invention; and Figure 19 illu=tes a wide dynarnic =-c r,--tifier based upon quarT--7-wa-ve trandormer.

DetaUed Description of the Prefemd Embodiments

An ove-rdimensioned matrix power ainplifier according to one ernbodin, ent of the p=nt invention is illustrated in Figure 4. A set of N input signals is connected to the input ports of an + M port butler matrix 20 40 chosen according to Babcock or other optijnum spacing. To apply Babcock spacing, the butler matrix ports are a=bered with increasing integers awording to the phase increments with whc, successive input signals appew combined at the outputs. For example, port 0 ref= to the butler matiix output wnding to the sum of the input signals with no 2-5 phase shift. Port 1 refers to the output conding to the sum of input signah with an increasing phase shift in the series 0, d P, 2d D, 3d 4, etc. Port 2 con-r-sMnds to phase shifts 0, 2d cD, 4-d P, 6d cD... etc. The M outputs of the butler matrix 40 are amplified in nen-linew power arnpU= 42. Amplifier outputs are combined using a similar M + M port butl= 1 kw k a matrix 44 and the -',; B.-abcock-.Tsaced output pcrT-s yield desired amplified signalls, while the remaining M - N signals are terminated to dissz)at-unwanted inter-modulation as heat or enerly in dummy loads 46.

A.i e,..ze-ne case of the above occurs when non-linear power 5 ampliners 42 are sa=ti-.d class-C amplifiers. TIne output signal of a class-c amplifler is c,=w,-i only by its instantaneous phasse, so the general. problern can be suced as findina M phases that wU result in the I; atbitrary signals being approximated as closely as pessible with intermoduladon and distortion being dirt--tcd to 2%f-'-NT tr-,-,=ated parts. This problem is solved in the present invention by realizing that to define N arbicwy output signals, 2N degrees of freedom are reczssary, as the arblL--a.-y signals have both a re.al and an irra.&-za,-y pait. Tbus, if each input sigrul can only Y2jy in phase, it has only one degree of freedom and 22NT must be specified, in order to synthesize the N arbitrary signals.

Consequently, a 2N-ch=el matrix power ampLifier with a 2N+2N pert Butler matrix combining its outputs allows N amplified output signals to be reproduced exactly at N of the output ports of the Butler matrix, with all intermodulation being collected at the other N port-s which are terminated in dummy loads, provding that a method of deriving 2. L,T phase-varyirg or constant-envelope signals can be found such that selected N combinations produced by the Butler matrix equate to the N desired signals.

One method Is to use N configmtions of cowtant- amplitude amplifier pairs to ampLify each of the N varying-amplitude signals. Each pair of amplifien in this arrangement is driven with a pair of constant amplitude signals having a mean pha-w equal to dw of a wanted signal and a phase diffrence equal to twice the ARCOSI;E of the ratio of the instzntar.ecus desired signal amplitude to its peak amplitude. A hybrid junction can then be used to form the sum and d!E=nct of the amplifier pair's output signals, the sum having the desiicd phase and ampLitude and the 0 d,.'iere,ic.e being toated in a dummy load or subjected to a waste eriergy recovery technique which will be described further below.

Figure 5 ilustrates a linew power amplifier using cl?-ss-' accordL-ia to one embodiment of the present invention. A log=-,ihmic am.PLifier/dett-,tor 21 can if more convenient ope.-,-L- at a lower, Intermediate frequency than that to be a.,npLifed by the power amzU.ers. In this casw, an optional upconvermr 52., consisting of a local oscillator 53, a MIXe: 54 and a selective filter 55, is used to wnvert the frequency to the final frequency. Othe.-wiw, the arnplifier 21 can in priic.r)le oper-ate d=itly at the final frequency and the upconvenor 23 may be ornitted.

The =,DLifier 21 is of a known ",j-?e and produces a hald-Lirruited output s g E Ignal which preserves phase information but his amplitude variations rennoved. The input signal's amplitude variations are encoded by the logahihinic ampUe-r's progressive detection proczss into a signal proporrional to the logarithm of the amplitude which, after optional lowpass filte- in low pass filter 60, is applied to a function generator 61 The function genentor 62 converts the time-varying logamplitude signal into two time varying signals labelled COSe and SINE) to indicate that the sum of their sq,=es is always unity. COSe is equal to the ratio of instantaneous amplitude to peak amplitude, or, expressed in terms of logamplitudes, is the antilog of the diffe=ce betww.-i in=taneous logarnplitude and peak log=plitude. The peak logamplitude can either be determined by the function generator 62 over a sufficiently long period, or set from outside by means of a "SC.A=" input. Tle SINe function is merely One means of implementing such a function generator is by di LOG(A) and then usig digital signal processing circuitry comprising function look-up tables. Analog means can also be used however, employing diode or transistor networks that synthesize a piect-wise linear approximation to the deswied functions.

1 0 ^ W j In the digital implementation, the LOCAly9LITUDE signal is sL-,ipled and digitized at a rate at least equal cc the total signal bandwidtli. -ne SCALE sianal is set. to the peak of the LOG.A-'LIYGL=ME sigInal and sub=ted from it to produce a value equal to the LOG of the ratio of 5 in=taneous to peak amplitude. This value is always negative but may of 1 - course be complemented to roduct a digital value that is always positive.

p 1 This binary value betwe--n 000... MC and 111... 111 is then used as the address to a precomputed table of corTzndL-ia cos(e) and sin(e) values. 'ne values are precomputed using the formula E)=2,ARCOSIFEXP(-?,A)] whe-.- A is the address and Xis a suitable Scaling value depending on the number of bits of the address and their significance is to rnake the re- sultant e value eQud to twice the ARCOSLNE of the instantaneous to 'Deak amplitude ratio. The digital cos(e) and sin(e) values am then converted to analog voltz.,c,,e waveforms using D to A convertors and low-pass filters.

7he COSE) and 5LNG functions are used to multiply the constant amplitude signal supplied by the logarithmic amplifier 50, the low-pass filter 51 and the upconvertor 52 if used. The multiplier 58 effectively re-applies the amplitude modulation removed by the hard limiting amplifier 50 on the upper channel while applying a complementary amplitude modulation to the lower ch=iel such that the sum of the squares of the new amplitudes AI and A22 is unity. This arises through choice of the functons COSe and SINE) whose = square is automatically unity. The new signals fxm the multiplien 58 are AI.EXPGP) and A2-EX:PGc) where J2 is the phase of the original input signal, Tliese two signals are combined using a quadrature 25 coupler 56 at the inputs of the class-C power amplifier to produce (AI +jA2)WG,) and (A2 +jAI)E3TGP). Since the sum of the squ= of the real and imaginary parts of these are by design unity, the class-C power amplifiers 57 receive constant amplitude drive signals so as to a=-ate at maximum efficiency. The entire chain of components described up to the W quadrature- coupler 56 is suitable for integration on a srnall, low---ost silicon cluip with a sizz. of only 3mm x 3mm.

The two matched, class-C power amplifier srages 57 amplif their cm=t amplitude signals and then recombine ther outputs using 5 the second of the couplen 56. One output of the coupler 56 is GAI.EC'JdcP) while the other cumut of the coupler 56 is GA-I.EX2G4) wh= G is:be ampliftc-,rion crain. The flint output is the desired output sizn which repres.ents the original input signal and its phase and amplitude variations. 71he second output is a waste energy sil-pial thar is either dissipated as heat in a durnmv lead 61 or subjected to the waszz energy rt--over-v proc= as will be de-.r,-:"Ded below. It will be apDr-.iated that, by u--,la the above technique, the peak power of the output signal is limit&- to the sum of powers of the two class-C power ampiL-lers 57.

Acc-ording to another embodiment of the prewrit irivendon, N signals C are defined which are the N+171; part Butler matrix combinations of N signals to be ampUed. This can be done if desired by dig-ital signal processing using a Fa3t WaU Transfom,. Then each of the N transformed output signals is amplified by converting to a pair of wristant amplitude vectors, for e=ple, using the technique illuscmted in Figure 5, that are ampLified using 2N class-C paw= amplifiers. T1he 2N power ampliEcr outputs are then cornb in pairs to produce IN wanted signals and N waste energy signals, and the N wanted signals recombined using an output N+N port Butler rnatrix. The N wasL-- ener-gy signals may be ten-nited in dummy loads whIch dissipate unted int=modulation as heat, light, other electroma-netic radiation, or subjected to a waste ener-gy rewve.-y process as will be described below.

1 The advantage of the above is that any signal can at the output momentarily reach a level equal to the sum of all 2INT amplifier output powers if needed. This arrangement is shown in Figure 6.

I0- %V N input signals are combined in an N--N port input Butler m atrix 70 to product N output signals. Miese are- spLit into DRIVE and CONTLENfENTARY DUVE signals iin drive signal splirters 71 according to the above described procedure so izc each siggnal is a constant amplitude 5 signal- These constant amplinide si-n2Is may be amplified efficiently by the N pairs of class-C power ampLijers 72, whose outputs are then pair-wise combined in combiners 73 to produw N wanted signals and. 2N waste energy signals. The waste energy sigplals are dissipated in dummy loads 74 U heat, while the N wanted -signals = de--orrbb'ied in output Butler matrix 75 to produw the original N input signals at an ampliffiled power level. As a result, unwanted interm oduladon products = dissipated as heat, light, or ot.her electromacrnetic rudiarion in d= v loads 74.

The combination of pair-wise combiners 74 and the N+IN pert Butler matrix 75 may berecognized as a 2N-JY port wrnbining network with 2N inputs (from the 2N class-C powe., amplifiers 72) and 2N outputs (of which N are terated in d=my loads). It is in fact a partial wnstruction of a 2N---2N port Butler matrix that has beeZ simplified by removing stages that merely complete the recombination of waste energy signals unnecessarily. In general, the 2N+21N output combining matrix of such a device does not need to be a full Butler matrix but can be simplified to its Fast Walsh Transform equivalent throug deletion of fractional phase shifters or separaton into groups of smaller transforms. In general, it is desirable that at least the wanted =put signals am wnstiruted as wrnbinations of all the power amplifier signals such that maximurn ficlbdiry exists to vary the output power of each signal up to the sum of all the powers of the power amplifiers, should that be instantaneously needed.

According to another embodiment of the pr=nt invention, the abovedescribed process can be used to increase the output power in c=tiin timeslots used for transmitting paging calls to mobile phones, to increase the probability of reception.

1 k.' In a TDNLk svsiem, each communicaLor. sic-wi is allocated a portionof a time-multiplex fram e period, called a -lime slot, or, a corrimon carrielfrequency. Ln a TDNLA system, It may be desirable to have the flexibility to raise the power level in cercaill time slots. A higher power level can, for example, be needed to communica:e with a inobile ground termirial that is tempora.rily suffering a slagnal attenuation due to buildnúrs, trees, or some other object. A higher power levei can also be desirable in the cimeslors used for broadcasting paging messages to mobile,erL-"lmls in order to guarantee recepti g on while their ar,e-,.nas are in a stowed position.

71iis embodiment of the present invention comeMises a multi-beam satellite antenm for dividinR the area illuminated on U'.e nto cells. A ina= power amplifier is provided with an output cor-,-spc)ndL-lq to each antenna beam and comprised of a pluraliry of power amplifiers connected by means of Butler matricles: at their ourpurs to provide C2Ch be= signaL Means are also provided to TDNL--modulate to vary the power level of the drive signals on a timeslot by timeslot basis. Means are also provided to inbit ti=mission on groups of unused timeslots on any TDNLA signal such that available power amplifier power can during that period be taken up by increasing the drive signals in another beam. The TDINIA modulation and power varying means are preferably located on the ground. A ground station comprises the necessary sim-al generadon for each beam signal and transmits these to the satellite. Moreover, the rnarrix-PA input Butler matrix combiners can also preferably be located on the ground whereby the ground station produces already cembined drive signals for dilectly Clfiving each of the power ampLifiers and these signals are transmitted to the satellite by means.of coherent feeder links as disclosed in U.S. Patent Application No. 08/179,953, endded "A Cellular/Satellite Communications System With Improved Frequency Re-use", filed January 11, 1994, which is expressly incorporated by reference.

1 ) 09 \WO p In che = where one or rwo drneslots in w.ch beam is allocated for a pa"M,- lcalLi.n. channel of higher power than a tr-dfic channel. Lie timeslets C-2 0 are dehbe:-,,tely staggered as betwe-en the -N different outputs such that no C two outputs demand high power in the same time-slot. In this way a laUe power incre= such as a factor of ten, may be achieved in one signal path at a tirne without dra,.,,-Lic, too much of the total available power amplifier -2 power.

In the formulation of the class-C rna-ix power amplifier, drive signals for 2.N- wnstant- wnpli rude power amDUric:- stages should be wnstructed in ordz., to producc N, signals with a desired mplirude and phaw modulation at N of the output ports of a 2N+2N per-, output combining ne-twork and a further N waste criergy signals at the other N output ports which may be dissipated as heat in du=y loads or subject to a waste energy recovery pro=s which will now be disclosed with reference to Figure 7.

A Completely aeneral passive wmbing s=cnire 80 wrnbines the outputs of IN power ampUer stages 82. to producc 1IT wanted signals and N unwanted signals that am dissipated in dummy loads 81 as heat or light. In a satellite application, the latter can be achieved by = of in=deseent filament lamps as the dummy loads, and the resultant light focused back on the solar ccJT array. The passive combining network can be the known prior art Butler matrix, or a reduced Butler matrix formed by ornitting cornbirdng waste energy signals, or a simplified Butler matrix C.Onding to a Fast Walsh Transform structure as apposed to the usual M struc=.

The general combining structure can be represented as a 2N by 2N matrix having wrnplex coefficients Q. The complex coefficients Cii describe both the phase and amount of signal voltage or currrnt transfer from power amplifier number j to output port number i. Thus the N desired cutput signals may be expressed as where GAIN is the int=nded amplification thraugh the device.

IQ - 1 1 - C21 Cl 2 Cl 3 Cl 4 C (1, 2n C21 C22................

... C (2, 2.=) C (2n, 1)...................... C (2n, 2=) X r (J7, @,? =) X GA-:

These equations have N wmplex (21N real) c.onditions to be satisfied (the N real and T irnaginar; pa= of the desired si als Si), and 2N degrees gn of (the IN phases E3j). Eowever the non-linear natwe of hese equarions hampers their direct solution for the pha-,--s jE). Instead, in the prewru Lnve-ition, the equations are diEe.-,--.idaLtd to obtain a set of linear equariOns for the changes in phases Dej needed such that the desL-ed signals Si %. LE char. ge from chez- previous values 5(k- 1) to their new inte- nded values (k). Thus we obtain:

k'k)--5(k-1) = jGAIN.C.T.dT where C is sharthwd notation for the co-efficient matrix, and T is the diaggonal matrix 0 d 6 0' 0 j .......

w(jel,) 0 0 0 a 0 =P(je2) 0 0 0 0 0 0 = (je3) 0 0 0 0 7 12 and dT is a vector of the 2N phase changes to be found.

By denoting the product jGAIN.C.T by the N by 2N complex matrix U, and the diffe.-=c:c between suczessive signal samples 5(k)-5(k-1) as -d5, 15 the following equation is obtained:

U-di = 5 The N by 2N wmplex matrix U may be regarded as a 11; by 2N real matr1x if the Nx2N real coefficients URij are row-interleaved with the Nx2N imas. b nary parts UEj in the following part= URI 1 UR12 UR13 UR(1,2n) L-Ill UI12 UI13 UI(1,2n) 1 4y UR-1 1 LTR12 UR-1 3..._ = 1 UI-72 1-11213..... UR---. UI...

UR..

IJI LTI(n>) Likewise, the N complex dS values comprise IN re--d values and can be regarded as a 22N point real vector by interlazing reJ,, and imaginary p= vertically. Thus, the N complex equations U.T = dS are actually 2N real equations that are sim ly solved by a real equation solving pro=s to yield p 1 0 dT = U. dS The dT values so found an added to corresponding e values to obtain new E) values. These can be used to calculate new values of EXTGe) which are the required drive signals for the power amplifiers. This conversion of E) values to drive signals can take placce by simply a 1 PD_ ying digidzed E) values to a COSISIN look-up table or ROM to obtain values for COS(e) and S2;(e) which are the real and imaginary pa= of EXPG6). Tle numerical values of COS(e) and SIN(E3) are then DtoA converted and used to drive a quadran= modulator to product the radio- frequency drive sipals as shown in Figure 8, which will be explained in detail below. Sinw mod= technology is capable of realizing DtoA convertors of adequate precision (e.g. 8 bits) which can run at 1000 Mecasamples, per second, such a digital implementation is practically useable for all practical bandwidths.

The new E) v-alues and the C matrLx can be used to calculate the Svalues actually achieved by the linearizing approximation of diff=ntiation. Alternatively, the S-values achieved may be measured at the wanted signal outputs. Either way, it is the actually achieved values of S that are used as S(k-1) to forin the dEere,,ic--s dS with the next desired sel.' of samples 5(k). In this way, c=ors do not propagate and are Lirriixd. In the case whe:e measured 5 values we subw-acted from the next set of desired value.s SCK), the system can be described as N-ch=el Cane-sian feedback. Cartesian feedback i-s a kno,.;-r technique for reducing distortion in a quasi line= power amplifier Llimrugh usins, a signal assessment demodulaator to measure the achieved wmple--,-jues of a signal at the output of a power amplifier and compahna thern with des:L--"-' values W produce error values. The c:-.or values are LntegraL--' and fed to a quadrature modulator to produce new values of the power amplIfler drive signals that will cause the power amplifier m= new.y to deiive., the desired c.omplex, signal output. An advantageous method for the above is described in U.S. Patent AppEcation No. 08,1068,087, 'Selfadjusting Modulator", which is incorporated herein by referencz. In the case of the above inventive matrix power a.-nplffier, feeding back mea-sured complex values of the N output signals so as better- to actneve the next set of complex values is a form of feedback for co=zinc, signal matr= instead of single complex si..

gnall values.

By de=mining the new dS values in the above-described manner, the new E) values are used with the C matrix to determine the new U Matrix. A simplified way to perform, this function is to note that the effect of adding DOI to D62 will be io rotate the previous values of Ulj through an angle DE)l, which ca=s a transfe., of a fraction of their imapina'7 parts to their real pari and vice ve--,a. For smaE values of DE), this allows the U matrix to be updated with no multiplies. The DE) values can always be kept sman by choasina su=sive sample values S(k-1), S(k), S(k+l)... to be sufficiently closely spaced in time. A high sp=d digital logic machine or computer may be envissaged for performing thew calculations in real time to give continuous drive waveforms to the power amplifier stag= of several MRz bandwidth. For a sat,--1Ute application, such a =hine is preferably 0C located on the ground and me resulting drIve waveforms only communicated to the sarellire via 21N mutually coherent feeder links as disclosed n U. S. Parent Application No. 08j'179,91;3, entided "A Cellular/Sarellite Communications System With Improved Frequency Re-use", filed January 5 11, 1994, which is expressly incorporated herein by reference.

^ne a=ú:e-zent for calculating the phase sipials is illustrated in Fizure 8. A difference calculator 100 computes the difference between the last approx=ation to the desired signals at insr=r k-1 and the new signal samples Si at instant k. 7he 11; complex differ,-nc--s are input to a matrix computer 102 that multiplies them by an inverse U-Matrix cc obtain 2N delta-pbme values. The delta phase values are accumulated in 21N, phase accumularcrs 204 to produce 2N E) values that ar-- converted to COSINE and SI2N,-E values usin. COS/SIN, ROMs 106 a-nd theri DcoA converted usL'na digital to analog convertors 108. The resulting 2N I,Q signals are passed to 2.N quadrarure modulators that impress the signals on the desired radio carrier frequency to obtain die signals E=ce) indicate in Figure 7 for amplification by the constant amplitude power amplifiers.

A simplified alternative exists when the sigmals to be communicated S1 are radio sienals modulated with dizital inform, ation streams. If the information streams on each of the signal paths are symbol or bit synchronous, then'the waveform.s Si at any inant depend on a limited number of past and future bits according to the smearing produced by the impulse response of the premoduladon filtering. In the limit, an S-value in the center of a symbel at least may depend only on that symbol. With binary signalling, the symbol can only take on one of two values, 0 or 1, corresponding to an Si of + 1 or - 1 with suitable scaling to the desired output power and fequency. For a matrix power amplifier of limited size, for example 16 channels, this means that there can only be 2 = 65536 different vectors S when sampled mid-bit. The 2N values of E) corresponding to each of these S vectors can be precomputed and stored in a lob ' V 6 - '71 - reasonably sizzed ROM, and may be rerrieved when needed by applying the current 16-bits forn the 16 channels as an address. For practical purposes, adequate shapdrig of the data transidons from one bit per-lod to the next in order to control the speca= may be obtairiable by smoodily =.sizio ="a 5 between these sets of E) values by means of interpolaiLcn. Thus, if a particularevalue for a current sec of 16 bits being rx=rnitied was 130 dea-r.-es and a next value was -170 decz-rees, the E) value would be moved from the old value to the new by means of the sequence I'J5,140, 14-;,150,155,160,165,170,175,180,-175,-170 nodna that the shortest path is taken. A E) value that had less flir to move would take, the same number of smaller steps. At each step, the values of THETA would be applied via COS/SLN' ROMis and DtoA convertors to a quadrature modulator in order to convert them to the desired radio-fequencv drive signals for the power amplifiers.

The general principle of the present Livention for obtaining multichannel linear signal amplificadon using non-linear and even saturating power ampLifier stages has now be-In explained together with a riurnber of implementadons. The aim of the present invention, to reduce the n=mission of unwanted intermodulation signals produced by the power amplifier non-Linearities, is achieved diroUah choice of drive signals in a coupled power ampLifier a=gement such that unwanted intermodulation is directed to an output not used for wanted siggnal n-ansm_ission, where the unwanted energy can be dissipated as heat or light in a dummy load, or subject to a waste energy recovery procedure.

In anotbc,- embodiment of the present invention, the amplifier outputs are not combined before beina fed to a multi-bea-m ante= but rather are fed directly to the elernerits of an antenna array. Figure 9 illustrates the use of a cylindrical array of slot ancenn., as might be used for a phased array base station as disclosed in U.S. Patent Application No. 08/179,953, cildtled "A Cellular/Satellite Co=unications System With Improved Frequency Re- 1 1 0 11 z: 1 use", which is incorporated herein by referencz above. In such a phased =, --v, a number of antenna elle-nents, such as 8 horizontal cclurnns around the c,;LLnder and 20 verdcal columns, are used for transmission or reception of sizr-als from mobile radio telephones. For transmission, the elennents in each, of the ver-cical columns may be driven in phase from the output of a single power ampliflex by use of a '20-way, passive power- splitrxr. Alternadvely, each element may be equipped with its own associated smaller power amplifier (of 1/2M the output power) and the 20 amplifiers in each colum.n are driven in phase using the sanne d.-ve signal.

Each colurnn thus produces radiation fwassed in the ver-ical plane but still spread over a fairly wide azimuth. When the radiation from all eight wlumns is evuluated however, by appropriate choice of relative amplirude and phase between the eight columns, focussing in the azimuth21 plane is also produced thereby n=wing the hcdwntal radiation pattern and inc=ing the di=fivity gain- When either of the above arranaements is used for tr,-ons,-mttina multiDle sianals each in a different desired dire--tion, the signal for each column comprises the sum of the different signals with an appropriate seet of amplitude and phaw (complex) weightings for that col=. For example, the signals S1,52,S3... for the eight columns may be formed from the signals T1J2,T3 to be transmitted as follows:

51 = 0.5jT1 + (0.7+0.1j)T-2 S2 = T1 + (0.6+0.3j)T2 + 0.5j73 S3 = 0.5jT1 + (0.1+0.4j)T-2 + T3 + (0.1+0.6j)T4 S4 = (-0.1-0.2j)n + 0.5j73 + (0.9-0.1j)T4..

etc., where j = V(-1) signifyg a 90-degree phase-changed component.

It can be = that the signal Si to be ampLffied for application to a col= of elements comprises the complex-weighted sum of the independent signals T1,72 etc. Tbus, the amplific-r or amplifien for that column have to faidiLEy reproduce not just a single signal (that could have been a constant ampLitude signal as with analog PY1) but the sum of independent signals dW j., is not a constant arnplitude signal. 7rius, in prior an phasecd arr-.v base s:acions, the: uw of line= amplifiers that reproduc-ed both the an, t)Li,.ude and phas,e variations of the composite drive signi-s is r.- uired. In U.S. Parent No. 3,917,998 to Welti, the use of a coupled matrix of power amuldfiers is 5 disclosed with the prope.-,v that no sin'-.'e ainplifier neczssarily has to gene,-,Lr-- the pc-nk power that any one antenna element or fe-ed may requw'e; however the possibility of using constant-amplitude:)owc-r amplifiers was not disclosed. In a previous c.- nbodi,-,-ient of the prewriz invention, the possibility of usina clws-C or wns-.--it amplitude power ampLL:lers in a coupled =&a is dLsclosed, through overdimensioning the matrix by a factor of two at l=r relative to the number of L"ndepe-ide,-ltlv-specle ampEfied output signals desired. T-lie previous er-,.bodimei,, consisted of a number at least 2N of class-C power amplifiers coupled at their outputs by an at least per.

Butler matrix or lossless coupling network, wh=in the 2N input ports being connected to said amplifier outputs and N of said coupling network output 0 ports being said dwired amplifled signal outputs, the rest being tenminated in dummy loads. As a result, the unwanted int-ermodulation products are dissipated in the dummy loads as heat.

In the present embodiment of the prewrit invention, the IN+21N Butler matrix is riot ne-ded. Instead, each column, of elements in the array is split into even and odd elements numbered vertically from 1 at the top to at le= 2N at the bottorn. In one implementation of this embodiment, elements 1,3,5 etc. of a colu= are connected by a passive N- way power splitter driven by a first power amplifier, while elements 2,4, 6,8... are connected by a second N-way pow-- spli= driven by a =nd power ampIffier. In a =nd implementation of the pr=rit embodiment, each element is equipped with a smaller power amplifier and the amplifiers attached to even clernents are all driven wilh a Erst drive signal while ampliflm attached to odd elements art all driven with a second drive sipial.

If the drive signals are the same, radiation from the even and odd elements 1 to will reinforce in the horizontal plane, while if the drive signal's are in antiphase, there will be no radiation in the horizontal plane. Thus it is possible by va.-ying the relative phasing berween in phase and antiphase to produce varying ampdrudles of sig--.21 radiation in the horizontal plane even thouch, the indivdual ele-nenLs of the column =ive constant amulitude signals from their respec:ve power amplifiers.

For the prewrit invention, the inve.ntive cellular base station comprises at IezLst one but prefe-ribly a numb,--- of such colurnns of vertically interLw-.--d elernents dise--J w,,th cvlLidrc symmetry around '360 de- of l.3 grets azimuth on top of an ante=a tower or mast. The angular spacnc, of the columns is chosen principally to avoid rnec.cal interferencz berween them, but may be grenier. t= diis miruirnurn spacing if this oc-Ves the amy greater horizontal directiviry. As will be described be-low, each column of elements must. be fed with two sippials, to the even and odd amplifien respwdyely, such that the net or=intal radiation will equate to the desired composite signals Si defined above.

If signal Si to be radiated by column i is expressed as a time-varying amplitude A(t) and a tirrie varying phase t(t), then Let 0 D1 - AC.EMP Cj (45 +e) 1 D2 - Ac. =. Cj (0 --e)] be the time varying drive signals for the even and odd elements of column i, 20 Where COS(e) = A(t)12Ao It may be ve:fficd that, when D 1 and D2 am added to calculate the radiation in the horizontal plane, we obtain D1 +D2 Ac. [9 C(j (0 -9)) 4. EMP (i (0 __@)) 1 Ao'. e) - EZP(-jE))]-=(j4) 2.Ac cos e') 2Ac. (A (C) /2A0) EXP (i 0) A(c).EX.PCjO), as desized.

j.

- The princ-1ple of the prewnt embodiment of the present invention may be extended by dividing a column of elements into four equal groups, i.e., nos. 1,5,9... 2,6,10... 3,7,11... 4,8,12 and so on. Each gT.oui) is conneczed to a single power an, pLifier for that group by means of a power splitter, or each element is equip wi its own power amplifier and the., ped arr.r)LL'lers of a are driven in phase with a drive signal adzarted for each group. Tben, a set of four drive signals is found such that desired signal radiation occ,= at up to two desired elevation angles, which may if desired both be in the horizontal plane. In gerieral, Ite problem of how to pro,-4,acz radiation of dmired s,,nals at N desired angles of elevation from a column of at least 2N cons-=.t-ar.-plitude radiating c-lennents rr, ;-:v be W1ved by a similar mathematical prowss to that deseribed above for gener-ating N desired signals of varyma amplitude and phaw using at least 2N signals of constant amplitude and varying phase.

Tbus, it has been shown above how each column of elements in a cylirdrical =v -.ay be =,iged to product a desii-ed contribution to radiation at a spezified angle of elevation, e.g., in the horizontal plane. It remains to be explained how the different contributions from each column are chosen so as to focus the resultant radiation also to desired angles of 2 0 azimuth. This is done simply by arranging the radiation from the colu= to have the complex conjugate relationship to each other wrnpared to the signals that would be rewived from the desired azimuth. For example, if col= 1 would rective from a mobile transmitter lying at 0 de-, -= of azimuth a signal of amplitude 0.5 and phase 120 dep= while W1urnn 2 would reccive a signal of 0.7 and phase 50 degr=s, that is amplitudes in the ratio 0.5:0.7 and ph= differing by -70 de.,-.=, then column 1 and column 2 should produo: radiation contributions in the same amplitude ratio 0.5:0.7 for transmission but with a rtlative phase of +70 deg=.

The amplitude and phase relationship of the array elements for 30 rer-pcion can be predicted from the theoretical or measured polar patte=s of 1 S Che constituent elements and heii physical disposition in the amv. 7nus, the relative phases and amplitudes for =srnission in any direc-don can be predetermined by changing the signs of the phases. It is oft,--n sufficient to quantize the nurnber of possible directions the army can be called upon to 5 transmit in to a mited number of b=s spaced, for c=ple, ar 5 degree intervals around 360 dec,= of azirruth. The phase relationships and p amplitude ratios can dus be prc--omputed for each of these 72 dL'r--- zions. If the a--.uv is formed by 8 columns of elements, there exists funhennore an 8- fold symme:ry such that said 72 ssible phase Lrld amplitude relationships 1 PC reduce to only 9 distinc, panerns that are repeated by shLmng the whole partern in steps of one wlu= around the =ay. It is thus a relatively straightforwp--d process to store these 9 patterns in a signal processor and to the pan= and the shift needed to radiate a given signal in a given direction to the ne--L-esk. 5 degrees. When this has been done for all signals desired to be radiated, and the results summed to deL--1e the composim radiation to be produced by each colum.n of elements, the drive signals for the constant-a.mplitude an, pLifien associated with each column are generated using the method described above. Figure 10 illustrates a block dlag= of a =.srnit signal proczssor designed to accomplish the above-descibed process. A set of signals to be =sinined T1J2... Tal along with associated di=don informationei,w... is applied to =smit matrix processor 110. Signals T1,72 etc. are preferably in the form of a di,7'dzed =ple stream produced by other signal processor(s) (not shown) that can include speech coding, e=r correction coding and digital converSion to modulated radio signal form. Tile latter repre=ts each signal as a str= of complex numerical samples having a real and an irnaginary part or alternatively in polar form using a phase and amplitude. The matrix proczsser 110 uses the direction information el, W etc. to select, or compute a set of complex weighting coefficients using stored d= in a memory 112 which is adapted to the array configuration. In its 1 1 0J simplest form. direction information el, e2 etc. can consist of a beam number, and the be= nurriber is used to select a sec of precomputed stored coefficients from the memory 112. The coef.,'ic;le,,irs Lre used in complex multiplication with signals TI,T2 to produce weighted S]= 51, 52... etc.

The complex we;,crhiing coefficients are calculawd for each signal accordiri- to its desired U'Lrection of radiation and usir-L7 the prinociple of pliase-conj"uqidna the signals that would be received from the desired direction, as disclosed above. The weighting coet,-icl-,--is may also be calculated such as to minir.ii e the transmit power cowamed to produce a given signal suength at the receive-,s, as disclosed in the incorper-ated U.S. Patent Application No. 08/179,953, entitled "A CeLlulail'Sarellirl Communications; System With Improved Frequency R---use".

71e number of sum signals produced is equal to the number of columns of array elements, which may be greater tban, or less d= the number of sim-als to be transmitted. Of course, sl--nals to be transmitted in the same direction at the same Erne are preferably modulated on different frequencies so that the same frequency is not used in the s=e be= number more than once, except in a CDNLA context. T1he frequencies of each signal are reflected in the nature of the modulated signals Ti. Alternatively, the signals TI can be nominally represented on a zero-frequency c=.er and the correct desired relative frequency applied widiin the matrix processor 110 by including a phase slope during a process of upsampling The composite output signals now represent the sum of signals at different frequencies and thus are wide band signals wirb a comme=rately increased sample rate. Each signal Si is then converted in drive signal splitters 114 to a pair of constant amplitude signals whose = has the desired instantaneous phase and amplitude of Si. This process of drive splitting is preferably still performed in the digital si-nal domain, but shortly tbereafter it is appropriate to convert said drive signals to analog form with 30 the aid of DtoA convertors. Since the numerical form consists of complex 0, %MW numbers, one convertor may be used for the real paii and another for the imaginary part. The two signals produced are termed in the kriown an as I, Q sistrials and may be applied to a known quadra=e modulator device to tr=late em to a desired radio frequency band. Translated signals are then 5 amplified to a transmit power level, using efficient, consunt-amplitude power amplifiers 116 and 118 for even and odd column ellements res- pectlively, These amplifiers may be distributed among the ar-.av elements themselves.

It will be appreciated thar, the present invention may be applied similarly to arrays of elements disposed on a cylindrical surface, a plana surface, or any other surface. The general principle is to provide a superfluity of elernents of preferably at least a factor of two over and above the number of distinctly different signal directions the army is desired to resolve. In this way, the array when trarismitting can comprise efficient con=t-arnplitude power amplifiers and unwanted si-anal intermodulation products that anise can be a=ged to be radiated in directions other d= those in which the array radiates wanted signal ener-gy. For example, an orbiting satellite carrying such ail array can be arranged to imdiate the earth with wanted signals at different locations wEle inrermodulatien products (unwanted signals) are radiated harmlessly into space. In this application, the invention disclosed in the incorporated U.S. Patent Application No. 08/179,953, entitled 'A Cellular/Satellite Communications Sys-rem With Improved Frequency Re-use", may be employed to place the generation of the drive signals for said array elements and associated amplifiers on the ground, the resulting signals being conveyed to the sateLlite using coherent feeder links from a ground-based hub-starion. An advantage of the present invention is a reduction of the beat that would normally have to be dissipated in a less efficient, linear power amplifier designed not to produce such unwanted intermodulation signals.

Another embodiment of the present invention relates to communications systems using relay stations and in particular to systems 0.

F where the relay station is an ean-orbitng sate-lilte,avrc, a commuruicaticns transpender. 7he corrinnufflications transponder signals broadcast from a first mund scadion in a first frequency band called the forward feeder link and =nslat-.s them to a sw-wond fequency band called the downlink for rellaving to a second ground-based station, wLeh. may be a small handporuble station- The =sponde- is fi=then-,iore eq,,.Lirped wi a multiplicity of such ch,-.rnel-s each having a =sr,-.it poweramplifler and connected to a multiple beamn antenna. T-he invention employs a combining ner,vork such as a Butle.-'lfacrlx to connect the mul.-'iplidty of ?ewer ampLifiers to the multiple be---,i antenna such that each amplifier amplifies a part of every desired be= siganal instead of each ampLifier being dedicated to amplify only a single dedred beam signal.

In contrast with the prior art, an inverse Butler combining device is employed at the first ground sta::ion in order to forrn weighted sums of the desL-ed be= signals for =sponding using the power amplifiriers. This has two advantages eve-- the prior ar. method of locating the inverse Butler mat:x in the satelEre; firstly, dymamic reallocation of power bew=n ante= be=. may be accomplished without lar-ge VMa tions in the corresponding forward feed= link signals; =ndly, pre-distortion of sis sent on the forward feeder links may be used to partially compensate for distortion in associated cdez cha=el powe-. arnpEfi=-s.

Additional advantages we provided in the prefe=ed implementation dirough use of a greater number of forwaid feeder bn-b and wr..esMndinc, =spender channels and pawer amplifien tlian the number of ante= beams, thus providing a degree of redundancy against kilures. The la= also allows distortion produ= to be directed towards the unused Butler matrix output ports Lhat are terminated in dummy loads and riot connected to anter.w beams.

The ability to radiate, beat from an orbiting satellite can often be the 0 dominant factor in limiting the capacity of a satelLre communications system.

0 ^ W 0M Satellite communications systems desig-ned to provide co=urici-dori with a lar2e numbe- of mobile stations arc- of so-called multiple access n-oe and may employ Fr.qi-,e-icy Division Mulliple Accless, Time-Division Multiple Access, Code Division Multiple Access, or any combination of these techniQues. In M'v1A or CDNIA svs-,ems, a larce number of sizzals must be radiated simultaneously leading to the problem of intermodulation in =rilirters. In TDtYLA sic-nals, a ET=e period is divided into dm.- slots, and each si-azal occupies a timeslot. Ibus, in a pure TDNIA systerri, 'it '15 not riec--ssary to radiated many signals si:multaneously, and efficient constant amplitude transmitters can be used. In practice, however, lack of available frequency spe--rr=. requires that a fourth multiple access method also be employed, called Space Division '.kvfultiple Access (SDIkIA) or 7requency Re-use". Frequency Re-use is the well-known cellular radio-telephone technique of dividing the earth into cells and pe='rtin,cr c-ells that are sufficiently separated to employ the same frequencies. Thus, even when a satellite system employs pure TDNIA within each ceLl, it may have to employ SDM.A to pe =.t other cells to use the same frequency; thus the satellite ends up having to radiate several signals at the same time in different directions. The invention described above may be employed to allow a phased array of antenna elements with associated efficient, class- C power amplifiers to generate a number of TDNIA signals for radiation instantaneously in different directions. The set of directions may change from one timeslot to the next as disclosed in U.S. Parent Application No. 08/179,953, entitled "A Cellular/Satellite Com:municarions System With Improved Frequency Re-use". The overall efficiency of such an inventive arrancrbm'enr expressed in terms; of conversion of DC power from a solar 0 array or battery into usefully radiated signal energy may be no greater than if a prior art a=gement using Linear amplifiers had been employed. However, the inefEciency shows up less in terms of heat dissipation, and 30 instead as the radiation of radio energy in the form of intermodularion

PJ products harmlessly inio space. the 'Ulrer-modulation may be subject to the waste energy rwovery proce-dure herein dese-i'Ded.

J 0.

TDMA is a pre.--,r.,ed choice for such a --ansronde,- as a communications system is often less than 100% loaded. In the MNLk case, the invenrive =spender is run at maximum efficiency for active timeslots and switched off, for timeslots not presently usf-.,4.. To acwmplish this, traffic is prefer-,-blv sorted so as to occupy as far as possible the sarne active timeslots on aE beams, wl,-jch ensures that the samne tirreslocs are inactive on every beim. 71e entire c-.Lisxnder array may then be powered do.,,-n for 10 timeslots that we inactive on all beams. For timeslots that are inactive on =v but not all beams, it may be efficient to power- down certain array elements only and arrange that the others are driven to max-imum output to 0 - support the other be=s, thus reducing heat dLssipation to a minimum.

In its simplest form, this advantage of the current invention may be obrained usina the a=gernent of Figure 11. A pair of efficient, constantamplitude powe-- amplifiers 12.2 and 124 ampEy signals produced either by drive splitter 120 or separately r=ived from the ground via two coherent feeder links. A hybrid junction 126 combines the output signals from the power amplifiers to produce a sum signal and a difference signal. The drive signals may be arranged such that either the sum or difference signal is the desired signal. In Figure 11, the sum signal is desired and is wnnwted to an antenria 130 such as a horn that illuminates the desired =get, i.e., the earth. The difference signal then contains unwanted intermodulation products. In prior ail systems, the difference signal has be-en dissipated in a dummy load as heat. However, the p=nt invention provide-s maximum reduction of heat by instead radiating the waste energy into space by use of a separate directorial antenna 128 pointed away from the earth. 7he simplest form of the invention may of course be extended in an obvious marine.- to multiple channel. transponders having many b=s pointed tov., ards the earth and waste energy dissipating beams pointed away from the earth.

0 z> 0 1 - 1.2 - a Alternatively, the dE=nc-- si.anal = be reed:led usinz the inventive rectifier of Raures 18 or 19 and the waste energy recycled to the battery.

All such a =..,em=ts are- conside.----' to be A-.Lliin die wirit and scoDe of the

Claims

inven.don as described by the follo,,;,' a Claims. 10= Another embodiment of he invention =1 tees to recycling waste c-,.e-.-y,ene--Ated in a power amplfier. In Figure 12, a power =,, pU-ier 140 is coupled through an.;selar.or/c;"-calaror 1-4 1 and an optional transmit filter 142 to an ante= 14-35. In duplex systems requiring sL-nulmneous tr=s,-i=on and reception on d5erent frequencies, the transmit filter may form part of the an:erna duplex filter, the other part of the filter 14-4 belne associated with dhe In rier-,-duDle=a systems 0 - ' = such as Time Division Duplex / Tline Divisicri Multiple Access systems or simply press-to-talk systems, there may be no duplex filter but rather a transmi t/ receive ante= The pre-sent invendon is applicable to both of these corifir=tions. In the p=-it invention, the dummy load no n y g rmall connected to the reflected power per, of a circulator is replaced by a =tific-r 145. 'ne rectifier 145 conve--j AC radio hzq,,iency ene-.gy into a direct current wluch is then fed back to the battery or power source that powen the transmitter. If the efficiency of the transmitter is E1 and the efficiency of the rectifier is E22, then a fraction EI.E21 of the total c-ier,7, will be rewve--ed in the rn where the anter= is a wrnplete mismatch, reflecting all the transmitted power fed to it- The ft-action El-E-1 can, of course never be greater than unity. In the cast winere the antenna refle= a fraction R of the power fed to it, the net wnsumpdon of ener.-y will be reduc-ed by the factor I-R-E-RI.E2. For example, if R= 10,75, EI =5575, and E2 =7017G, the battery consumption is reduced by 3.85%, which is a sig.m"ficant amount of savings. When the efective energy is calculated as the ratio of the wer 0 P0 actually radiated from the antenna to the net power consumption, the efficiency curves with and without the use of the t invention are plotted in Figure 13. Figure 13 shows that the effectye efficiency is 1= C7 A 03 sensitive to antenna misrnacch, wheri Che presen. invention is used.. Figure 14 WL,st.-,it--s the Percentage of Lnc.-eued mlk-tir-,,e by using the preserit invention, as a function of the pe,-ce-iL,,ae of eriercry reflected by the ante=. Another embodiment of the present invention is illustr,-ii!--d in Figure 15. Two similar power amplifier stages 150 and 152 are combined using a MB dir=tional coupler 154 in order to obtain the sum of theiz ourput Powers at the antenna 156. An unwanted differ-.ncz signal is produc-td which, ins=d of being dissipated in a dummy load as in Figure 2, is converted- to a DC cur-,ent in rectifier 158 and the current is fed back to the bar---.v or power source feeding the amplifiers. The =c curves illustrated in Figures 13 and 14 apply to this embodiment as well. Another embodiment of the present invention is illustr,,L-.d in connection with feedforward linearization of pawer amplifiers. Fe., tdforward linearization is us-ed to improve the linearity over that achievable wi.th simple amplifiers, or in order to obtain improved efficiency for a given linearity. An improved efficiency f--dforward amplifier is illustrated in Figpire 16 wherein the main amplifier 160 can be a class C amplifier gener-ating a constant amplitude signal. When a higher amplitude is momentarily needed, an error ampLiflex 162 producces an output signal that adds to the output sinal of the class C amplifier 160 in directional coupler 164. W-hen a low amplitude is momentarily desired, the phase of the =r signal is reversed, causin. the =r signal to be subtracted from the main output 166 in the di=tional coupler 164, while increwing the level of the difference signal 168. Normally, the energy in the difference signal will be dissilpated in a dummy load and wasted. Eowever, the present invention uses a rectifier 170 to conven the difference signal to a DC =nt which U fed back to the b=r, thus recov =.g the energy. The s=' of the error arnplifier 162 in re- lation to the main amplifier 160 and the choice of the coupling ratio of the directional coupler 164 for maximurn overall efficiency must be made with the knowledge of the signal amplitude statistics.

1 1 0J Without the present invention, the ef-,ciency is Inven by:

Z 1 ( 1 -k 2) T773F A'/ IC - (1 -k 2) -31. 3ck / (k '..EZ) wh- = k is the voltace c-nwLiza factor of the dL-ectional couple.r 164, Ec is the efficiency of the amplifier 160, EL is the efficiency of the amplifier 162 at its maximum output amplitude Bpk, A is the wnstant signal amplitude produced by the amplifier 160, B is the difference bervw.,i A and the dewe,-' output signal waveform that must be ciontd"outed by ampLific.. 162, Bpk is the peak amplitude of signal wayeform B, and overlining sies m= value over dire.

When using the p=al. -Invention, the effective efficiency is given by:

( 1 -k 2) -3 A 212c + (1 -k 2).S. 3 ck / (k-EL) -R.

wh=-e R is the rectfler efficfency and D is the difference signal giverby D = k(A+ B) - Blk Another embodiment of the present invention relates to recovering waste energy in high power base station =smit= for multiple carrier cpcmdon. When a cellular base station is required to support multiple conversations with mobile stations on different frequencies, the first choice for the transnutter design is between a single, high-power multi- carrier power amplifler and a multiplicity of lowex power single carrier amplifiers. 71e multi-carriez ampLLEer must be linezr and is very inefficient because of the bdgh peak to m= ratio of the composite, multicarrier signal. In the caw where ft-dforward linearization is used, the prewnt invention can be employed to recover waste energy from the diEe=ce port of an output wuplcr.

)lr If the single c2rzne.- amplifier approach Is selected. mews must be pided to connect the plurality of amplifiers cc the antenna.

rovi antennas are not favorable because:.!v L-nd to be lar-ge. expensiveand nt-d more real estate. One way to coupie multiple amplifiers to the ar:---Lna is to 5 use a multicoupling fUtr--, which uses frequency selectilvlry, to isOL-:-- the ar.lpli.."7.ers from one another. Multicoupling fj;r--rs are large, and ex-Densive and an only fe:;Lsible when the "iue-icy dii'ie,-ence berw--n the muldole is not too small. The alrernadve is to use dissipacive combining, in which ampL:.,7.iers may be combined in pairs using hybrid couplers or di-r--tional wujle,--s, wEich are essentially sum and di.-:-e. -enw -ier,vcrk.s. Using dissipative combining for two signals, for e=ple, a power ampEfier of power 2P is used for a first signal and a second power amplifier of output power 21P for a &--ond The outputs of the two power ainplifiers are then combined using a MB coupler. Tle coupler allows half of the power of each power amplifier, i.e., P from each, power amplifier, to reach the antenna while the other half is normally dissipated in a dummy load. In general, dissipative combining of N, signals requires that each signal be amplified to a power N times the desired power P, so that the total power of all N ampUiers is NiNP, while only NTP (P from each amplifier) reaches the antenna. Where N is a power of 2, the dissipative cembiner is a binary trec which combines pairs of signals and then pairs of pairs and so an to the final output. Each pair-combinin. network may be a MB coupler.

At each coupler, half of the power is extracted in a sum sigTiall and passed to the next couplei or to the final output, while the other half is normally wasted in a dummy load. Tbus, half the total power is wasted in combining two carrien, 3/4 of the total power is wasted in combining four carriers, and in ge.-icnl a fraction (N-1)1N of the total paw= is wasted when N c=.'ers an combined. By employing the current invention, this normally wasted power is recovered at least to the extent of the rectifier efficiency R.

r p Tbus, instead of the rier cf:.-,-i-l;ency being only F-N where E is the effic, lency of one single carr.= power amplifier, the effective cf.-,^,cieicy becomes Far e=ple, when E=60% and R=70%, the effective efficiency is plotted versus '11; in Figure 17 with and wthout the present 5 nve-.,rion. The figurt Tustrates that efficiency can be improved by a fac:or of more than 1.5 by using the pre-wnt invention.

In multicarrier transr M-itters it is generally de-sired to minimize the production of unwanted frequency components caused by interrnodulation between the c=.'er frequencies. These cornponerits ar-pew at frequencies such as 2fl-r- or 22-fl if the hybrid couplers do not perfectly isclate the transmitter on fl from the transmitter on f2. However, the inclusion of a reztffier on the dLg=ence port of the coupler can, if no care. is taken, greatly contribute to the generation of intermodulation components. This phenomenon is related to the more general problem of how to product a rectifier that transfers energy from a variable voltage AC source to a fixed DC voltage. Of course, a solution that can be used is to incorporate a regulator, preferably a pulse width modulation type, i.e., switched-mode type, between the vaiiable DC output of the rectifier and the fixed voltage to which it is to be connected. However, an alterriative and novel solution is disclosed below which employs no active ci=uitry.

Fig= 18 Tu=tes a wide dynamic mge rectifying circuit with a cascade of diode rectifier stages, each stage being composed of a quadramre coupler driving a pair of identical diode rectifien, and preferably fuLl wave rectifiers. A quadranirr- wupler has the prop" that when terminated with 2.5 two identical load impedances, any ener.gy reflected from a load iinpedance mismatch is output to the so-called isolated port and not returned to the source. The AC load impedance presented by each full-wave rectifier is in fact equal to the DC load impedance, which in tum is equal to the output voltage, i.e., the fixed supply volta-ge to which the rectffied energy is to be recycled, divided by the rectified c=nt. Because the rectified c=nt h - W 1 ' vahes with the AC signa-l LmpLirud.e.;vn2,e the output voltage is consuained to a fixed value, the reezifier presenis a variable load l.,rped=ce which is not therefore matched to the source, wIE;ch causes Lineffliciency and interniodulation. However, in the ncvel configuration illustrated 'W Figure 18, the power reflected when the rec::ifier load deviates f-.om a Perfect match is transferred to the isclaed port and not rer.Lned to the sourc.e. As a result, this enerty has a =nd chance to be in the second rectifier stage, and so on. By designing each rectifier stage to be efficient at the level of residual power passed to it from a pr-.--,-iing szzge, the ne: reflection effect of the cascade of rectifiers is the product of that of the individual stages, ensuring that at all source power levels in queston, Limle enercry is reflected and that the great majoriry is rec-"fed and =. sferred,o the fLx--d voltage level. This novel rectifier circuit thus adapts a variable voltage AC source to a fixed DC voltage without the use of active regulators.

is Other cons=c:ions of such rectifiers can alm be designed. For example, another embodL-nent of the present invention is illustrated in Figure 19. A fint rectifier stage 200 is connected via a quarter wave matching U=forme.- 202 to the new end of a transmission line 204 and is matched to the source impedance. This first rectifier is adapted by choice of matching transformer 202 efficiently to rectify small signals. For larger signals, the rectifier presents a reduced impedance which, at the junction of 202 and 204, =slit into an in=twed impedance by virtue of the redprocal impedance =sfonnation pedormed by a quarter wave transformer. The-refore, at larger signal levels, the first rectifier does not load the transmission line 204 as much. A second rec:ifier stage 206 is connected via a second quarter wave matching transfornner, 208 to a point 1/4 w-avelength downstr= on the transmission line 204. Thi-s second rectifier is adapted to be below the threshold of rectification for the small signal levels for which the first reetifler is adapted. Tbus, it presents a high impedance to the seccond quarter wave matching transformer 208 which =slates to a low impedance r 1 1 MW at the juncton of 208 wd 204. One quarter wavele,,izt- nearcr the source at the junction of 20,4 and 202, this =slar-,s to a high L-npedance again, thus not loading 2.00. Consequently, source --,er--y is primarily absorbed by rectifier 200 at small signal levels and primarily by rectifier 206 at large signal levels. The principle may be ex:ended by the addition of ftu-ther r.jetifie.-s W-lth associated matching transformen pro,-,,-ssively at 1/4 wavelength ncreml-.-its dawm transrnission Line 2G4 in order to produce a rectifiler de,,ice thac convert-s energy from an AC source of any voltage level - to a fixed DC vclr---e. A pe-rsen ordinarily skilled in the a-n will also recognize that component equivalents of the disclosed =smssion line or coi:.jle- -ler,4.orb can be built, as well as waveguide circuits or c=uit.s c-nolovina circulators in order co achieve the wme lgoall, and all such corstructions which achieve adaptation of a vari=.le voltage AC source to a fixed DC voltage lc;e.' are wnsidered to fall within the scope of the relevant invention.

It be a=re- by those of ordin2--y skill in the art that the present invention can be embodied in other specific forms without department from the spirit or essential character thereof. The prewntly disclosed embodirne-lts are therefore considered in all respt--ts to be illustrative and not restrictive. The scope of the invention is indicated by the appended claims rather d= the foregoing description, and all changes which come within the meaning and range of equivalence thereof are intended to be embraced therein- 1 CLAIMS 1. A method of reducing waste energy in a system for communicating using a first station with a first plurality of second stations using a phased array antenna, comprising the steps of:

generating a first plurality of signals for transmission to said second stations using radiowave modulation; combining said first plurality of signals to produce a second plurality of signals, said combined signals being complex-weighted sums of said first plurality of signals, wherein the number of said second plurality of signals is equal to a number of antenna arrays disposed along a first dimension of said phased array antenna; generating a third plurality of signals from said second plurality of signals associated with different groups of elements disposed along a second dimension of said phased array antenna; amplifying said third plurality of signals to produce wanted and unwanted signals; and radiating said wanted signals in desired directions and said unwanted signals in other directions.

2. The method of claim 1, wherein said first station is an orbiting spacecraft.

3. The method of claim 2, wherein said antenna is a directional antenna with at least one beam pointing away from the earth.

4. The method of claim 2, wherein said radio wave modulation comprises speech coding, errorcorrection coding, error-detection coding and digital modulation.

5. The method of claim 1, wherein said antenna is a directional antenna.

6. A system for communicating using a first station with a first plurality of second stations using T 0 ^ C7 a phased array antenna comprising:

signal generation means for generating a first plurality of signals for transmission to respective second stations using radio-wave modulation; combining means for producing a second plurality of combined signals being complex-weighted sums of said first plurality of signals, the number of said second plurality of signals being equal to a number of antenna array elements disposed along a first dimension of said phased array antenna; drive signal generation means for producing from each of said second plurality of signals a third plurality of signals associated with different groups of elements disposed along a second dimension of said phased array antenna; amplifying apparatus for amplifying said third plurality of signals using respective transmit power amplifiers adapted efficiently to transmit constant amplitude signals; and said phased array antenna connected to said amplifying apparatus such that wanted signals are radiated in desired directions and unwanted signals are dissipated in other directions.

7. The system of claim 6, wherein said first station is an orbiting spacecraft.

8. The system of claim 6, wherein said antenna is a directional antenna.

9. The system of claim 6, wherein said antenna is a directional antenna with at least one beam pointing away from the earth.

10. The system of claim 6, wherein said radio wave modulation comprises speech coding, errorcorrection coding, error-detection coding and digital modulation.

11. A system for communicating using a first station with a first plurality of second stations using 1 0 ^ ww is 1 a phased array antenna comprising:

signal generation means for generating a first plurality of signals for transmission to respective second stations using radio-wave modulation; signal processing means for processing said first plurality of signals to produce a second plurality of constant amplitude drive signals; amplifying apparatus for amplifying said second plurality of drive signals using transmit power amplifiers adapted efficiently to transmit constant amplitude signals; and antenna means connected to said amplifying apparatus for radiating said power amplified signals such that wanted signals are radiated in desired directions and unwanted signals are dissipated in other directions.

12. A system according to claim 11, wherein said first station is an orbiting satellite.

13. A system according to claim 11, wherein said second stations are mobile or portable wireless telephones.

14. A system according to claim 11, wherein said radio-wave modulation is selected from the group comprising speech coding, error-correction coding, error-detection coding, and digital modulation.

15. A satellite communications system for communicating from at least one ground station via at least one orbiting satellite to a plurality of mobile stations, comprising:

signal generation means in said orbiting satellite for producing a plurality of constant amplitude signals in dependence on signals desired to be communicated to said mobile stations; power amplifier means connected to said signal generation means for amplifying said constant amplitude signals to produce desired and undesired signals; and r 01 4-2 antenna means coupled to said power amplifier means such that said desired signals are radiated in the direction of said mobile stations and the undesired signals produced by said signal generation means and power amplifier means are radiated out of said orbiting satellite in other directions.