GB2335321A - Modulation systems - Google Patents

Abstract

In an amplitude and phase modulator, an input signal at 2 is modulated by an IF signal at 7 and up-converted at 9 before application to an amplitude modulator 11 and a phase modulator 12. Non-linearities introduced by a power amplifier 100 are corrected by applying respective amplitude and phase correction signals derived from the input signal via shape processors 14, 16 to associated amplitude, phase modulators 11, 12. In a further embodiment (fig. 3 not shown), the amplitude and phase modulators may be combined into a single unit. The amplitude modulator 11 (fig. 2 not shown) may comprise a splitter for splitting an input signal, circuits for applying equal and opposite phase shifts to the split signals and a circuit for combining the split and oppositely phase shifted signals to produce an amplitude modulated signal. A shape processor 14 or 16 may include circuits to divide an input signal into a plurality of selected signal portions, circuits for adjusting the shape of each signal portion and a circuit to combine the shape adjusted signal portions to produce a composite signal.

Description

2335321 PATENTS ACT 1977 P12124GB-ALMILI1/mf "IMPROVEMENTS IN OR RELATING

TO TRANSMITTER SYSTEMS" This invention relates to improvements in or relating to transmitter systems and more particularly to pre-correcting transmitter drives, amplitude modulators, combined amplitude and phase modulators, correction signal generation and signal shape processors.

In broadcast systems non-linear power output amplifiers driven by transmitter drivers produce an output signal which does not have linear transfer characteristics. It is desirable, however, for the output signal which is to be broadcast to have linear transfer characteristics. To attempt to achieve this aim, it is known to combine a signal to be modulated with a correction signal, the transfer characteristics of which are the exact inverse of the transfer characteristics of the downstream non-linear power amplifier. Thus, the resultant signal, having passed through the non-linear power amplifier should have substantially linear transfer characteristics and be suitable for broadcast.

Particular problems are experienced when attempting to use correction signals to cancel out the phase shift which is generated by solid state amplifiers used in C017D1A (Coded Orthogonal Frequency Domain Multiplex).

It is an object of the present invention to seek to provide correction signals, amplitude modulators, combined amplitude and phase modulators, and shape processors such that the resultant signal has substantially linear transfer characteristics for all types of modulation techniques.

Accordingly, one aspect of the present invention provides an amplitude modulator for amplitude modulating a signal comprising: means to split the signal into two signals; means to apply a first phase shift to one of the split signals; means to apply a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; and means to combine the split and oppositely phase shifted signals together to produce an amplitude modulated signal having substantially no phase shift with respect to the original signal.

Another aspect of the present invention provides a combined amplitude and phase modulator for amplitude and phase modulating a signal comprising: means to split the signal into two signals; means to apply a first phase shift to one of the split signals; means to apply a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; means to apply a further common phase shift to both the split signals; and means to combine the split and phase shifted signals together to produce an amplitude modulated and phase modulated signal.

A further aspect of the present invention provides a transmitter driver circuit with pre-correction comprising: means to upconvert a signal to be modulated and pre-corrected; amplitude shape processor means to adjust the shape of the signal to be modulated and thereby produce an amplitude correction signal; phase shape processor means to adjust the shape of the signal to be modulated and thereby produce a phase correction signal; and modulation means to combine the signal to be pre-corrected and the amplitude and phase correction signals to produce a pre-corrected modulated signal output.

3 Another aspect of the present invention provides a shape processor for modifying the shape of a signal, the shape processor comprising: means to divide the signal into a plurality of selected signal portions having respective limits; adjustment means to adjust the shape of each signal portion; and means to combine the shape adjusted signal portions together to produce a composite output signal.

A further aspect of the present invention provides a method of amplitude modulating a signal comprising the steps of. splitting the signal into two signals; applying a first phase shift to one of the split signals; applying a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; and combining the split and oppositely phase shifted signals together to produce an amplitude modulated signal having substantially no phase shift with respect to the original signal.

Another aspect of the present invention provides a method of amplitude and phase modulating a signal comprising the steps of. splitting the signal into two signals; applying a first phase shift to one of the split signals; applying a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; applying a further common phase shift to both the split signals; and combining the split and phase shifted signals together to produce an amplitude modulated and phase modulated signal.

A further aspect of the present invention provides a method of shape processing a signal to generate an amplitude or phase correction signal comprising the steps of. dividing the signal into a plurality of selected signal portions having respective limits; adjusting the shape of each signal portion; and combining the shape adjusted signal portions together to produce a composite output signal.

4 Another aspect of the present invention provides a method of generating a transmitter driver signal with pre-correction firom a signal to be modulated and pre-corrected comprising the steps of. feeding the signal to be modulated and pre-corrected to at least two branches of a transmitter driver circuit; a first branch upconverting the signal; and the at least one other branch adjusting the shape of the signal to produce an amplitude correction signal or adjusting the shape of the signal to produce a phase correction signal; and combining the upconverted signal and the amplitude or phase correction signal to produce a pre-corrected signal output.

In order that the present invention may be more readily understood, embodiments thereof will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 shows a pre-correcting transmitter drive circuit with final carrier amplitude and phase modulation embodying the present invention.

Figure 2 shows an amplitude modulator embodying the present invention for use with the circuit of Figure 1; Figure 3 is a further pre-correcting transmitter drive with final carrier combined amplitude and phase modulation embodying the present invention.

Figure 4 is a combined amplitude and phase modulator embodying the present invention for use with the circuit of Figure 3; Figure 5 is an amplitude or phase corrector embodying the present invention; Figure 6 is an example of an amplitude correction signal showing a division of the signal into regions by a number of onset values; and Figure 7 is an example of a phase correction signal showing a division of the signal into regions by a number of onset valves.

Referring to Figure 1, a pre-correcting transmitter drive circuit 1 embodying the present invention with so-called pre-correction has an input 2 to which a signal to be modulated is applied. The signal to be modulated is fed to the input 2, though a buffer 3. The circuitry then splits into three branches 4, 5, 6.

The first branch, an upconvert branch 4, comprises the necessary components to upconvert the signal for modulation and broadcast. In the example shown in Figure 1, these components comprise an intermediate frequency (IF) oscillator source which feeds a first mixer 7 such that the output of the mixer 7 comprises the input signal modulated by the IF oscillator frequency. The output of the mixer 7 is fed to a first bandpass filter 8 and then into a second mixer 9 for mixing its input with the local oscillator frequency fed from a local oscillator (L0) source. The output of the second mixer 9 is fed through a further bandpass filter 10 to a linear wide band amplitude modulator 11 for providing amplitude modulation of the input signal. The circuitry necessary to upconvert the signal is available as a converter, preferably a Harris upconverter, incorporating IF and LO oscillators, identified by Harris drawing no. 843-5466-323 Rev.B of 24/12/96. The output of the amplitude modulator 11 is fed to a linear wide band phase modulator 12, such as that available from RF Power, for applying phase modulation to the signal.

6 The second branch comprises an amplitude correction signal branch 5 and includes a delay circuit 13 for applying a time delay to the input signal which is substantially equivalent to the time taken for the input signal to propagate through the two mixers 7,9 and the two bandpass filters 8, 10 in the upconvert branch 4. Preferably, the delay circuit 13 is of the type available and commonly used in video delay circuits. The delayed input signal is then fed to a corrector or, as termed herein a shape processor 14, for adjusting the shape of the delayed signal. The output of the amplitude shape processor 14 comprises an amplitude correction signal which is fed back to the amplitude modulator 11 in the upconvert branch 4. The transfer characteristics of the shape adjusted amplitude correction signal are chosen to be substantially the inverse of the transfer characteristics of a power output amplifier 100 located downstream of the phase modulator 12. The transfer characteristics of the amplitude correction signal are changed by altering the shape of the amplitude correction signal in a manner to be described later in the detailed description of the shape processor 14.

The third branch comprises a phase correction signal branch 6 which includes another delay circuit 15 for delaying the input signal by a time substantially equivalent to the time taken for the input signal to propagate through the two mixers 7,9, the two bandpass filters 8,10 in the upeonvert branch 4 and the amplitude modulator 11. The delayed signal is fed to another corrector, as termed herein a shape processor 16, for altering the shape of the delayed signal. The output of the phase shape processor 16 comprises a phase correction signal which is fed directly to the phase modulator 12 connected to the upconvert branch 4. The transfer characteristics of the shape adjusted phase correction signal are chosen to be substantially the inverse of the transfer characteristics of the power output amplifier 100. The phase shape processor 16 is of the same construction as the amplitude shape processor 14 and alters 7 the transfer characteristics of the phase correction signal by altering the shape of the phase correction signal.

The particular structure and operation of a preferred shape processor 14, 16 is discussed in detail later.

Preferably, the power output amplifier 100 is a Harris bipolar Class A/B lkW UHF Amplifier identified as pan no. 3913 467 09820.

The output from the phase modulator 12 comprises the pre-corrected signal output of the correction circuit 1. The pre-corrected signal is fed directly to the power output amplifier 100 for broadcast.

In operation, an input signal to be upconverted, modulated and precorrected is fed into the three separate branches and the upconvert branch 4 is used to up-convert the signal as required for modulation and subsequent broadcast. The signal fed to the amplitude correction branch 5 is shaped by the amplitude shape processor 14 to achieve the desired transfer characteristic and the signal fed to the phase correction branch 6 is shaped by the phase shape processor 15 to achieve the desired transfer characteristic. Thus, the amplitude modulator 11 receives as inputs the up-converted input signal and the amplitude correction signal. The amplitude correction signal is tailored such that its transfer characteristics are substantially the inverse of the transfer characteristics of the power output amplifier 100 so that when the amplitude correction signal from the amplitude shape processor 14 is modulated with the output of the up-converted signal, the output of the amplitude modulator 11 has a pre-corrected amplitude signal, the transfer characteristics of which are substantially the inverse of the transfer characteristics of the power output 8 amplifier 100, resulting in a linear amplitude signal output from the power output amplifier 100.

A similar process is carried out in the phase correction branch 6 such that the output of the phase shape processor 16 has transfer characteristics which are substantially the inverse of the transfer characteristics of the power output amplifier 100. Thus, the output signal from the amplitude modulator 11, when combined with the phase correction signal from the phase shape processor 16 in the phase modulator 12, produces a resultant output signal which will correct for phase non-linearities in the power output amplifier 100, resulting in a linear phase output from the power output amplifier 100.

Thus, the corrections necessary to ensure linear transfer characteristics because of any amplitude variations can be corrected using the amplitude shape processor 14 and any non-linear transfer characteristics of the resultant signal due to phase variations can be corrected by varying the output of the phase shape processor 16. Thus, the steps of amplitude and phase correction are carried out independently of one another.

Whilst the above-mentioned embodiment using pre-correction provides a workable solution for a pre-correcting transmitter drive circuit it does have a disadvantage in that the amplitude modulator 11 always introduces an amount of phase shift so that the amplitude correction signal does not produce a truly inverted transfer characteristic for the amplitude modulator since the phase variation is not taken into account. Amplitude modulators can introduce a phase shift of at least 300 and generally more. This substantial phase shift is not taken into account by the phase correction signal generated by the phase shape processor 16 and is, in some cases, impossible to correct for by manipulation of the shape of the phase correction signal owing to the 9 magnitude of the phase shift. The phase correction signal applied to the phase modulator 12 to account for variations in phase shift also introduces amplitude variations so, again, a truly inverted transfer characteristic for the phase modulator 12 is not possible. Use of the system shown in Figure 1 means that an iterative correction process involving iterative adjustment of the shape processors 14, 16 must be carried out to achieve the necessary corrected signal output having transfer characteristics which are substantially the inverse of the power amplifier 100.

It should be appreciated that no amplitude modulator 11 is presently available, especially for COMM systems, which does not introduce some level of phase shift. However, one embodiment of the present invention does provide an amplitude modulator 11 which does not introduce an appreciable phase shift and is, therefore, particularly suited for use asthe amplitude modulator 11 in the embodiment of Figure 1. Such an amplitude modulator is shown in Figure 2.

Referring to Figure 2, the amplitude modulator 11 has two inputs 17, 18 and a single output 19. The first input 17 is to be fed with the signal to be corrected, that is the up-converted signal from the upconvert branch 4 of the transmitter drive circuit 1. The input 17 is fed to a conventional splitter 20 which splits the signal waveform into two separate waveforms each having half the amplitude of the original waveform. The two outputs from the splitter 20 are each connected to a MB coupler 21, 22. The outputs from the two MB couplers 21, 22 are connected back to a common MB coupler 23 which has, as its output the corrected signal output 19. Preferably, the MB couplers are RF Power S03B70OW1 and the splitter 20 is an RF Power S03B70OW1 MB coupler configured as a signal splitter. The splitter 20 may, however, be a resistive splitter.

The other input 18 to the amplitude modulator 11 provides the amplitude correction signal from the amplitude shape processor 16 to two separate branches of the amplitude modulator 11.

The two branches of the amplitude modulator 11 fed by the input 18 are substantially identical to one another. The first branch comprises a noninverting buffer 24 connected to a summer 25 to which an offset adjustment means 26, a potentiometer and operational amplifier, is also connected. The output of the summer 25 is connected to an impedance shifter 27 in the form of a series tuned circuit comprising an inductor and a variable capacitance diode for setting the impedance which is derived from the signal at input 18. The output of the impedance shifter 27 is fed to the aforementioned 3 dB coupler 2 1. The impedance shifter 27 is forced to operate over the linear operating region of the variable capacitance diode by the bias set by the offset adjustment means 26.

In relation to the other branch of the amplitude modulator 11, this is substantially identical to the first branch except that the buffer 28 in the second branch is an inverting buffer which means that the signal applied by the impedance shifter 31 to the other MB coupler 22 is of the opposite sense to that applied to the other MB coupler 21. Like reference numerals are used to denote like components in the two branches.

The effect of these two branches of the amplitude modulator is to apply, in the respective MB couplers 21, 22, a phase shift to each of the split signals from the splitter 20. The phase shift applied by the first branch is a positive phase shift and the phase shift applied by the second branch is a negative phase shift. The phase shifts are of exactly the same amounts but are opposite to one 11 another. The outputs of the two MB couplers 21, 22 are fed into the fifflal MB coupler 23 to combine the two oppositely phased shifted"signals together so as to produce a corrected signal output which is in phase but has either a higher or a lower amplitude as required. The amount of amplitude modulation is dictated by the amount of equal and opposite phase shift applied to the separate branches of the amplitude modulator 11. This step of applying opposite phase shifts to respective halves of the split signals and then recombining the phase shifted split signals provides an almost zero phase shift at UHF and RF frequencies. Phase shift due to the amplitude modulator 11 embodying the present invention is in the order of 2'. Phase shifts of 5' or less are regarded as not appreciable for pre-correction purposes so the achieved phase shift of 20 is very low. Thus, an amplitude modulator 11 is provided which does not suffer from phase shift. This form of amplitude modulator 11 is ideal for use with the modulation circuit of Figure 1.

Another form of pre-correcting transmitter driver circuit embodying the present invention, shown in Figure 3, is very similar to that previously described in relation to Figure 1. Indeed, like reference numerals are used in Figure 3 to denote the same components as were used in Figure 1. The difference between the two arrangements is that the transmitter driver circuit of Figure 3 utilises a combined amplitude and phase modulator 32 instead of a discrete amplitude modulator 11 and a discrete phase modulator 12. Thus, the outputs of the amplitude shape processor 14 and the phase shape processor 16 are both fed to the combined amplitude and phase modulator 32. An example of such a combined amplitude and phase modulator 32 is shown in Figure 4.

The combined amplitude and phase modulator 32 of Figure 4 is very similar to the amplitude modulator 11 of Figure 2. Again, like reference numerals are used in Figure 4 to denote the same components as were used in 12 Figure 2. Effectively, the amplitude modulator of Figure 2 is modified to include a further input from the phase shape processor 16, input 33, which is itself fed to both of the branches of the amplitude modulator 32. A summer 34 connected in series with a further buffer 35 is connected between the buffer 24 and, the summer 25 present in the amplitude modulator 11 of Figure 2. The input for the summer 34 is fed by the phase correction signal from the phase shape processor 16 from input 33. Similarly, a further summer 36 and noninverting buffer 37 are connected in series between the inverting buffer 28 and the summer 251 present in the amplitude modulator 11 of Figure 2. The summer 36 is also fed with the phase correction signal from the phase shape processor 16 from input 33. Thus, the. phase correction signal from the phase shape processor 16 in Figure 3 is fed directly to the amplitude and phase modulator 32 by the summers 34, 36 and applies an intentional common phase shift to the respective split signals, the amount of the common phase shift being determined by the amplitude of the phase correction signal. Thus, whilst the MB couplers 21, 22 still apply equal phase shifts of opposite sense to the respective split signals, a further resultant phase shift is achieved by adding the common phase shift as an increased dc offset to both branches of the modulator.

Thus, depending upon the phase correction signal from the phase shape processor 16 as input to the combined amplitude and phase modulator 32 on input 33, the resultant phase shift applied by the modulator 32 can be readily altered. Unlike the separate modulator system shown in Figure 1 where an adjustment of the amplitude necessitates an adjustment of the phase shift, and once the phase shift has been adjusted, a commensurate adjustment of amplitude must be effected, the combined amplitude and phase modulator 32 shown in Figure 4 does not require such iterative adjustment of both amplitude and phase correction signals since both amplitude and phase can be adjusted 13 fully independently of one another. The amplitude is adjusted by the settings of the MB couplers 21, 22 and the resultant overall pha se shift is determined by the additional amount of dc offset applied by the phase correction signal from the phase shape processor 16. The amplitude section of the combined amplitude and phase modulator 16 benefits from the same non-appreciable phase shift as the amplitude modulator 11 of Figure 2.

Shape processors 14, 16 for producing correction signals for driving amplitude and combined amplitude and phase modulators such as those shown in Figures 2 and 4 are well known and conventionally involve filtering a top portion of a waveform by using a transistor to "bend" the waveform. The "benf' waveform is then applied to both the top and bottom of the waveform, the top and bottom of most modulated waveforms being substantially symmetrical about the time base, thereby producing a customised correction signal. However, some forms of signal modulation such as COMM do not have a symmetrical waveform and are not therefore, well suited to such conventional methods of correction signal generation.

Referring to Figure 5, there is shown a shape processor which can be used as the amplitude shape processor 14 or the phase shape processor 16 in Figures 1 and 3 to produce the necessary correction signals for feeding to the amplitude or combined amplitude and phase modulators 11, 32. The shape processor does not require the presence of a carrier signal to function properly nor does it require a substantially symmetrical waveform. The shape processor can also function through zero carrier.

The shape processor 14, 16 has a single input 38 and a single output 39. The shape processor 14, 16 is made up of a number of sections and only one such section is shown in Figure 5. Other sections of the shape processor 14, 16 14 (all of the same configuration) are connected in parallel to the illustrated sections. The reason for this will be explained in due course.

The section of the shape processor shown in Figure 5 receives the input signal from the input 38 at a pair of buffers comprising a non-inverting buffer 40 and an inverting buffer 4 1. These two buffers 40, 41 mark the beginnings of two separate branches of the illustrated section of the shape processor, which two branches are substantially identical to one another. Thus, only the first branch connected to buffer 40 will be described. Like reference numerals are used to denote the same components used in the second branch starting at buffer 4 1.

The output of the buffer 40 is connected to a precision rectifier 42. The precision rectifier 42 rectifies the input signal such that only the top half (i.e. above zero carrier) of the input signal is considered by the first branch of the illustrated section of the shape processor. The rectifier is connected at its output to a difference detector 43. The difference detector 43 has two other inputs from apair of adjustment means 44, 45 (both potentiometers) which set, respectively, a lower onset value and an upper onset value. The output of the difference detector 43 is therefore a selected portion of the output signal of the precision rectifier 42 from the lower onset value to the upper onset value. That is to say that referring to Figure 6, the output of the difference detector 43 only comprises that part of the output signal from the precision rectifier 42 which is above the lower onset value (L,) and below the upper onset value (UJ. The difference detector therefore discards all other parts of the signal above the upper onset value and below the lower onset value.

The example of Figure 6 shows a typical amplitude correction signal, amplitude voltage against time, where the signal is shown as being divided into a number of portions by the various onset values set in the various sections of the shape processor.

Only the selected portion of the input signal is passed through a further buffer 46 to a shape adjustment section of the branch. The shape adjustment section of the branch comprises a shape adjustment means 47, a potentiometer, which sets a shape adjustment factor. This shape adjustment factor comprises a first input to a two quadrant multiplier 48 (preferably configured from an Elantec EL4083C multiplier) in the shape adjustment section. The output of the buffer 46 comprises the other input to the two quadrant multiplier 48. Thus, the output of the two quadrant multiplier 48 comprises the selected portion of the input signal multiplied by the shape adjustment factor to set the new shape of that particular portion of the correction signal.

A four quadrant multiplier 49 is connected to the output of the buffer 46 and has the selected portion of the input signal as a first input and the shape adjusted signal firom the two quadrant multiplier 48 as its second input. The output of the four quadrant multiplier 49 can, therefore, provide a signal which is the square (or, as required, any specified multiplication index up to the physical limitation of the multiplier chip which is preferably an Elantec EL4083C multiplier) of the selected portion of the signal multiplied by the amplitude adjustment factor. Thus, the output of the four quadrant multiplier 49 can provide a parallel signal to the output from the two quadrant multiplier 48 and can be configured, as required, to provide selected power law signals.

A correction shape selector switch 50 has two inputs, a first input being the output of the two quadrant multiplier 48 and the second input being the output of the four quadrant multiplier 49. The correction shape selection switch 50 can be switched to select either of these two inputs, the first input 16 providing the pure selected portion of the signal multiplied by the shape adjustment factor and the second input providing the pure selected portion of the signal squared (in the present example) and multiplied by the shape adjustment factor.

The output of the correction shape selector switch 50 is connected to a finther four quadrant multiplier 51 which has as its other input an amplitude adjustment means 52 for altering the amplitude of the shape adjusted and selected portion of the signal. The amplitude adjustment means 52, a potentiometer, provides an amplitude adjustment factor which multiplies the selected portion of the signal to alter the amplitude of the selected portion of the signal. Thus, the output of the first branch of the illustrated section of the shape processor provides an amplitude adjusted and shape adjusted selected portion of the input waveform. The output of the second parallel branch of the illustrated section

incorporates the inverting buffer 41 and a precision rectifier 421. The precision rectifier 421 in the second branch, rectifies the input signal such that only the bottom half (i.e. below zero carrier) of the input signal is adjusted by the second branch of the illustrated section of the shape processor. Another difference between the two branches is that, in the second branch, the last four quadrant multiplier 511, is configured to multiply the amplitude adjusted and phase adjusted selected portion of the input waveform by a factor of - 1.

The outputs of the two branches of the illustrated section of the shape processor 14, 16 comprise a selected portion of an amplitude adjusted and shape adjusted top half of each signal waveform and a selected portion of an amplitude adjusted and shape adjusted bottom half of each signal waveform.

17 The use of the lower and upper onset adjustment means 44, 45 in each of the branches of the sections of the shape processor allows each of the branches of each section to concentrate on a respective selected part of the input signal. For example, and referring to Figure 6, a shape processor 14, 16 having three sections could divide the top half of the input signal into three distinct portions from the lower onset value to the upper onset value set by the respective onset adjustment means in each branch of each section. In the example shown in Figure 6, L 1 is the lower onset value set by the low' er onset adjustment means 44 in the first branch of the first section, UI is the upper onset value set by the upper onset adjustment means 45 in the first branch of the first section, L2 is the lower onset value set by the lower onset adjustment means in the second section (not shown) U2 is the upper onset value set by the upper onset adjustment means in the second section (not shown) and so on. The onset values in the separate branches of each section need not to be set to the same value and, in practice, depending on the waveform structure, the onset values are more likely not be set to the same values within each section.

Thus, the top half of each waveform and bottom half of each waveform are each divided up into three portions: a top portion; a middle portion and a bottom portion, each of the portions being independently adjusted for amplitude and shape. In this manner, different parts of the input signal can be adjusted independently of one another using respective amplitude adjustment factors, shape adjustment factors and power law settings so as to obtain exactly the shape of waveform required over the particular selected portion of the waveform to produce an accurate correction signal to drive the amplitude modulator or combined amplitude and phase modulator fed by the shape processor.

18 A summing junction 53 connects together the outputs of all of the branches of the various sections of the shape processor and sums all the amplitude adjusted and shape adjusted selected portions of the waveform together to complete a composite waveform for driving the ' amplitude modulator or combined amplitude and phase modulator. The output of the summing junction 53 passes through a buffer 54 before going to the output 39 of the shape processor 14, 16.

it should be appreciated that the use of the thresholding means comprising the lower onset adjustment means and upper onset adjustment means 44, 45 in the various branches and sections of the shape processor 14, 16 allows particular areas of a waveform to be focused on as shown in Figure 7. Figure 7 shows portions of a phase correction signal being divided, in the top, part of the waveform, using a fairly narrow band for the first portion adjacent the time scale between the lower onset value L 1 and the first upper onset value U1, whereas the other bands between L2, U2 and L3, U3 are somewhat larger. This allows precise manipulation of the shape of the selected portion of the signal between L1 and U1 and a more general manipulation of the shape of the other selected portions of the signal between L2, U2 and L3, U3.

The shape processor 14, 16, can be produced by using off-the-shelf and cheap components such as high frequency operational amplifiers in the precision rectifiers and regular high frequency multipliers.

The shape processor was found to be very temperature stable because of the extensive use in the circuitry thereof of operational amplifiers and diodes in feedback loops. It was found that very little overall temperature drift occurred.

19 The shape processor can manipulate signals such as video, COMM, audio and any other type of broadcast modulation standard and does not require a symmetrical type waveform. Thus, the shape processor is extremely flexible and can be used in many applications.

The shape processor functions over an extremely broad range of frequencies, particularly firom DC to 10MHZ. The maximum frequency of operation would be determined by present operational amplifier specifications, i.e. in the region of 30lvffiz. A flat frequency response was achieved over a test range of DC to 1 OMHz.

The shape processor 14,16 amplitude modulator 11, combined amplitude and phase modulator 32 and transmitter drive circuit 1 are all suited for use with digital radio and television broadcast systems.

Claims (54)

CLAIMS:

1. An amplitude modulator for amplitude modulating a signal comprising: means to split the signal into two signals; means to apply a first phase shift to one of the-split signals; means to apply a second phase shik equal to but of the opposite sense to the first phase shift, to the other of the split signals; and means to combine the split and oppositely phase shifted signals together to produce an amplitude modulated signal having substantially no phase shift with respect to the original signal.

2. An amplitude modulator according to Claim 1, wherein an input signal is provided to determine.the amount of phase shift applied to the split signals.

3. An amplitude modulator according to.Claim 2, wherein the input signal is an amplitude correction signal which determines the amount of phase shift appli ed to the split signals.

4. An amplitude modulator according to Claim 2 or 3, wherein means are provided to feed the input signal to a first branch and a second branch of the amplitude modulator, the input signal being inverted in one of the branches such that the phase shift applied to the split signal associated with that branch is equal but opposite to the phase shift applied to the other split signal associated with the other branch.

5. An amplitude modulator according to any preceding claim, wherein the means for applying the phase shifts to the respective split signals comprise respective impedance shifters.

21

6. An amplitude modulator according to Claim 5, wherein each impedance shifter comprises a series tuned circuit comprising an inductor and a variable capacitance diode forced to operate over the linear operating region of the variable capacitance diode.

7. An amplitude modulator according to any preceding claim, wherein the amplitude modulator comprises means to phase modulate the signal, the phase modulation means comprising means to apply a further common phase shift to both the split signals to produce a resultant intentional phase shift in the modulated output signal.

8. An amplitude modulator according to Claim 7, wherein a further input signal determines the amount of resultant phase shift applied to the modulated output signal.

9. An amplitude modulator according to Claim 8, wherein the further input signal comprises a phase correction signal to determine the amount of resultant phase shift applied to the modulated signal.

10. An amplitude modulator according to Claim 8 or 9, wherein the modulator has a first and a second branch and means are provided to feed the fin-ther input signal to both branches equally such that the further common phase shift is applied to both the split signals.

11. A combined amplitude and phase modulator for amplitude and phase modulating a signal comprising: means to split the signal into two signals; means to apply a first phase shift to one of the split signals; means to apply a second phase shift, equal to but of the opposite sense to the first phase shift to the other of the split signals; means to apply a further common phase shift to 22 both the split signals; and means to combine the split and phase shifted signals together to produce an amplitude modulated and phase modulated signal.

12. A modulator according to Claim 11, wherein an input signal is provided to determine the amount of phase shift applied to the split signals.

13. A modulator according to Claim 12, wherein the input signal is an amplitude correction signal which determines the amount of phase shift applied to the split signals.

14. A modulator according to Claim 12 or 13, wherein means are provided to feed the input signal to a first branch and a second branch of the amplitude modulator, the input signal being inverted in one of the branches such that the phase shift applied to the split signal associated with that branch is equal but opposite to the phase shift applied to the other split signal associated with the other branch.

15. A modulator according to any one of Claims 11 to 14, wherein the means for applying the phase shifts to the respective split signals comprise respective impedance shifters.

16. A modulator according to Claim 15, wherein each impedance shifter comprises a series tuned circuit comprising an inductor and a variable capacitance diode forced to operate over the linear operating region of the variable capacitance diode.

17. A modulator according to any one of Claims 11 to 16, wherein a further input signal determines the amount of resultant phase shift applied to the modulated signal.

23

18. A modulator according to Claim 17, wherein the further input signal comprises a phase correction signal to determine the amount of resultant phase shift applied to the modulated signal.

19. A modulator according to Claim 17 or 18, wherein the modulator has a first and a second branch and means are provided to feed the further input signal to both branches equally such that the further common phase shift is applied to both the split signals.

20. A shape processor for modifying the shape of a signal, the shape processor comprising:, means to divide the signal into a plurality of selected signal portions having respective limits; adjustment means to adjust the shape of each signal portion; and means to combine the shape adjusted signal portions together to produce a composite output signal.

21. A shape processor according to Claim 20, wherein the- adjustment means comprises an amplitude adjustment means.

22. A shape processor according to Claim 20 or 21, wherein the adjustment means comprises a shape adjustment means.

23. A shape processor according to any one of Claims 20 to 22, wherein the adjustment means comprises a power law adjustment means.

24. A shape processor according to Claim 23, wherein a shape selector is provided to select the power law adjustment means such that the output of the shape selector is subject to a power law.

24

25. A shape processor according to any one of Claims 20 to 24, wherein the shape processor has a number of sections, each section having means to divide the signal into signal portions.

26. A shape processor according to Claim 25, wherein each section is divided-into two branches, the first branch for selecting a signal portion from the signal above zero carrier and adjusting the shape of the selected portion of the signal and a second branch for selecting a signal portion from the signal below zero carrier and adjusting the shape of the selected portion of the signal.

27. A shape processor according to Claim 26, wherein the first branch has a rectifier to remove any below zero part of the signal and the second branch has an inverter and a rectifier to remove any above zero part of the input signal.

28. A shape processor according to any one of Claims 20 to 27, wherein the respective limits of each selected signal portion comprise a lower limit and an upper limit, which limits are pre-determined and set by respective lower and upper onset adjustment.

29. A shape processor according to any of Claims 20 to 28, wherein a summer is provided to add all the adjusted signal portions from the sections together to provide the composite output signal.

30. A transmitter driver circuit with pre-correction comprising: means to upconvert a signal to be modulated and pre-corrected; amplitude shape processor means to adjust the shape of the signal to be modulated and thereby produce an amplitude correction signal; phase shape processor means to adjust the shape of the signal to be modulated and thereby produce a phase correction signal; and modulation means to combine the signal to be pre-corrected and the amplitude and phase correction signals to produce a pre-corrected modulated signal output.

31. A transmitter driver circuit according to Claim 30, wherein a power output amplifier is provided having non-linear transfer characteristics, the signal output being fed to the power output amplifier and the transfer characteristics of the amplitude correction signal and phase correction signal being substantially the inverse of the non-linear transfer characteristics of the power output amplifier such that the output of the power output amplifier has substantially linear transfer characteristics.

32. A transmitter driver circuit according to Claim 30 or 3 1, wherein the modulation means is a discrete amplitude modulator and a discrete phase modulator, the amplitude modulator combining the signal to be precorrected with the amplitude correction signal and the phase modulator combining the signal to be pre-corrected with the phase correction signal.

33. A transmitter driver circuit according to Claim 30 or 3 1, wherein the modulation means is a combined amplitude and phase modulator operable to combine together the signal to be pre-corrected, the amplitude correction signal and the phase correction signal.

34. A transmitter driver circuit according to any one of Claims 30 to 33 incorporating at least one shape processor according to any one of Claims 20 to 29.

35. A transmitter driver circuit according to any one of Claims 30 to 34 incorporating an amplitude modulator according to any one of Claims 1 to 6.

26

36. A transmitter driver circuit according to any one of Claims 30 to 34 incorporating a combined amplitude and phase modulator according to any one of Claims 11 to 19.

37. A shape processor according to any one of Claims 20 to 29 in combination with an amplitude modulator according to any one of Claims 1 to 6.

38. A shape processor according to any one of Claims 20 to 29 or 37 in combination with a phase modulator.

39. A shape processor according to any one of Claims 20 to 29 for producing an amplitude correction signal and a shape processor according to any one of Claims 20 to 29 for producing a phase correction in combination with a combined amplitude and phase modulator according to any one of Claims 11 to 19.

40. A method of amplitude modulating a signal comprising the steps of. splitting the signal into two signals; applying a first phase shift to one of the split signals; applying a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; and combining the split and oppositely phase shifted signals together to produce an amplitude modulated signal having substantially no phase shift with respect to the original signal.

41. A method of amplitude and phase modulating a signal comprising the steps of. splitting the signal into two signals; applying a first phase shift to one of the split signals; applying a second phase shift, equal to but of the opposite sense to the first phase shift, to the other of the split signals; applying a further common phase shift to both the split signals; and combining the split and phase 27 shifted signals together to produce an amplitude modulated and phase modulated signal.

42. A method of shape processing a signal to generate an amplitude or phase correction signal comprising the steps of. dividing the signal into a plurality of selected signal portions having respective limits; adjusting the shape of each signal portion; and combining the shape adjusted signal portions together to produce a composite output signal.

43. A method of generating a transmitter driver signal with precorrection from a signal to be modulated and pre-corrected comprising the steps of. feeding the signal to be modulated and pre-corrected to at least two branches of a transmitter driver circuit; a first branch upconverting the signal; and the at least one other branch adjusting the shape of the signal to produce an amplitude correction signal or adjusting the shape of the signal to produce a phase correction signal; and combining the upconverted signal and the amplitude or phase correction signal to produce a pre-corrected signal output.

44. A method according to Claim 43, wherein the at least one other branch comprises two branches: ' one branch adjusting the shape of the signal to produce an amplitude correction signal and the other branch adjusting the shape of the signal to produce a phase correction signal; and the method includes the step of combining the upconverted signal with the amplitude correction signal and the phase correction signal to produce a precorrected signal output.

45. A method according to Claim 43 or 44, wherein the transmitter driver circuit is the transmitter driver circuit of any one of Claims 30 to 36.

28

46. An amplitude modulator substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

47. A combined amplitude and phase modulator substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

48. A shape processor substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

49. A transmitter driver circuit substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

50. A method of amplitude modulating a signal substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

51. A method of amplitude and phase modulating a signal substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

52. A method of shape processing a signal to generate an amplitude or phase correction signal substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

53. A method of generating a transmitter driver signal with precorrection from a signal to be modulated and pre-corrected substantially as hereinbefore described with reference to and as shown in the accompanying Figures.

54. Any novel feature or combination of features disclosed herein.