The invention has been developed primarily for use in the field of telecommunications, and two-way mobile telecommunications in particular. However, the invention is not limited to use in this field and can be employed in other fields, such as digital broadcasting, where there is a need to amplify a signal with an amplitude modulated component.

BACKGROUND TO INVENTION Current and proposed systems for mobile communication are increasingly using modulation formats that include amplitude and phase components. In addition, it is often' necessary to vary the power level at which such signals are transmitted, particularly where it desirable to reduce or minimise power consumption such as when batteries are used.

This requirement for variation in the amplitude of the signal, both for modulation and for overall power control, means that the transmitter power amplifier is often operating at below its maximum power level. In conventional amplifier designs, this inevitably leads to a reduction in electrical efficiency compared to that obtained when operating at the amplifier's maximum power level. This in turn produces unnecessary (and often undesirable) heat dissipation and, in a battery-powered device such as a mobile phone, shortened battery life. As the transmitter power amplifier typically consumes the majority of the electrical power in a mobile phone and a substantial amount of power in a cellular base-station, these are significant disadvantages.

It is therefore desirable to provide an amplifier design that can operate with high efficiency over a suitably large range of output power levels for use in radio frequency transmitters.

SUMMARY OF INVENTION In accordance with a first aspect of the invention, there is provided an amplifying circuit for generating a radio frequency modulated output signal carrying an amplitude-modulated signal, the amplifying circuit including : a first modulator for accepting a first I/Q component signal Il, Ql and a carrier signal, and modulating the first I/Q component signal with the carrier signal to generate a first modulated signal; a first amplifier for amplifying the first modulated signal; a second modulator for accepting a second I/Q component signal 12, Q2 and the carrier, and modulating the second I/Q component signal with the carrier signal to generate a second modulated signal; a second amplifier for amplifying the second modulated signal; output combining circuitry for combining the amplified first and second modulated signals, thereby to generate the output signal; and phase changing means for selectively altering relative phases of the first and second modulated signals prior to being combined, such that the output signal is phase modulated in accordance with a data signal to be transmitted and amplitude modulated on the basis of a desired transmission power.

Preferably, Il=I2 and Q1=Q2 and the phase changing means includes a fixed phase delay between the first modulator and the first amplifier. More preferably, the fixed phase delay is 90 degrees. The fixed phase delay can alternatively be implemented after the first amplifier, or can be implemented by two smaller fixed phase changes each side of the first amplifier.

In a preferred embodiment: the first modulator includes a first mixer for accepting the carrier and Il as inputs and generating a first mixer output, and a second mixer for accepting the carrier and Ql as inputs and generating a second mixer output, the first and second mixer outputs being added together to generate the first modulated signal; and the second modulator includes a third mixer for accepting the carrier and Ql as inputs and generating a third mixer output, and a fourth mixer for accepting the carrier and Q2 as inputs and generating a fourth mixer output, the third and fourth mixer outputs being added together to generate the second modulated signal.

Preferably, the carrier signal includes a phase modulated data signal.

In a preferred form, the amplifier is configured such that a phase change at the output signal is introduced by changing the relative phases of the Il, Ql and I2, Q2 pairs in accordance with a phase modulated data signal.

It is particularly desirable that the output combining circuitry includes a reactance network, the characteristics of which are selected to improve efficiency at one or more predetermined frequencies or ranges of frequencies. In this case, it is preferable that the characteristics of the reactance network are selected to improve efficiency at a predetermined output power or range of output powers. In a preferred form, the characteristics of the reactance network can selectively be altered on the basis of instantaneous desired output power or characteristics of a data signal modulated onto the output signal.

In a second aspect, the present invention provides a method of generating a radio frequency modulated output signal carrying an amplitude-modulated signal, the method including the steps of : accepting a first I/Q component signal Il, Ql and a carrier signal, and modulating the first I/Q component signal with the carrier signal to generate a first modulated signal; amplifying the first modulated signal with a first amplifier; accepting a second I/Q component signal I2, Q2 and the carrier, and modulating the second I/Q component signal with the carrier signal to generate a second modulated signal; amplifying the second modulated signal with a second amplifier; combining the amplified first and second modulated signals, thereby to generate the output signal; and selectively altering relative phases of the first and second modulated signals prior to being combined, such that the output signal is phase modulated in accordance with a data signal to be transmitted and amplitude modulated on the basis of a desired transmission power.

At least in a preferred embodiment, in general terms the invention involves two amplifiers being driven with substantially constant amplitude variable phase signals. The outputs of the amplifiers are then summed, and the variation in phase difference between the signals creates varying amplitude in the final output signal (see Figure 1). A reactive combining network means that the amplifiers are operated at their peak efficiency both at maximum output power and at a lower power level. The power at which this lower peak occurs is detennined by the reactive components of the output network.

BRIEF DESCRIPTION OF DRAWINGS Figure 1 is a vector diagram showing the summing of individual phasors to generate a vector sum; Figure 2 is a schematic of a first embodiment of an amplifier in which a single I/Q component signal is used to generate an output signal, in accordance with the invention; Figure 3 is a schematic of a second embodiment of an amplifier in which a pair of I/Q component signals is used to generate an output signal, in accordance with the invention; Figure 4 is a graph illustrating the relationship between output power and efficiency for various reactance networks; Figure 5 is a schematic of a variable reactance network for use in the combiner shown in Figures 2 and 3; Figure 6 is a schematic of an alternative variable reactance network for use in the combiner shown in Figures 2 and 3; Figure 7 is a schematic of yet another variable reactance network for use in the combiner shown in Figures 2 and 3; and Figure 8 is a schematic of a generalised alternative embodiment of a variable reactance network implemented in accordance with the invention.

DETAILED DESCRIPTION OF PREFERRED AND OTHER EMBODIMENTS Refening to the drawings, and Figure 2 in particular, there is provided an amplifying circuit 200 for generating a single-phase radio frequency modulated output signal 201 carrying a phase-modulated data signal.

The amplifying circuit includes a first modulator 202 having a first mixer 204 and a second mixer 206. The first mixer 204 accepts as inputs a carrier signal 208 and the Q1 component, whilst the second mixer 206 accepts as inputs the same carrier signal 208 (delayed 90 degrees by delay unit 207) and the I1 component. The outputs of the first mixer 204 and the second mixer 204 are added by adder 209 to generate a first modulated output 218.

The amplifier also includes a second modulator 212 having a third mixer 214 and a fourth mixer 216. The third mixer accepts as inputs the carrier signal 208 and the Il component, whilst the fourth mixer accepts as inputs the carrier signal 208 (delayed 90 degrees by delay unit 217) and the Q1 component. The outputs of the third and fourth mixers are added by adder 211 to generate a second modulated output 210.

The first modulated output 218 is passed through phase changing means in the form of a 90 degree phase delay circuit 220, the output of which is fed to a first amplifier 222. The second modulated output 210 is fed directly to a second amplifier 224.

Each of the amplifiers 222 and 224 amplifies the respective modulated outputs, and pass the amplified modulated outputs to and output combiner 226. It will be appreciated that the amplifiers will usually be of similar, or identical design. In the preferred embodiment, the amplifiers are a matched pair to reduce distortion at the output. It is particularly desirable that the phase responses of the amplifiers are consistent with each other to avoid the need for phase compensation elsewhere.

The output combiner includes a reactive network, such as one of those shown in Figures 5 to 7, which is configured to combine the two phasors embodied in the amplified first and second modulated outputs 210 and 218. The output combiner will be described in more detail later.

In this embodiment, a data signal is introduced into the system for phase modulation onto the output signal by way of the carrier. The data signal can be modulated onto the carrier in any of the many ways that will be known to those skilled in the art.

In use, when Il is at full amplitude and Ql is at zero, the first and second modulated outputs 210 and 218 are 180 degrees out of phase with each other. The result is a zero amplitude after combining. Conversely, when Il is at zero and Q1 is at full amplitude, the first and second modulated outputs are in phase with each, and combine to give maximum output amplitude.

By altering the Il and Q values appropriately, any desired output amplitude can be generated, whilst the phase component of the signal is provided by the carrier.

An alternative embodiment is shown in Figure 3, in which like numerals are used to indicate corresponding features from the previous embodiment. However, in this case, two pairs of inputs, Il, Ql and I2, Q2, are provided. Again, each of the mixers 204,206, 214 and 216 accept the carrier signal. However, the first mixer 204 accepts the 12 component, the second mixer 206 accepts the Q2 component, the third mixer 214 accepts the Il component and the fourth mixer 216 accepts the Q1 component.

In contrast to the first embodiment, the carrier signal in the embodiment shown in Figure 3 is unmodulated. The data signal in this case is fed into the amplifier by way of the Il, Ql and I2, Q2 components. In particular, the phase relationships of the I and Q components are dynamically altered to generate the desired instantaneous amplitude and phase characteristics of the output signal. For example, to effect a phase change at the output after combining, it is necessary to cause an equal phase change between the vectors represented by Il, Ql and I2, Q2, whilst to effect an amplitude change it is necessary to cause an equal but oppositely signed phase change between the vectors represented by Il, Ql andI2, Q2.

Both of the embodiments shown in Figures 2 and 3 can be used with any of a number of combiners. Usually, the combiner will include a reactive network to reduce losses between the active stages and the overall output. This is necessary due to the interaction of the output impedances of the first and second amplifiers.

When the phase between the two modulated output signals is not either 0° or 180°, then the individual amplifiers see a reactive load impedance. The reactive components in the output combiner are designed to alter this in such a way as to create a predetermined efficiency characteristic. The change of efficiency versus power curve with changing reactive network is illustrated in Figure 4.

A signal that is mostly at or near maximum power will achieve best efficiency with the output network tuned to give the secondary peak at a relatively high power, similar to the dashed line above. A signal operating at reduced power or with a high peak to average ratio will benefit from a peak in efficiency at a lower power level such as the solid or dotted lines above.

The optimum values of the reactive components in the output combining network are therefore dependant not only on the usual current, voltage and frequency terms, but also on the characteristics of the signal being amplified and the desired output power. As one or both of these are likely to vary during operation, it is desirable that the reactance of the combiner be selectively variable in accordance with these parameters.

This could be done in a manner so as to track changes in state, such as spreading codes in a CDMA signal or number of constellation points in a QAM signal. Alternatively it could be done in a manner that follows the envelope of the data signal to be modulated. It will be appreciated by those skilled in the art that the latter approach will give a theoretically greater efficiency but is likely to be more difficult to implement.

Some specific output combiner arrangements are shown in Figures 5 to 7. It will be appreciated that only the reactance-varying portion of the combiner is shown in each case, and that the combiner may also include a fixed part 500 for buffering or providing a fixed reactance that interacts with the varying reactance part.

In Figure 5, a tuned varicap 501 (variable capacitance) diode is used in conjunction with a choke 502 and a capacitor 503 to vary the reactance of the network. The control input can be binary, a series of discrete levels, or continuously variable. This control input is co- ordinated with the data signal and the output power so as to keep the diode (s) tuned to the optimum value for the operating conditions. This co-ordination can be undertaken in any of the ways mentioned above.

Using one or more varicap diodes to vary the reactances in the combining network has the potential disadvantage of the diodes causing non-linear effects when operating at high output powers.

An alternative implementation is shown in Figure 6, in which a network of capacitors 600 are switchable by way of RF switches 601. It will be appreciated that tuning of the reactance network is then limited to a single step or a series of discrete steps, but otherwise the principle of operation is the same as for the varicap tuned implementation. It will be appreciated that parallel combinations of capacitors can be used to increase the number of steps for a given number of available capacitors, and that switched inductors can be used alternatively or in addition.

Yet another alternative is shown in Figure 7, which is similar to the embodiment of Figure 5 except that a variable inductance 700 is used rather than a variable capacitance. The disadvantage of this approach is that currently available components capable of useful swings in inductance are limited in number. However, it is hypothesised by the inventor that a ferrite-cored inductor can selectively be driven into saturation by a control current, which would then reduce the inductance seen by high-frequency signals.

There are other ways of implementing a reactive network. For example, in Figure 8 there is shown an arrangement in which variable reactance circuitry 800 and 802 are implemented for the respective outputs of the amplifiers 224 and 222. The reaction components 800 and 802 can include any combination of fixed reactance and variably reactive components, and can include one or more of the variably reactive circuits and components shown in Figures 5 to 7. It will be noted that circuitry 800 and 802 can be changed independently, which increases the flexibility of the system as a whole. For example, amplifiers having different output characteristics, or characteristics that change with age or temperature, can actively be compensated for.

Although the invention has been described with reference to a number of particular embodiments, it will be appreciated by those skilled in the art that the embodiments may take many other forms.