Amplifiers – Modulator-demodulator-type amplifier
Reexamination Certificate
2002-10-21
2004-09-21
Mottola, Steven J. (Department: 2817)
Amplifiers
Modulator-demodulator-type amplifier
C330S136000
Reexamination Certificate
active
06794931
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to amplifiers and methods for amplifying signals, and in particular to amplification of a signal provided in terms of polar components.
2
. Description of the Related Art
The use of envelope restoration in a linear transmitter (or envelope elimination and restoration—EER—in the case of an amplifier), as an efficiency boosting mechanism, has been proposed. The essence of the technique is to utilise high efficiency RF amplifiers (e.g. class-C, D or E) and to apply the required envelope variations by modulating the amplifier supply. Assuming that this varying high power supply can be supplied with appropriate high efficiency, then the technique is capable of very high levels of overall DC-RF power conversion efficiency. The mechanism for supplying this high power envelope is therefore critical to the process.
FIG. 1
shows the configuration of a conventional EER amplifier. The input signal to be amplified contains both phase and amplitude modulation is of the form V
in
(t)=V(t)cos[&ohgr;
c
t+&phgr;(t)]. The input signal is split into a baseband signal S
1
(t)=V(t), representing the envelope of the input signal, and a RF signal, S
2
(t)=cos[&ohgr;
c
t+&phgr;(t)], which represents the phase (or frequency) modulation of the original input signal, and is a constant-envelope, phase-modulated signal. The RF signal S
2
may be created simply by limiting the RF input V
in
to remove the amplitude modulation, thus leaving only the phase (or frequency) modulation. The baseband signal S
1
corresponding to the amplitude variations of the input signal may be created either by diode detection of the input signal, or by coherent detection using the carrier signal after the above mentioned limiter. The latter method will produce more accurate (lower distortion) results, but will lead to increased complexity for the system overall. For this reason, diode based envelope detection has been employed in many EER transmitters to date.
The constant-envelope, phase-modulated signal S
2
is then amplified by a high efficiency RF amplifier
100
, e.g. of class-C, D or E. This will preserve the phase modulation information and transmit it to the output of the system. The baseband amplitude modulation signal S
1
is amplified by a suitably efficient audio amplifier
110
, or is used to feed a pulse-width modulator with subsequent class-D power amplification. Finally, the resulting high power audio signal is used to modulate the collector or power supply of RF amplifier
100
. This high-level modulation process restores the signal envelope to the output of amplifier
100
and, assuming that the relevant delays between the two paths conveying signals S
1
and S
2
are suitably equalised, results in a high power replica V
out
=GS
1
(t)S
2
(t) of the input signal being produced at the output.
There are several main sources of intermodulation distortion in the type of amplifier shown in FIG.
1
. These are the bandwidth of the envelope modulator (e.g. a class-S amplifier), the differential delay between the signals S
1
and S
2
, AM-PM conversion in amplifier
100
(when being high-level amplitude modulated), “encoding” error in the high-level modulator (i.e. errors between the actual and ideal response of the modulator at a given input amplitude level), and cut-off occurring in the amplifier
100
at low envelope levels.
SUMMARY OF THE INVENTION
The invention aims to reduce some or all of these forms of distortion and thereby provide more linear amplification.
According to one aspect, the invention provides apparatus for amplifying an input signal to produce an output signal, comprising means for providing the input signal in polar format comprising amplitude and phase components, amplifying means and modulating means, wherein the modulating means modulates the phase component with a pulse width modulated (PWM) signal derived from the amplitude component and the amplifying means receives the modulated phase component for amplification to produce the output signal.
According to another aspect, the invention provides a method of amplifying an input signal to produce an output signal, comprising providing the input signal in polar form comprising an amplitude component and a phase component, modulating the phase component with a pulse width modulated (PWM) signal derived from the amplitude component and amplifying the modulated phase component to produce the output signal.
Unlike the conventional EER technique of
FIG. 1
, the PWM signal of the invention does not need to be a high power signal. In
FIG. 1
, the PWM signal needs to be a high power signal because it is used to supply power to the amplifier
100
for amplifying the phase-modulated signal. As mentioned previously, the PWM signal of
FIG. 1
is amplified to a high power signal (i.e. at the full power required of the amplifier) and it must therefore be amplified by an appropriate high power switching amplifier. The bandwidth capability of this (usually class-D) amplifier is fundamentally restricted by the device capacitance of the high-power device employed. With current technology, the bandwidth limitation is generally of the order of 10 MHz, limiting the usable bandwidth of the overall system to the order of 100 kHz to 1 MHz.
Where a PWM signal is used to restore the bandwidth of an amplified signal, the intermodulation distortion (IMD) level of the overall amplification system is directly related to the “over sampling” rate of the PWM process relative to the bandwidth of the envelope signal (i.e. the amplitude component of the input signal). The oversampling rate should generally be at least 10 times, but can advantageously be increased to 50 or 100 times in order to achieve high linearity (over 60 dBc in theory). It is much easier to achieve a high sampling rate using the invention with a low power PWM signal and a switching process.
In traditional EER systems, such as that shown in
FIG. 1
, it is essential to time-align the signals conveying the envelope and phase information at the amplifier
100
, where the envelope is restored to the signal. For a 30 MHz bandwidth signal (e.g. a multi-carrier base station signal), with a desired IMD performance of −70 dBc, this time alignment must be better than about 330 ps. This can be difficult to achieve when it is considered that the envelope processing (PWM) processing, amplification, filtering etc. may have a delay measured in &mgr;s. Using the lower power envelope processing system used in the invention, this delay is considerably reduced, making the time alignment operation much easier.
In the amplification system of
FIG. 1
, significant AM-PM distortion can be introduced by amplifier
100
, when being high-level amplitude modulated. The present invention exhibits less AM-PM distortion since the PWM signal is not supplied at high level. Furthermore, the amplification system of
FIG. 1
may exhibit cut-off when the power supply to amplifier
100
is modulated to a low level. This problem is avoided by the present invention since the amplifier power supply is no longer being modulated.
Advantageously, the modulating means comprises switching means for selectively supplying the phase component to the amplifier, the switching action of the switching means being controlled by the PWM signal, thereby modulating the phase component with the PWM signal.
A linearisation scheme may be employed to counteract distortion in the system. In a preferred embodiment, the input signal can be predistorted to counter distortion in the system. Either the amplitude or phase component of the input signal or both may be predistorted. Predistortion of the phase component of the input signal assists in the further reduction of AM-PM distortion, such as may be introduced by the switching action of the switch means.
Where a linearisation scheme is employed, it is possible to use a feedback signal from the amplifier output to provide adaptive control of the linearisation scheme.
The provisio
Andrew Corporation
Mendelsohn Steve
Mottola Steven J.
LandOfFree
Switched amplifier does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Switched amplifier, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Switched amplifier will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3220318