Amplifiers – Hum or noise or distortion bucking introduced into signal...
Reexamination Certificate
2002-02-26
2003-05-20
Shingleton, Michael B (Department: 2817)
Amplifiers
Hum or noise or distortion bucking introduced into signal...
C330S136000, C375S297000
Reexamination Certificate
active
06566948
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field of the Invention
The present invention relates in general to the field of signal generation, and in particular, by way of example but not limitation, to reduction of non-linearities in signal-generation circuits. A major objective of the present invention is to reduce non-linearities in signal-generation circuits.
2. Description of Related Art
Attempts to improve the linearity of a series combination of a modulator and power amplifier (PA) include: (a) choosing components that are more linear; and (b) using linearization techniques such as, for example, feedback, feedforward, and pre-distortion. Feedback generally refers to feeding back at least a part of a signal to a delayed version of the signal in order to correct the signal substantially in real time. Feedforward generally refers to correcting a signal in advance so the future signals are corrected. Because components that exhibit greater linearity are often more expensive than components that exhibit lesser linearity, linearization techniques that correct non-linearities of the less-linear components have been pursued.
Feedback linearization generally has two forms: 1) polar; and 2) Cartesian. Linearization can be applied to either a base-band signal, an IF (intermediate-frequency) signal (if one exists), or a radio-frequency (RF) signal. In general, feedback loop delays tend to limit the available signal bandwidth and can lead to instability if bandwidth restrictions are exceeded.
Feedforward linearization operates by measuring distortion from non-linearities, delaying a main signal, and subtracting the distortion from the main signal using the measured distortion. Feedforward linearization tends to be used for reducing effects of non-linearities in RF components, such as, for example, a PA. Modulator errors are not typically corrected using feedforward linearization. Feedforward linearization performance is typically limited by the quality of a second, reference amplifier, used to amplify the measured distortion. Disadvantages of using feedforward compensation include expense and the need for precise balances between reference and signal paths in order to obtain acceptable performance.
Pre-distortion linearization involves the introduction of non-linearities into a signal path. The non-linearities are then used to counteract non-linearities present in the original components, such as, for example, the modulator and the PA. An objective of pre-distortion is to keep constant the gain of the series combination of a pre-distorter (PD), the modulator, and the PA, despite instantaneous signal power-level variations due to time-varying (e.g., amplitude-modulation (AM)) components of the signal. Pre-distortion linearization can be applied anywhere along the signal path; however, in certain applications, pre-distortion linearization of the base-band or of the RF signal might be preferred.
Pre-distortion linearization typically compensates for the modulator and PA non-linearities by modeling as a parametric function a time-varying inverse of the non-linear gain of the system. Pre-distortion parameters that minimize the output distortion are also usually estimated. When the pre-distortion is perfect, the parameters of the inverse non-linear gain are identical to those minimizing the output distortion. Adaptive estimation of the pre-distortion parameters is desirable because the gain can drift over time.
The adaptive parameter estimation typically requires that the gain variations be measured accurately. Inaccurate measurements of the gain variations can bias the parameter estimation, thereby impairing the performance of pre-distortion linearization.
An inherent problem with adaptive approaches to linearization is that the system being linearized cannot distinguish distortion induced by non-linearities of, for example, the modulator or the PA within the signal path from non-linearities induced by measurement components used to measure the non-linearities within the signal path. As a result, many currently-known adaptive pre-distortion linearization approaches require ultra-linear measurement components. These ultra-linear measurements are often prohibitively expensive.
From the foregoing it can be seen that non-linearities of signal-generation circuits have required ultra-linear measurement components, which are often prohibitively expensive. What is needed is a method and system for reducing non-linearities in signal-generation circuits that improve the linearity of such circuits without requiring ultra-linear measurement components.
SUMMARY OF THE INVENTION
The present invention provides a method and system for improving the linearity of circuits such as signal-generation circuits using a measurement circuit that is not required to be ultra-linear and is therefore less expensive than other approaches. A time-varying signal with an AM component is input to a circuit in order to excite non-linearities of the circuit. The measurement circuit is balanced to remove the AM component so that the resulting signal does not excite any non-linearities of the measurement circuit. Any such non-linearities of the measurement circuit are minimized by extracting a portion of the signal that passes through the circuit. The extracted portion does not have a substantial time-varying component. Thus, non-linearities of the measurement circuit are not excited, while non-linearities of the circuit are excited and can therefore be measured and subsequently corrected.
In an embodiment of the present invention, a method of reducing non-linearities of a circuit includes providing and pre-distorting an input signal. The input signal includes a constant component and a time-varying component. The pre-distorted input signal is input to the circuit and an output signal is obtained from the circuit. A carrier signal is subtracted from the output signal, thereby yielding an error signal. An error signal envelope is determined and is demodulated using a demodulating signal related to the input signal. The parameters of the pre-distorted input signal are adjusted with reference to the demodulated converted error signal, thereby causing an amplitude of the error signal envelope to approach a constant value.
In another embodiment of the present invention, a system for reducing non-linearities includes an input-signal-generation circuit and a carrier-signal-generation circuit. The signal-generation circuit is adapted to generate an input signal having a constant component and a time-varying component. The carrier-signal-generation circuit is adapted to generate a carrier signal. The circuit includes an input connected to the input-signal-generation circuit and adapted to receive the input signal and a modulator connected to the carrier-signal-generation circuit and adapted to modulate the input signal with the carrier signal. The circuit also includes an amplifier, connected to the modulator, adapted to amplify the modulated input signal. An output is connected to the amplifier and is adapted to output the modulated amplified input signal. The system includes a pre-distorter that is adapted to adjust signal parameters. The pre-distorter is connected to the circuit. There is also a measurement circuit connected to the carrier-signal-generation circuit and to the output. The measurement circuit is adapted to yield an envelope of an error signal. The envelope comprises a function of an imbalance of the measurement circuit. Pre-distortion parameter adjustments are used to cause an amplitude of an envelope of the error signal to approach a constant value, thereby reducing non-linearities of the circuit.
The above-described and other features of the present invention are explained in detail below with reference to illustrative examples shown in the accompanying Drawings. Those of ordinary skill in the art will appreciate that the described embodiments are provided for purposes of illustration and understanding and that numerous equivalent embodiments are also contemplated in this patent application. Furthermore, the present
Agilent Technologie,s Inc.
Shingleton Michael B
LandOfFree
Method and system for reducing non-linearities does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for reducing non-linearities, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for reducing non-linearities will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3068900