Amplifiers – With control of power supply or bias voltage – With control of input electrode or gain control electrode bias
Reexamination Certificate
2002-04-19
2003-07-15
Choe, Henry (Department: 2817)
Amplifiers
With control of power supply or bias voltage
With control of input electrode or gain control electrode bias
C330S129000, C327S307000
Reexamination Certificate
active
06593812
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to amplifiers, and in particular, to efficient power amplifying circuits and methods that compensate for nonlinear distortions produced by a power amplifier between an input signal and an output signal.
2. Description of the Related Art
Amplifiers are used in a variety of applications requiring small signal amplification. Low distortion in amplifier output is of particular importance in applications requiring linear processing or reproduction of signals containing information in the amplitude and phase of a signal. Amplifier output should exhibit low distortion in amplitude and phase to maintain integrity of this information upon amplification. In other words, to maintain high signal fidelity, an amplifier should exhibit a near linear input-to-output characteristic over its operating range.
Linear amplification is of particular importance in communication devices that transmit amplified signals having information encoded in the amplitude and phase of the signal. Signals transmitted from a source to a destination are often modulated and amplified before they are transmitted. Several existing and prospective wireless digital communication systems are based on modulation schemes with both varying amplitude and phase are often referred to as linear modulation schemes. Compared to modulation schemes having only phase or frequency modulation, linear modulation schemes provide higher spectral efficiency for a given throughput (number of bits per Hz per second).
Unfortunately, in a communication device the spectral properties of a signal can only be preserved if the entire transmitter chain is linear. If the transmitter is not linear, intermodulation distortion (IMD) products will be generated in the transmitted signal causing spectral growth of the signal that interferes with users in adjacent channels. If the nonlinearity is particularly strong, signal integrity will be jeopardized and lead to increased bit-error rates in the receiver. However, in practice the output of an amplifier is nonlinear because the output eventually saturates at some value as the amplitude of the input signal is increased. As the amplifier output is driven into saturation, IMD products in the output increase resulting in undesirable distortion.
A traditional method of obtaining linear amplification is to use class A power amplifiers operating far below saturation. However, this type of operation is inefficient since a class A amplifier will dissipate power even with a zero input signal or in a quiescent condition. This is a particularly significant power drain in portable devices operating on battery power.
The power amplifier in a transmitter is the main contributor of distortion because the design is a tradeoff between linearity on one side and power efficiency on the other. Recent attention has been directed to design of power amplifier configurations having linearization circuitry applied to power efficient, but nonlinear power amplifiers to obtain both linear amplification and high power efficiency. One method of achieving increased linearity is by using envelope feedback circuits, such as those disclosed by H. Kosugi et al. in “A High-Efficiency Linear Power Amplifier Using an Envelope Feedback Method,” Electronics and Communications in Japan, Part 2, vol. 77, no. 3, 1994, pp. 50-57, and B. Shi et al. in “Linearization of RF Power Amplifiers Using Power Feedback,” Proceedings of the 49
th
IEEE Vehicular Technology Conference, May 1999, pp. 1520-1524, both of which are hereby incorporated by reference.
FIG. 1
shows a block diagram of an RF power amplifier configuration
100
illustrative of the principle of envelope feedback. As shown in
FIG. 1
, an RF input signal s
i
is weighted in a variable gain amplifier (VGA)
124
prior to a nonlinear power amplifier (PA)
128
so as to achieve an overall linear behavior. The weighting is obtained by a feedback circuit provided between an input node
120
on which is impressed an input signal s
i
and an output node
130
receiving an amplified reproduction s
o
of the input signal s
i
. Between node
122
and node
130
, the input signal s
i
is adjusted by amplification or attenuation by the VGA
124
. The adjusted signal is output to node
126
and then supplied to PA
128
to produce amplified output s
o
.
Amplifier configuration
100
includes an envelope feedback circuit to supply a signal for controlling the VGA
124
. The feedback circuit includes a first signal path along node
132
and a second signal path along node
133
. The output signal s
o
is coupled by a coupler
131
from output node
130
to node
132
. The decoupled output of s
o
is then supplied to a fixed attenuator
134
. The fixed attenuator
134
scales the RF output so and defines the linear gain of the topology (assuming a very large loopgain). The scaled output is then provided to regular or power squared envelope detector
136
.
The input signal s
i
on node
120
is similarly coupled by a coupler
121
to node
133
, which in turn is coupled to a regular envelope or squared power envelope detector (D)
135
. The detected input and scaled output (regular or power) envelope signals are subtracted in a differencing element
140
, such as a difference amplifier, to produce an error signal r
e
. The error signal is then filtered in loop filter
142
, such as a low pass filter (LPF), and amplified in an amplifier
144
. Using a summing element
146
, an optional preset offset K
c
also may be applied to the amplified error signal prior to the control of the VGA
124
. The amplified error signal is then provided to control the amplification level of the VGA
124
, which conditions the input signal s
i
prior to input into the PA
128
.
FIG. 2
shows a block diagram of a power amplifier configuration
200
that is illustrative of an alternative to using a VGA to control an output amplitude of a nonlinear power amplifier. Elements with like reference numerals and their corresponding functions are described above. As shown in
FIG. 2
, the output power of a PA
228
is directly controlled by the error signal r
e
. In this case, the error signal r
e
, for example, may control the operating point or the supply power voltage. Direct control of the PA
228
, for example, by controlling the operating point, power supply and/or another direct control method, is a way of implementing VGA functionality within the PA.
As discussed below in detail, the linearity in the combined gain of the PA and the VGA may be adversely affected depending on the choice of loop component parameters. Thus, the effect that the loopgain has on the overall gain of the amplifier configurations of
FIGS. 1 and 2
merits further investigation.
The techniques described above have the same feedback topology. What does differ between the implementations of FIG.
1
and
FIG. 2
is the relationship between the VGA/PA control signal and the output amplitude of the PA. In any case, it can be assumed that this relationship will be more or less nonlinear. Thus, without any loss of generality, it is sufficient to consider the topology in
FIG. 1
in the following analysis outlining effects that small and large signals have on the linearity of the PA in either feedback configuration.
With respect to the detectors
135
and
136
, either regular envelope detectors or power detectors have been proposed. Starting with the envelope detector and considering the DC characteristics of the loop (i.e., disregarding the loop filter), the complex baseband equivalent of an ideal envelope detector is the absolute value (a real value) of a complex valued signal. Thus,
D
(
s
x
)=|
s
x
|≡r
x
(equation 1),
and for this case the output amplitude is given by
r
o
=
A
G
(
K
c
+
A
c
r
i
)
1
+
A
G
A
c
r
i
β
·
r
i
.
(
equation
2
)
With A
G
A
c
&bgr;r
i
>>1, where A
G
A
c
&bgr;r
i
may be defined as the loopgain of the system, and for A
G
A
c
r
i
>>K
c
, the outp
Burns Doane , Swecker, Mathis LLP
Choe Henry
Telefonaktiebolaget LM Ericsson
LandOfFree
Automatic optimization of linearity for envelope feedback RF... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic optimization of linearity for envelope feedback RF..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic optimization of linearity for envelope feedback RF... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3003468