Amplifiers – With amplifier bypass means
Reexamination Certificate
2000-04-21
2002-03-19
Lee, Benny (Department: 2817)
Amplifiers
With amplifier bypass means
C330S149000
Reexamination Certificate
active
06359509
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to distortion or error canceling amplifiers. More particularly, the invention relates to a distortion-canceling amplifier that has low spectral regrowth and enhanced power efficiency.
Modern digital communications systems provide high spectral efficiency, clarity, and fade resistance that is unmatched by older analog systems. Nevertheless, increasing consumer demand requires even better systems. To achieve further improvements in performance, communications systems such as CDMA (code division multiple access) and GSM (global system for mobile communication) will require amplifiers that provide relatively high-power signals with limited or minimal spectral regrowth. As is known, spectral regrowth refers to the amplification of signals outside a desired frequency range. Ideally, an amplifier would amplify signals without creating noise, particularly noise outside the frequency range of the input signal. In practice, this ideal has not yet been achieved, and spectral regrowth often causes interference between adjacent communication channels. Limiting or reducing spectral regrowth is an important factor to improving spectral efficiency. When spectral regrowth is low, interference is reduced. With reduced interference, channel separation may be narrowed and the number of channels in a given bandwidth may be increased.
Simple class A or class AB amplifiers have been used in communications applications. When using a class A or class AB amplifier, spectral regrowth can be controlled by operating a simple amplifier at 8 to 10 dB below compression (the point where the amplifier clips or saturates). However, simple class A and class AB amplifiers waste about 90% of the available output power.
Another type of amplifier, a feed-forward amplifier, can be operated at higher power levels with higher efficiency. Feed-forward amplifiers (“FFAs”) use two amplifiers: a main amplifier and a distortion-canceling amplifier. The main amplifier is operated at a relatively high power level and generates an amplified, but distorted or noisy signal. A feed-forward circuit or path is used to estimate the distortion generated by the main amplifier. The estimated distortion is inverted, amplified, and then summed with the output from the main amplifier to remove the distortion in the amplified signal. An exemplary feed-forward amplifier
10
is shown in FIG.
1
.
The amplifier
10
receives an input signal
12
. The input signal
12
is delivered to a coupler
14
. The coupler
14
outputs a signal
16
to a gain and phase block
18
. The output of the gain and phase block
18
is delivered to a main amplifier
20
, which could be a class A or class AB amplifier. The main amplifier
20
amplifies the input signal
12
by a predetermined gain and outputs an amplified signal
21
to a coupler
22
. The amplified signal
21
includes a main signal component (the amplified input signal
12
) and a noise or error component. The gain and phase control block
18
and main amplifier
20
comprise a main signal path
24
.
The coupler
14
outputs a second signal
26
that is delivered to a delay
28
. The amount of time delay presented by the delay
28
is approximately equal to the time required for the signal
16
to propagate through the main signal path
24
. The delayed output signal of the delay
28
is delivered to a coupler
30
. The coupler
30
also receives an input from the coupler
22
. Each signal entering the coupler
30
is phased such that the main signal component in each input signal is canceled (or nearly canceled), leaving only an error signal
31
(the distortion created by the main amplifier
20
). The coupler
30
outputs the error signal to an error path
32
that includes a gain and phase block
34
and an error or distortion canceling amplifier
36
. The error path
32
generates a second error signal
38
. The second error signal
38
is a gain and phase adjusted version of the signal
31
. At the output, the amplitude of the second signal
38
matches or nearly matches the amplitude of the error component of the signal
21
. Also, the second signal
38
is 180° out-of-phase with the en-or component of the signal
21
.
The signal
21
is output by the coupler
22
to an error delay
40
. The error delay
40
provides a time delay approximately equal to the time it takes for the error signal
31
to propagate through the error path
32
. The delayed signal from the error delay
40
and the error signal
38
are input to a coupler
42
. The two signals are combined in the coupler
42
and, due to the phase and gain adjustments made to each, the error signal
38
substantially cancels the error component of the signal
21
, creating an output signal
44
with only the main component of the original input signal
12
.
While FFAs provide improved amplification over simple amplifiers, FFAs are still relatively inefficient. Loses in the final coupler (e.g., coupler
42
) of an FFA waste a large amount of the error signal. In addition, the distortion amplifier in an FFA requires a relatively high power output capability to prevent the error amplifier from creating a significant level of independent distortion. Accordingly, there is a need for improved methods and amplifiers for amplifying communications signals.
SUMMARY OF THE INVENTION
The present invention provides a highly efficient, linear amplifier for communications and other applications in the form of a balanced error correction (“BEC”) amplifier. (Sometimes “error correction” and “error cancellation” will be used interchangeably herein.) The BEC amplifier includes a main amplifier operable to receive a main signal and to generate an amplified signal having a main component and an error component. The BEC amplifier also includes a second amplifier coupled in a feed-forward arrangement to the main amplifier and operable to receive an input signal and to generate an output signal having a main component and an error component. A balancing network is coupled to the main amplifier and to the second amplifier. The balancing network isolates a sample of the error component of the output signal of the main amplifier, inverts the sample, and combines the sample with the input signal to the error correction amplifier. A summing point combines the output signal from the main amplifier and the output signal of the error correction amplifier such that the error components of the two output signals substantially cancel one another and the main components of the output signals are added to one another.
The invention also provides a method of providing a first signal path with a gain and a phase, providing a second signal path that is 180° out-of-phase to the first signal path and that has a gain that is about two times greater than the gain of the first signal path, providing a third signal path with a gain and a phase, and providing a fourth signal path that has a gain and phase that is substantially the same as the gain and phase of the third path. The method may also include providing a fifth signal path having a gain and a phase, and providing a sixth signal path that is 180° out-of-phase to the fifth signal path and that has a gain of about 1.5 times the gain of the fifth signal path.
Because the amplifier and method produce an output signal having a high power level and little or no noise component, spectral regrowth, which is typically analyzed based on the power level of a signal within a desired bandwidth as compared to the power level of signals outside the desired bandwidth, is reduced.
As is apparent from the above, it is an advantage of the present invention to provide an efficient amplifier with improved spectral regrowth. Other features and advantages of the present invention will become apparent by consideration of the detailed description and accompanying drawings.
REFERENCES:
patent: 3906401 (1975-09-01), Seidel
patent: 4379994 (1983-04-01), Bauman
patent: 4583049 (1986-04-01), Powell
patent: 4591800 (1986-05-01), Opas
patent: 4595882 (1986-06-01), Silagi et al.
patent: 4629
Lee Benny
Michael & Best & Friedrich LLP
Netcom, Inc.
Nguyen Patricia T.
LandOfFree
Balanced error correction amplifier and method of removing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Balanced error correction amplifier and method of removing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Balanced error correction amplifier and method of removing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2822310