HK1224439A1 - A multi-band digital predistortion system and a method for amplifying signals - Google Patents

Abstract

A high performance and cost effective method of RF-digital hybrid mode power amplifier systems with high linearity and high efficiency for multi-frequency band wideband communication system applications is disclosed. The present disclosure enables a power amplifier system to be field reconfigurable and support multiple operating frequency bands on the same PA system over a very wide bandwidth. In addition, the present invention supports multi-modulation schemes (modulation DEG agnostic), multi-carriers and multi-channels.

Description

Digital predistortion system and method for amplifying signal

The application is a divisional application of an invention patent application with the application date of 21/12/2010, the application number of "201080064338.3" (international phase application number of PCT/IB2010/003448) and the name of a multiband broadband power amplifier digital predistortion system and method.

RELATED APPLICATIONS

This application claims the benefit of the following applications:

US patent applications filed on 21.12.2009 under the name of 61/288,838 entitled "Multi-BANDWIDEBANDPOWERAWEAMPLIFIERDIGITALPHEREDISTORTIONSYSTEMS AND filed by Wan-JongKim, Kyoung-JoonCho and ShawnPatrickStapleton.

This application is incorporated herein by reference for all purposes.

Technical Field

The present invention generally relates to wireless communication systems using complex modulation techniques. More particularly, the present invention relates to power amplifier systems for wireless communications.

Background

A wideband mobile communication system using complex modulation techniques such as Wideband Code Division Multiple Access (WCDMA) and Orthogonal Frequency Division Multiplexing (OFDM) has a large peak-to-average power ratio (PAPR) specification, and thus requires a highly linear power amplifier for its Radio Frequency (RF) transmission. Conventional Digital Predistortion (DPD) techniques have operating bandwidth limitations.

Conventional DSP-based DPD schemes use an FPGA, DSP or microprocessor to estimate, calculate and correct the non-linearity of the amplifier (PA): they track and adjust signals in the PA system quickly. However, the variation of linearity performance of the power amplifier over a wide bandwidth due to environmental variations such as temperature and asymmetric distortion of the PA output signal caused by memory effects presents challenges to conventional DSP-based DPD schemes. Conventional DPD algorithms are based on a wideband feedback signal and these algorithms require a high speed analog-to-digital converter (ADC) to capture the required information. Multi-frequency band applications (or simply multi-band applications) can have their operating frequencies significantly spaced apart. Conventional DPD architectures use ADC sampling rates that are greater than twice the nonlinear distortion bandwidth of the input signal. The sampling rate is typically greater than twice the operating bandwidth of the composite modulated signal by a factor of 5. The factor of 5 results in spectral regrowth due to nonlinear distortion produced by the power amplifier. This constraint on the sampling rate limits the feasibility of conventional predistortion architectures for single band applications. The higher the sample rate ADC, the lower the resolution, consume more power, and be more expensive.

Disclosure of Invention

Accordingly, the present invention has been made keeping in mind the above problems, and an object of the present invention is to provide a high-performance and cost-effective method of power amplifier system with high linearity and high efficiency for multi-frequency band broadband communication system applications. The present disclosure enables a power amplifier system to be field reconfigurable and support multiple operating frequency bands on the same PA system over a very wide bandwidth. In addition, the present invention supports multiple modulation schemes (agnostic), multiple carriers, and multiple channels.

To achieve the above object, according to the present invention, the technique is based on a method of adaptive digital predistortion to linearize an RF power amplifier. The invention is based on using different signals of different frequencies (multi-band signals). These multi-band signals will experience distortion through the power amplifier and produce nonlinear distortion centered on each carrier that is approximately 5 times the respective bandwidth of the multi-band signal. The feedback signal from the power amplifier is down-converted to an Intermediate Frequency (IF) that ensures that the fundamental bandwidths do not alias with each other after sampling in the ADC. The present invention can accommodate aliasing of nonlinear distortion of individual carriers.

Various embodiments of the invention are disclosed. In one embodiment, a combination of Crest Factor Reduction (CFR), DPD, power efficiency boosting techniques and coefficient adaptation algorithms are used within the PA system. In another embodiment, an Analog Quadrature Modulator (AQM) compensation structure is also used in order to enhance performance.

In one embodiment, there is provided a multi-band digital predistortion system comprising: a multi-band input signal, wherein the frequency bands are centered around spaced frequencies and the bandwidth of each frequency band is substantially smaller than the frequency spacing between the frequency bands; at least one power amplifier for providing an amplified output comprising distortion characteristics; input aliasing logic to produce an aliased image for each frequency band, wherein the aliased image for a first frequency band is in one nyquist zone and the aliased image for a second frequency band is in another nyquist zone, each of the one and other nyquist zones having a width of half a sampling rate of the input signal; a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and predistortion logic responsive to the aliased image, the predistortion logic to generate predistortion coefficients that linearize the output of the power amplifier.

In one embodiment, a method for amplifying a signal is provided, the method comprising: receiving a multi-band input signal, wherein the frequency bands are centered about spaced frequencies and the bandwidth of each frequency band is substantially less than the frequency spacing between the frequency bands; pre-distorting the multi-band input signal using pre-distortion coefficients generated by pre-distortion logic such that a first aliased image of a first frequency band of the multi-band input signal is in one nyquist zone and a second aliased image of a second frequency band of the multi-band input signal is in another nyquist zone, each of the one and the other nyquist zones having a width of half a sampling rate of the input signal; amplifying the predistorted multiband input signal to generate an amplified output, the amplified output comprising distortion characteristics; wherein the predistortion coefficients are updated using a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristics.

In one embodiment, a digital predistortion system is provided, comprising: a multi-band input signal; at least one power amplifier for providing an amplified output comprising distortion characteristics; a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and predistortion logic responsive to an aliased representation of the feedback signal, the predistortion logic to generate predistortion coefficients that linearize the amplified output of the power amplifier, wherein a sampling rate of the aliased representation of the feedback signal is less than twice a maximum bandwidth of the feedback signal.

In one embodiment, a method for amplifying a signal is provided, the method comprising: receiving a radio frequency input signal; detecting an amplified output from a power amplifier, the output being responsive to the radio frequency input signal and comprising distortion characteristics; deriving a feedback signal from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and generating, via predistortion logic responsive to an aliased representation of the feedback signal, predistortion coefficients that linearize the amplified output of the power amplifier, wherein a sampling rate of the aliased representation of the feedback signal is less than twice a maximum bandwidth of the feedback signal.

In one embodiment, a digital predistortion system is provided, comprising: a digital predistorter configured to receive a radio frequency input signal and output a predistorted output signal; a power amplifier configured to receive the pre-distorted output signal and output an amplified output; a digital predistortion estimator configured to receive an aliased representation of a feedback signal derived from the amplified output and to compute predistortion coefficients based on the aliased representation of the feedback signal.

In one embodiment, a method for amplifying a signal is provided, the method comprising: receiving a first radio frequency input signal; detecting a predistorted output signal from a digital predistorter, the predistorted output signal being responsive to the first radio frequency input signal; detecting an amplified output from the power amplifier, the output including distortion characteristics; deriving a feedback signal from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and generating, by a digital predistortion estimator, predistortion coefficients linearizing the amplified output of the power amplifier based on the feedback signal and an aliased representation of the predistorted output signal.

Some embodiments of the invention enable monitoring of fluctuations in the characteristics of the power amplifier and self-adjustment by means of adaptive algorithms. One such adaptive algorithm that is presently disclosed is referred to as the adaptive DPD algorithm, which is implemented in the digital domain and the teachings of which are given in the application incorporated herein by reference and attached as an appendix.

Applications of the present invention are suitable for use with all wireless base stations, access points, mobile equipment wireless terminals, portable wireless devices, and other wireless communication systems such as microwave and satellite communications.

Drawings

Further objects and advantages of the invention will be more fully understood from the following detailed description taken in conjunction with the accompanying drawings, in which:

fig. 1 is a block diagram showing a basic form of a digital predistortion power amplifier system.

Fig. 2 is a block diagram illustrating a simple digital predistortion block diagram for a power amplifier system according to one embodiment of the invention.

Fig. 3 is a block diagram illustrating the polynomial based predistorter of the present invention.

Fig. 4 is a block diagram of an adaptive digital predistortion direct learning algorithm applied in the digital predistortion power amplifier system of the present invention.

Fig. 5 is a block diagram of an adaptive digital predistortion indirect learning algorithm applied in the digital predistortion power amplifier system of the present invention.

Fig. 6 is a graphical representation of a frequency domain signal present in an aliased digital predistortion system.

Fig. 7 is an embodiment of a quadrature modulator compensation block architecture.

Glossary

ACLR adjacent channel leakage ratio

ACPR adjacent channel power ratio

ADC analog-to-digital converter

AQDM analog quadrature demodulator

AQM analog quadrature modulator

AQDMC analog quadrature demodulation corrector

AQMC analog quadrature modulation corrector

BPF band-pass filter

CDMA code division multiple access

CFR crest factor reduction

DAC digital-to-analog converter

DET detector

DHMPA digital mixed mode power amplifier

DDC digital down converter

DNC down converter

DPADoheerty power amplifier

DPD digital predistortion

DQDM digital quadrature demodulator

DQM digital quadrature modulator

DSP digital signal processing

DUC digital up-converter

EER envelope elimination and restoration

EF envelope following

ET envelope tracking

EVM error vector magnitude

FFLPA feed-forward linear power amplifier

FIR finite impulse response

FPGA field programmable gate array

GSM global mobile communication system

I-Q in-phase-quadrature

IF intermediate frequency

Linear amplification using non-linear components for LlNC

LO local oscillator

LPF low pass filter

MCPA multi-carrier power amplifier

MDS multidirectional search

OFDM orthogonal frequency division multiplexing

PA power amplifier

Peak to average power ratio (PAPR)

PD predistortion

PLL phase-locked loop

QAM quadrature amplitude modulation

QPSK quadrature phase shift keying

RF radio frequency

SAW surface acoustic wave filter

UMTS universal mobile communication system

UPC up converter

WCDMA wideband code division multiple access

WLAN wireless local area network

Detailed Description

The present invention is a new multi-band predistortion system using an adaptive digital predistortion algorithm. The invention is a hybrid system of digital and analog modules. The interaction of the digital and analog modules of the hybrid system linearizes the spectral regeneration and enhances the power efficiency of the PA while maintaining or increasing the wide bandwidth. Thus, the present invention achieves higher efficiency and higher linearity of wideband complex modulated carriers operating simultaneously over multiple different frequency bands.

Fig. 1 is a high-level block diagram illustrating a basic system architecture, which may be considered to include, at least for some embodiments, digital and analog modules, and a feedback path. The digital module is a digital predistortion controller 101 comprising a DPD algorithm, other auxiliary DSP algorithms and associated digital circuitry. As discussed additionally in the applications incorporated by reference, the analog modules are the main power amplifier 102, other auxiliary analog circuits such as DPAs, and related peripheral analog circuits of the overall system. The present invention is a "black box", plug and play type system because it accepts as its input an RF modulated signal 100 and provides as its output a substantially identical but amplified RF signal 103, which is thus RF in/RF out. According to one embodiment of the present invention, the baseband modulation signal may be directly applied to the digital predistorter controller. According to one embodiment of the present invention, the baseband input signal may be applied directly to the digital predistorter controller. According to one embodiment of the present invention, the optical input signal may be applied directly to the digital predistorter controller. The feedback path essentially provides a representation of the output signal to the predistortion controller 101.

In either input mode, memory effects due to self-heating, bias network and frequency dependence of the active devices are compensated for by the adaptive algorithm 204 in the DPD201 of fig. 2. In FIG. 2, the multi-band RF input x [ n ] is provided to DPD201 and DPD algorithm logic 204. The output z [ n ] of DPD201 is provided to DAC202 and logic 204. The output of DAC202 provides an input to power amplifier 203. The distortion characteristics of the PA are sensed by feedback samples y (t) which are converted to data ya [ n ] in ADC206 and provided to alignment logic 205. After alignment, data ya [ n ] provides feedback data to algorithm logic 204.

The coefficients of the DPD are adjusted by synchronizing the wideband acquisition aliased output multiband signal ya [ n ] (sampled feedback aliased signal) from the feedback path with the reference multiband signal x [ n ] (input signal). The DPD algorithm performs synchronization and compensation. The synchronization aligns the reference signal with the aliased feedback signal in the alignment block. In one embodiment of the DPD algorithm, the reference signal and the aliased sampled feedback aliased signal ya [ n ] are used in a direct learning adaptive algorithm. In another embodiment of the DPD algorithm, the aliased pre-distorted signal za [ n ] (pre-distorted output aliased signal) and the sampled feedback aliased signal ya [ n ] are used in an indirect learning adaptive algorithm.

In some embodiments, Crest Factor Reduction (CFR) is applied prior to DPD with an adaptive algorithm in a digital processor to reduce PAPR, EVM and ACPR and compensate for memory effects and changes in linearity due to temperature changes of the PA. The digital processor may take almost any form; for convenience, an FPGA implementation is typically used, but a general purpose processor is also acceptable in many implementations. The CFR implemented in the digital module of an embodiment is based on proportional iterative pulse cancellation (scaled iterative pulse cancellation) as proposed in the following patent applications: this patent application is US61/041,164 entitled "AnEffectiontPeak CancellationsMethod for Reducingthe Peak-To-AveragePowerRationWidebandCommunicationSystems", filed 3, 31.2008, which is incorporated herein by reference. CFR is included to enhance performance, and thus is optional. CFR may be deleted from an embodiment without affecting overall functionality.

In all embodiments, memory effects due to self-heating, bias network and frequency dependence of the active devices are compensated by an adaptive algorithm in DPD. The coefficients of the DPD are adjusted by synchronizing the wideband captured output signal from the feedback path with the reference signal. The digital predistortion algorithm performs synchronization and compensation. The predistorted signal passes through DQM to generate a real signal, which is then converted to an IF analog signal via a DAC. In all embodiments, DQM need not be implemented in an FPGA or at all. If DQM is not used in the FPGA, an AQM implementation can be implemented with two DACs to generate real and imaginary signals, respectively.

Fig. 3 is a block diagram illustrating a Predistortion (PD) portion in the DPD system of the present invention. The PD in the present invention typically uses a polynomial based adaptive digital predistortion system. Another embodiment of the PD uses a look-up table (LUT) based digital predistortion system. More specifically, the PD shown in fig. 3 is processed in a digital processor by the following adaptive algorithm: the adaptive algorithm is set forth in U.S. patent application No. 11/961,969 entitled "advanced for base and predistrionation LinearationSystems". The PD of the PD system in fig. 3 has a plurality of Finite Impulse Response (FIR) filters, i.e., FIR1301, FIR2303, FIR3305, and FIR 4307. The PD also includes a third order product generation block 302, a fifth order product generation block 304, and a seventh order product generation block 306. The output signals from the FIR filters are combined in a summation block 308. The coefficients of the plurality of FIR filters are updated by a digital predistortion algorithm.

Digital predistorter algorithm

Digital Predistortion (DPD) is a technique for linearizing a Power Amplifier (PA). Fig. 2 shows a block diagram of a digital predistortion PA system. In the DPD block, a memory polynomial model is used as the predistortion function (fig. 3).

Wherein, a_ijAre DPD coefficients.

In the DPD estimator block, the least squares method is used to solve the DPD coefficient a_ijThen the DPD coefficient a_ijAnd sending to the DPD block. Fig. 4 and 5 show the DPD algorithm in detail. These coefficients are obtained using a QRRLS adaptation algorithm in the DPD estimation block.

FIG. 4 shows one of the multi-band digital predistortersAn embodiment is described. The direct learning adaptive algorithm has two inputs into the DPD estimator. DPD estimator uses aliased sampled multiband signal xa [ n ]](sampled input aliased signal) as a reference and using the sampled feedback aliased signal ya [ n ]]As an input. Thus, the x (n) signal is provided to DPD400 and frequency translation/aliasing logic 420. DPD output signal z (n), while aliasing logic 420 outputs xa (n) and provides xa (n) to integer delay logic 402, integer delay logic 402 then provides signals to fractional delay logic 403 and multiplexer 414, multiplexer 414 also receiving the output of logic 403. The multiplexer also receives a control signal from the delay estimator 406 and provides an output to a block xa '404, which block 404 determines xa' (n-m). The output of the multiplexer is also provided to a data buffer 405, the data buffer 405 providing its output to a DPD estimator 412 and control signals to a phase shift block 410 and a gain correction block 411. Feedback signal y (t) and sampling frequency F_sIs provided to ADC421, where F_sIs selected to produce a suitable Nyquist zone for aliased signals. The ADC output is provided to a feedback data buffer 408, and ya (n) data from block 407 is also provided to the feedback data buffer 408. The data buffer 408 provides data to the delay estimator 406 and also to the phase shifter 410 and thus to the gain correction block 411. The gain correction is provided to the DPD estimator which is multiplexed with the already in-memory predistortion coefficients shown in 401 back to the DPD 400.

Fig. 5 shows another embodiment of a multi-band digital predistorter. The indirect learning adaptive algorithm of fig. 5 has two inputs into the DPD estimator. The DPD estimator uses the predistorted output aliased signal za n as a reference and the sampled feedback aliased signal ya n as an input, but is otherwise very similar to fig. 4 and will not be described further.

Fig. 6 shows a diagram of a spectral domain diagram. Reference input signal x [ n ]]Showing a center frequency of F_aAnd F_bTwo different frequency bands. The operating bandwidth of each carrier is small compared to the frequency spacing between carriers. Non-linearities in a power amplifier will arise as inThe spectral regeneration shown in the analog feedback multiband signal y (t) (analog feedback signal). The analog feedback signal being down-converted to an intermediate frequency F_iMedium frequency F_iIs selected to be in the first nyquist zone as shown in fig. 6. Frequency F_iDepending on the ADC sampling rate or F used_sOperating bandwidth of the carriers and frequency spacing (F) between carriers_a-F_b). ADC sampling rate F_sIs usually F_iThe limiting factor at the time of selection. For a two carrier system, in at least some embodiments, F_aLocated in the first Nyquist zone and F_bIn the second Nyquist zone, however, depending on the particular implementation, signal F_aOr F_bCan be located in the third, fourth or nth nyquist zone. The basic requirement is that only the two signals are located in different nyquist zones. F_iSelection constraint F_aAnd F_bHow far apart from each other.

Sampling of the analog feedback signal y (t) generates an image as shown in the spectrum of the sampled feedback aliased signal ya [ n ]. The nonlinear distortions from the various carriers are allowed to alias with each other as long as the aliased part of the signal does not adversely affect the original multi-band signal. The direct learning algorithm uses the difference between xa [ n ] and ya [ n ] to minimize the resulting error signal. In the DPD estimator, the QRRLS algorithm uses the error to adjust the predistorter coefficients. The indirect learning algorithm first models the power amplifier using the pre-distorted output aliased signal za [ n ] and the sampled feedback aliased signal ya [ n ]. The modeled power amplifier coefficients are then used to calculate predistorter coefficients.

Fig. 7 is a block diagram illustrating an analog quadrature modulator compensation structure. The analog quadrature modulator converts the baseband signal output from the DAC into an RF frequency. Input signal is divided into in-phase components X_IAnd the orthogonal component X_Q. The analog quadrature modulator compensation structure comprises four real filters { g11, g12, g21, g22} and two DC bias compensation parameters c1, c 2. The DC bias in AQM will be compensated by the parameters c1, c 2. The frequency dependence of AQM will be compensated by filters g11, g12, g21, g 22. Order of real filterDepending on the level of compensation required. Output signal Y_IAnd Y_QWill be presented to the in-phase and quadrature ports of the AQM.

In summary, the multi-band wideband power amplifier predistortion system of the present invention can significantly reduce feedback ADC sampling rate requirements. This will enable multi-band broadband applications and reduce power consumption and cost. The system is also reconfigurable and field programmable because the algorithms and power efficiency enhancement features can be adjusted in the digital processor at any time like software, as discussed in more detail in the application incorporated by reference and attached as an appendix.

Furthermore, the multiband wideband DPD system does not need to know the modulation scheme such as QPSK, QAM, OFDM, etc. in CDMA, GSM, WCDMA, CDMA2000 and wireless LAN systems. This means that DPD system can support multiple modulation schemes, multiple carriers and multiple channels. Other benefits of DPD systems include: correction of PA non-linearity in repeaters or indoor coverage systems that do not have readily available necessary baseband signal information.

Further, the techniques provided by the present disclosure may be configured as follows:

(1) a digital predistortion system comprising:

a radio frequency input signal;

a power amplifier for providing an amplified output including distortion characteristics;

a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

predistortion logic responsive to the feedback signal and the input signal, the predistortion logic to generate an aliased representation of the input signal and to derive predistortion coefficients based on the aliased signal.

(2) A multi-band digital predistortion system comprising:

a multi-band input signal, wherein each band is centered around a spaced frequency and the bandwidth of each band is much lower than the frequency spacing between bands;

at least one power amplifier for providing an amplified output comprising distortion characteristics;

input aliasing logic to produce an aliased image for each frequency band, wherein the image for a first frequency band is in a first nyquist zone and the image for a second frequency band is in a different nyquist zone;

a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

predistortion logic responsive to the aliased image, the predistortion logic to generate predistortion coefficients that linearize the output of the power amplifier.

Although the present invention has been described with reference to preferred embodiments, it is to be understood that the invention is not limited to the details described in the preferred embodiments. Various alternatives and modifications have been suggested in the foregoing description, and others will occur to those skilled in the art. Accordingly, all such substitutions and modifications are intended to be included within the scope of the present invention as defined in the appended claims.

Claims (60)

1. A multi-band digital predistortion system comprising:

a multi-band input signal, wherein the frequency bands are centered around spaced frequencies and the bandwidth of each frequency band is substantially smaller than the frequency spacing between the frequency bands;

at least one power amplifier for providing an amplified output comprising distortion characteristics;

input aliasing logic to produce an aliased image for each frequency band, wherein the aliased image for a first frequency band is in one nyquist zone and the aliased image for a second frequency band is in another nyquist zone, each of the one and other nyquist zones having a width of half a sampling rate of the input signal;

a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

predistortion logic responsive to the aliased image, the predistortion logic to generate predistortion coefficients that linearize the output of the power amplifier.

2. The multi-band digital predistortion system of claim 1 wherein the one nyquist zone is in a first nyquist zone extending from 0Hz to half the sampling rate of the input signal.

3. The multi-band digital predistortion system of claim 1 wherein the other nyquist zone is in a second nyquist zone extending from half the sampling rate of the input signal to the sampling rate of the input signal.

4. The multi-band digital predistortion system of claim 1 wherein one band of the multi-band input signal is centered about a frequency within the one nyquist zone and another band of the multi-band input signal is centered about a frequency within the other nyquist zone.

5. The multi-band digital predistortion system of claim 1 wherein the predistortion logic generates the predistortion coefficients using an adaptive, polynomial based algorithm.

6. The multi-band digital predistortion system of claim 1 wherein the predistortion coefficients are stored in a look-up table.

7. The multi-band digital predistortion system of claim 1 wherein the predistortion coefficients are updated using a direct learning adaptive learning algorithm that receives the multi-band input signal as an input.

8. The multi-band digital predistortion system of claim 1 wherein the predistortion coefficients are updated using an indirect learning adaptive learning algorithm that receives the aliased image as an input.

9. The multi-band digital predistortion system of claim 1 further comprising alignment logic that aligns the feedback signal with the multi-band input signal.

10. The multi-band digital predistortion system of claim 1 further comprising a digital-to-analog converter that converts the multi-band input signal after having been processed using the predistortion coefficients into an analog signal.

11. A method for amplifying a signal, the method comprising:

receiving a multi-band input signal, wherein the frequency bands are centered about spaced frequencies and the bandwidth of each frequency band is substantially less than the frequency spacing between the frequency bands;

pre-distorting the multi-band input signal using pre-distortion coefficients generated by pre-distortion logic such that a first aliased image of a first frequency band of the multi-band input signal is in one nyquist zone and a second aliased image of a second frequency band of the multi-band input signal is in another nyquist zone, each of the one and the other nyquist zones having a width of half a sampling rate of the input signal;

amplifying the predistorted multiband input signal to generate an amplified output, the amplified output comprising distortion characteristics;

wherein the predistortion coefficients are updated using a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristics.

12. The method of claim 11, wherein the one nyquist zone is in a first nyquist zone extending from 0Hz to half of the sampling rate of the input signal.

13. The method of claim 11, wherein the other nyquist zone is in a second nyquist zone, the second nyquist zone extending from half the sampling rate of the input signal to the sampling rate of the input signal.

14. The method of claim 11, wherein one band of the multi-band input signal is centered about a frequency within the one nyquist zone and another band of the multi-band input signal is centered about a frequency within the another nyquist zone.

15. The method of claim 11, wherein the predistortion logic generates the predistortion coefficients using an adaptive, polynomial based algorithm.

16. The method of claim 11, wherein the predistortion coefficients are stored in a look-up table.

17. The method of claim 11, further comprising:

comparing the multi-band input signal and the feedback signal using a direct learning adaptive learning algorithm; and

updating the predistortion coefficients using the comparison.

18. The method of claim 11, further comprising:

comparing an aliased image of the multiband input signal and an aliased image of the feedback signal using an indirect learning adaptive learning algorithm; and

updating the predistortion coefficients using the comparison.

19. The method of claim 11, further comprising aligning the feedback signal with the multiband input signal.

20. The method of claim 11, further comprising converting the multi-band input signal after having been predistorted from an analog signal to a digital signal.

21. A digital predistortion system comprising:

a multi-band input signal;

at least one power amplifier for providing an amplified output comprising distortion characteristics;

a feedback signal derived from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

predistortion logic responsive to an aliased representation of the feedback signal, the predistortion logic to generate predistortion coefficients that linearize the amplified output of the power amplifier, wherein a sampling rate of the aliased representation of the feedback signal is less than twice a maximum bandwidth of the feedback signal.

22. The digital predistortion system of claim 21 wherein the sampling rate of the feedback signal is less than the sampling rate of the multiband input signal.

23. The digital predistortion system of claim 21 wherein the predistortion logic is responsive to a sampled representation of the multi-band input signal.

24. The digital predistortion system of claim 21 further comprising a digital predistorter receiving the multiband input signal, the digital predistorter affecting the predistortion coefficients derived by the predistortion logic.

25. The digital predistortion system of claim 21 wherein the digital predistorter utilizes an adaptive polynomial based digital predistortion system that utilizes the predistortion coefficients generated by the predistortion logic.

26. The digital predistortion system of claim 25 wherein the digital predistorter utilizes a look-up table based digital predistortion system having a finite impulse response filter whose coefficients include the predistortion coefficients generated by the predistortion logic.

27. The digital predistortion system of claim 21 further comprising alignment logic that aligns the feedback signal relative to the multi-band input signal.

28. The digital predistortion system of claim 21 wherein the frequency of the aliased representation of the feedback signal is in the nyquist zone.

29. The digital predistortion system of claim 21 wherein the predistortion coefficients are derived using a direct learning adaptive algorithm.

30. The digital predistortion system of claim 21 wherein the predistortion coefficients are derived using an indirect learning adaptive algorithm.

31. The digital predistortion system of claim 21 wherein the predistortion coefficients are derived to compensate for memory effects due to one or more of frequency dependence of active devices in the digital predistortion system, bias networks and self-heating.

32. A method for amplifying a signal, the method comprising:

receiving a radio frequency input signal;

detecting an amplified output from a power amplifier, the output being responsive to the radio frequency input signal and comprising distortion characteristics;

deriving a feedback signal from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

generating, via predistortion logic responsive to an aliased representation of the feedback signal, a predistortion coefficient to linearize the amplified output of the power amplifier, wherein a sampling rate of the aliased representation of the feedback signal is less than twice a maximum bandwidth of the feedback signal.

33. The method of claim 32, wherein a sampling rate of the feedback signal is less than a sampling rate of the radio frequency input signal.

34. The method of claim 32, wherein the predistortion logic is responsive to a sampled representation of the radio frequency input signal.

35. The method of claim 32, further comprising:

receiving a second radio frequency input signal; and

pre-distorting the second radio frequency input signal using the pre-distortion coefficients derived by the pre-distortion logic with a digital pre-distorter.

36. The method of claim 32, wherein predistorting the second radio frequency input signal comprises utilizing an adaptive, polynomial-based digital predistortion system that utilizes the predistortion coefficients derived by the predistortion logic.

37. The method of claim 32, further comprising aligning the feedback signal relative to the radio frequency input signal.

38. The method of claim 32, wherein a frequency of the aliased representation of the input signal is in a nyquist zone.

39. The method of claim 32, wherein the predistortion coefficients are derived using a direct learning adaptive algorithm.

40. The method of claim 32, wherein the predistortion coefficients are derived using an indirect learning adaptive algorithm.

41. A digital predistortion system comprising:

a digital predistorter configured to receive a radio frequency input signal and output a predistorted output signal;

a power amplifier configured to receive the pre-distorted output signal and output an amplified output;

a digital predistortion estimator configured to receive an aliased representation of a feedback signal derived from the amplified output and to compute predistortion coefficients based on the aliased representation of the feedback signal.

42. The digital predistortion system of claim 41 further comprising frequency transform logic configured to receive the predistorted output signal and generate an aliased representation of the predistorted output signal.

43. The digital predistortion system of claim 42, wherein the digital predistortion estimator is further configured to receive the aliased representation of the predistorted output signal to compute the predistortion coefficients.

44. The digital predistortion system of claim 43 wherein the digital predistortion estimator is configured to compute predistortion coefficients based on the aliased representation of the predistorted output signal in addition to the aliased representation of the feedback signal.

45. The digital predistortion system of claim 41 wherein the sampling rate of the aliased representation of the predistorted output signal is less than twice the maximum bandwidth of the feedback signal.

46. The digital predistortion system of claim 41 wherein the radio frequency input signal is a multi-band input signal.

47. The digital predistortion system of claim 41 wherein the digital predistorter is configured to affect the predistortion coefficients derived by the digital predistortion estimator.

48. The digital predistortion system of claim 41 wherein the digital predistorter utilizes an adaptive polynomial based digital predistortion system that utilizes the predistortion coefficients generated by the digital predistortion estimator.

49. The digital predistortion system of claim 41 wherein the digital predistorter utilizes a look-up table based digital predistortion system having a finite impulse response filter, wherein coefficients of the finite impulse response filter comprise the predistortion coefficients derived by the digital predistortion estimator.

50. The digital predistortion system of claim 41 further comprising alignment logic that aligns the feedback signal relative to the radio frequency input signal.

51. The digital predistortion system of claim 41 wherein the predistortion coefficients are derived to compensate for memory effects due to one or more of frequency dependence of active devices in the digital predistortion system, bias networks and self-heating.

52. A method for amplifying a signal, the method comprising:

receiving a first radio frequency input signal;

detecting a predistorted output signal from a digital predistorter, the predistorted output signal being responsive to the first radio frequency input signal;

detecting an amplified output from the power amplifier, the output including distortion characteristics;

deriving a feedback signal from the amplified output, the feedback signal comprising a representation of at least a portion of the distortion characteristic; and

generating, by a digital predistortion estimator, predistortion coefficients linearizing the amplified output of the power amplifier based on the feedback signal and an aliased representation of the predistorted output signal.

53. The method of claim 52, further comprising: generating, by frequency transform logic, the aliased representation of the predistorted output signal from the predistorted output signal.

54. The method of claim 52, further comprising outputting the predistortion coefficients to the digital predistorter.

55. The method of claim 54, further comprising:

receiving a second radio frequency input signal; and

pre-distorting the second radio frequency input signal using the pre-distortion coefficients.

56. The method of claim 55, wherein predistorting the second radio frequency input signal comprises utilizing an adaptive polynomial based digital predistortion system that utilizes the predistortion coefficients derived by the digital predistortion estimator.

57. The method of claim 55 wherein predistorting the second radio frequency input signal comprises utilizing a look-up table based digital predistortion system having a finite impulse response filter, wherein coefficients of the finite impulse response filter comprise the predistortion coefficients derived by the digital predistortion estimator.

58. The method of claim 52, further comprising aligning the feedback signal relative to the first radio frequency input signal.

59. The method of claim 52, further comprising converting the feedback signal from analog to digital form.

60. The method of claim 52, further comprising converting the predistorted output signal from a digital form to an analog form.