Digital predistortion methods for wideband amplifiers

• US 6,587,514 B1
• Filed: 06/16/2000
• Issued: 07/01/2003
• Est. Priority Date: 07/13/1999
• Status: Expired due to Term

• Current Assignee: Maxlinear Asia Singapore Pte Ltd. (Maxlinear Incorporated)
• Original Assignee: PMC-Sierra Incorporated (Microchip Technology Incorporated)
• Inventors: Wright, Andrew S., Klijsen, Bartholomeus T. W., Yee, Paul V., Bennett, Steven J., Hung, Chun Yeung Kevin
• Primary Examiner(s): Trost, William
• Assistant Examiner(s): Ly, Nghi H.
• Application Number:
  US09/595,988
• Time in Patent Office: 1,110 Days
• Field of Search: 375/296, 375/297, 375/295, 375/254, 375/278, 375/284, 375/285, 455/501, 455/63, 455/67.3, 455/295, 455/126, 330/149, 330/129, 330/136
• US Class Current: 375/296
• CPC Class Codes: H03F 1/3241 using predistortion circuits H03F1/3211, H03F1/3217 take precedence
• CPC Class Codes: H03F 1/3247 using feedback acting on predistortion circuits H03F1/3264 takes precedence
• CPC Class Codes: H03F 1/3294 Acting on the real and imaginary components of the input signal
• CPC Class Codes: H03F 2201/3233 Adaptive predistortion using lookup table, e.g. memory, RAM, ROM, LUT, to generate the predistortion

Abstract

A wideband predistortion system compensates for a nonlinear amplifier'"'"'s frequency and time dependent AM-AM and AM-PM distortion characteristics. The system comprises a data structure in which each element stores a set of compensation parameters (preferably including FIR filter coefficients) for predistorting the wideband input transmission signal. The parameter sets are preferably indexed within the data structure according to multiple signal characteristics, such as instantaneous amplitude and integrated signal envelope, each of which corresponds to a respective dimension of the data structure. To predistort the input transmission signal, an addressing circuit digitally generates a set of data structure indices from the input transmission signal, and the indexed set of compensation parameters is loaded into a compensation circuit which digitally predistorts the input transmission signal. This process of loading new compensation parameters into the compensation circuit is preferably repeated every sample instant, so that the predistortion function varies from sample-to-sample. The sets of compensation parameters are generated periodically and written to the data structure by an adaptive processing component that performs a non-real-time analysis of amplifier input and output signals. The adaptive processing component also implements various system identification processes for measuring the characteristics of the power amplifier and generating initial sets of filter coefficients. In an antenna array embodiment, a single adaptive processing component generates the compensation parameters sets for each of multiple amplification chains on a time-shared basis. In an embodiment in which the amplification chain includes multiple nonlinear amplifiers that can be individually controlled (e.g., turned ON and OFF) to conserve power, the data structure separately stores compensation parameter sets for each operating state of the amplification chain.

Claims

1. A method of predistorting a wideband digital input signal to a power amplifier to compensate for nonlinearities in an amplification process, comprising:

generating a first index that represents a first characteristic of the wideband digital input signal, and a second index that represents a second characteristic of the wideband digital input signal;

using at least the first and second indexes to look up a set of compensation parameters from a multi-dimensional data structure such that each index corresponds to a respective dimension of the multi-dimensional data structure, wherein the compensation parameters including filter coefficients for filtering the wideband digital input signal; and

predistorting the wideband digital input signal in real time based at least in part on the set of compensation parameters;

whereby a predistortion function applied to the wideband digital input signal is dependent upon at least the first and second characteristics of the wideband digital input signal.

2. The method as in claim 1, further comprising repeating the method on an input sample by input sample basis.

3. The method as in claim 1, further comprising generating a third index using an output of a temperature sensor that measures a transistor die temperature of the power amplifier, and using the third index in combination with the first and second indexes to look up the set of compensation parameters from the multi-dimensional data structure.

4. The method as in claim 1, further comprising preconditioning and compressing the wideband digital input signal to reduce an amplitude peak-to-average ratio of the input signal.

5. The method as in claim 1, wherein predistorting the wideband digital input signal comprises filtering the input signal with a finite impulse response filter using filter coefficients read from the multi-dimensional data structure.

6. The method as in claim 1, wherein the set of compensation parameters includes IQ modulator correction coefficients.

7. The method as in claim 1, further comprising applying automatic gain control to the input signal after the input signal has been predistorted.

8. The method as in claim 1, wherein the wideband digital input signal has a spectral occupancy that exceeds 0.1% of the input signal'"'"'s RF carrier frequency.

9. The method as in claim 1, wherein the wideband digital input signal has an envelope that varies in amplitude and phase.

10. The method as in claim 1, wherein predistorting the wideband digital input signal comprises correcting for at least amplitude dependent and frequency dependent variations in the gain and phase rotation introduced by the power amplifier.

11. The method as in claim 1, further comprising preconditioning and compressing the wideband digital input signal to reduce an amplitude peak-to-average ratio.

12. The method as in claim 1, further comprising, during normal operation of the power amplifier:

storing samples of signals that correspond to the digital input signal and an output of the amplifier;

processing the samples in non-real-time to generate multiple sets of compensation parameters; and

storing the multiple sets of compensation parameters in the multi-dimensional data structure for subsequent use.

13. The method as in claim 12, wherein processing the samples in non-real-time comprises applying at least one of the following algorithms to the samples:

LMS, RLS, Kalman.

14. The method as in claim 1, further comprising generating a third index which represents a third characteristic of the digital input signal, and using the third index in combination with the first and second indexes to look up the set of compensation parameters from the multi-dimensional data structure.

15. The method as in claim 14, wherein the first characteristic is an instantaneous amplitude or power of the input signal, the second characteristic is an integrated signal envelope of the input signal, and the third characteristic is a rate of change of the input signal.

16. The method as in claim 1, wherein predistorting the wideband digital input signal comprises processing the signal with a quasi-static filter that is cascaded with a dynamic filter, wherein coefficients of the dynamic filter are updated more frequently than coefficients of the quasi-static filter.

17. The method as in claim 16, wherein coefficients used by the dynamic filter are read from the multi-dimensional data structure, and the coefficients of the quasi-static filter are generated dynamically without being stored in the multi-dimensional data structure, such that use of the quasi-static filter reduces coefficient data storage requirement.

18. The method as in claim 1, wherein the first characteristic is an instantaneous characteristic of the digital input signal, and the second characteristic is an averaged characteristic of the digital input signal.

19. The method as in claim 18, wherein the first characteristic is an amplitude or power of the input signal, and the second characteristic is an integrated signal envelope of the input signal.

20. The method as in claim 19, further comprising, during normal operation of the power amplifier, adaptively updating filter coefficients of an integration filter used to generate the integrated signal envelope.

21. The method as in claim 1, wherein the power amplifier comprises a power splitter that divides a predistorted input signal into a plurality of component signals, and further comprises an array of nonlinear amplifiers in which each nonlinear amplifier of the array amplifies a respective one of the plurality of component signals, and wherein the method further comprises adjusting operating points of individual nonlinear amplifiers of the array based on current traffic conditions.

22. The method as in claim 21, wherein adjusting operating points comprises turning off individual nonlinear amplifiers of the array during relatively low traffic conditions.

23. The method as in claim 21, further comprising storing within the multi-dimensional data structure (a) a first set of compensation parameters that corresponds to a first set of amplifier operating points of the array, and (b) a second set of compensation parameters that corresponds to a second set of amplifier operating points of the array.

24. A method of predistorting a wideband digital input signal to a power amplifier to compensate for nonlinearities in an amplification process, comprising:

providing a data structure which contains sets of compensation parameters that are indexed according to at least one signal characteristic;

providing a compensation circuit that digitally predistorts the wideband input signal prior to upconversion and amplification using sets of compensation parameters read from the data structure;

during operation of the amplifier, repeatedly generating at least one index that is based on a selected characteristic of the wideband digital input signal, and using the at least one index to load a set of compensation parameters from the data structure into the compensation circuit for use;

capturing sequences of signal samples that represent the wideband input signal and a corresponding output of the power amplifier, and processing the signal samples to generate updates to the sets of compensation parameters stored within the data structure; and

storing the updates to the sets of compensation parameters in the data structure for subsequent use.

25. The method as in claim 24, wherein the compensation circuit comprises a finite impulse response (FIR) filter, and the method comprises loading sets of FIR filter coefficients from the data structure into the FIR filter based on the at least one index.

26. The method as in claim 24, wherein storing the updates to the sets of compensation parameters in the data structure comprises using an interpolation method to smooth transitions between current and new sets of compensation parameters.

27. The method as in claim 24, wherein storing the updates to the sets of compensation parameters in the data structure comprises storing sets of compensation parameters into a table portion that is not currently available for use, and then switching the table portion into use to replace an outdated table portion.

28. The method as in claim 24, wherein the data structure is a multi-dimensional data structure, and the method comprises:

generating a first index that represents a first characteristic of the wideband digital input signal and a second index that represents a second characteristic of the wideband digital input signal; and

using the first and second indexes in combination to look up a set of compensation parameters from the multi-dimensional data structure.

29. The method as in claim 28, further comprising generating a third index, and using the third index in combination with the first and second indexes to look up the set of compensation parameters from the multi-dimensional data structure.

30. The method as in claim 29, wherein generating the third index comprises using an output of a temperature sensor that measures a transistor die temperature of the power amplifier.

31. The method as in claim 28, wherein the first characteristic is an instantaneous characteristic of the wideband digital input signal, and the second characteristic is an averaged characteristic of the wideband digital input signal.

32. The method as in claim 31, wherein the second characteristic is an integrated signal envelope of the wideband digital input signal.

33. The method as in claim 32, further comprising, during normal operation of the power amplifier, adaptively updating filter coefficients of an integration filter used to generate the integrated signal envelope.

34. In a wideband amplifier system that includes a digital predistortion circuit, a method of adaptively adjusting the digital predistortion circuit during operation of the amplifier, comprising:

providing a data structure that stores sets of compensation parameters, including filter coefficients, that are used by the digital predistortion circuit, wherein sets of compensation parameters are read from the data structure for use in real time based on at least on characteristic of a digital input signal to the amplifier system;

capturing and storing digital signal sequences that represent the digital input signal and a corresponding output of the amplifier system;

processing the digital signal sequences in a non-real-time mode to generate updates to the sets of compensation parameters; and

storing the updates to the sets of compensation parameters in the data structure during operation of the amplifier.

35. The method as in claim 34, wherein storing the updates to the sets of compensation parameters in the data structure comprises storing sets of compensation parameters in a table portion that is not currently available for use, and then switching the table portion into use to replace an outdated table portion.

36. The method as in claim 34, further comprising applying an interpolation algorithm to the compensation parameters to reduce noise caused by transistions between parameter sets.

37. The method as in claim 34, wherein the updates include finite impulse response (FIR) filter coefficients used by the digital predistortion circuit.

38. The method as in claim 37, wherein storing the updates in the data structure comprises storing sets of FIR coefficients within a multi-dimensional data structure in which at least two of the dimensions correspond to different respective characteristics of the digital input signal.

39. The method as in claim 38, wherein said characteristics include (1) an instantaneous amplitude or instantaneous power, and (2) a past power profile.