Signal Transmitter Linearization

• US 20130034188A1
• Filed: 10/04/2012
• Published: 02/07/2013
• Est. Priority Date: 04/04/2006
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: van Zelm, John-Peter, Tholl, David M., Rashev, Peter Zahariev
• Granted Patent: US 8,737,938 B2
• Time in Patent Office: Days
• US Class Current: 375/297
• CPC Class Codes: H03F 1/3258 based on polynomial terms
• CPC Class Codes: H04L 25/03063 using fractionally spaced delay lines or combinations of fractionally and integrally spaced taps
• CPC Class Codes: H04L 25/03343 Arrangements at the transmitter end
• CPC Class Codes: H04L 25/03828 Arrangements for spectral shaping; Arrangements for providing signals with specified spectral properties partial response systems H04L25/497

Abstract

A linearizer for a non-linear transmitter includes a tap delay line that provides samples of an input signal at selected times. At least one Volterra tap is coupled to the tape delay line. The Volterra tap includes a lookup table representation of a polynomial. An adaptive controller is coupled to the Volterra tap for modifying values in the lookup table.

Claims

1. A linearized transmitter, comprising:

a tap delay line configured to provide samples of an input signal at selected times;

a plurality of Volterra taps coupled to the tap delay line, each Volterra tap comprising a lookup table representation of a polynomial;

an adder coupled to the plurality of Volterra taps, wherein the adder is configured to sum outputs of the plurality of Volterra taps to generate a predistorted version of the input sample;

an adaptive controller coupled to the plurality of Volterra taps, wherein the adaptive controller is configured to modify values in each lookup table; and

a non-linear amplifier coupled to the adaptive controller and configured to amplify the predistorted version of the input sample to provide an amplified and linearized output.

2. The transmitter of claim 1, wherein the tap delay line comprises an array of delay elements with tap-spacing of N samples.

3. The transmitter of claim 1, wherein the tap delay line comprises elements distributed across the Volterra tap inputs and outputs.

4. The transmitter as defined in of claim 1, further comprising:

an error module coupled to an output of the non-linear amplifier and a delay compensated input signal, the error module being configured to provide an error signal to the adaptive controller.

5. The transmitter of claim 4, further comprising:

a causality delay module configured to shift the location of the present tap by a desired number of samples back in time and virtually create non-causal inverse models which work on past, present and future input samples.

6. The transmitter of claim 5, wherein the adaptive controller is configured to provide a sample-by-sample lookup table value optimization.

7. The transmitter of claim 6, wherein the adaptive controller is configured to modify a lookup table Volterra tap location value by reading its current value, modifying it with a processed error, and writing the new value to the same lookup table location.

8. The transmitter of claim 1, wherein main X-terms are created by connecting Volterra tap inputs to different outputs from the tap-delay line.

9. The transmitter of claim 1, wherein the tap delay line is configured to hold a history of the evolution of the input such that different polynomials may be applied across time.

10. The transmitter of claim 1, wherein the adaptive controller is configured to use at least two Volterra taps for a sample to provide main X-terms for the sample.

11. A method of linearizing a transmitter, comprising:

applying an input signal to a tap delay line to provide samples of the input signal at selected times;

coupling a plurality of Volterra taps to the tap delay line, each Volterra tap comprising a lookup table representation of a polynomial; and

adding outputs from the plurality of Volterra taps to provide a predistorted version of the input sample;

modifying values in each of the lookup tables; and

applying the predistorted version of the input signal to a non-linear amplifier to amplify the predistorted version of the input sample to provide an amplified and linearized output.

12. The method of claim 11, wherein the tap delay line comprises an array of delay elements with tap-spacing of N samples.

13. The method of claim 11, wherein the tap delay line comprises elements distributed across the Volterra tap inputs and outputs.

14. The method of claim 11, further comprising:

comparing an output of the non-linear amplifier and a delay compensated input signal to provide an error signal to the adaptive controller.

15. The method of claim 14, further comprising:

shifting the location of the present tap by a desired number of samples back in time to virtually create non-causal inverse models which work on past, present and future input samples.

16. The method of claim 15, further comprising:

providing a sample-by sample lookup table value optimization.

17. The method of claim 16, further comprising:

modifying a lookup table Volterra tap location value by reading its current value, modifying it with a processed error, and writing the new value to the same lookup table location.

18. The method of claim 11, further comprising:

creating main X-terms by connecting Volterra tap inputs to different outputs from the tap-delay line.

19. The method of claim 11, further comprising:

holding a history of the evolution of the input in the tap delay line such that different polynomials may be applied across time.

20. The method of claim 11, further comprising:

using at least two Volterra taps for a sample to provide main X-terms for the sample.