Least mean squares adaptation of a concurrent multi-band pre-distorter using overlapping spines

Patent metadata

Field	Value
Publication number	US-10447211-B2
Application number	US-201515737004-A
Country	US
Kind code	B2
Filing date	Jun 17, 2015
Priority date	Jun 17, 2015
Publication date	Oct 15, 2019
Grant date	Oct 15, 2019

Abstract

A method and system for digital pre-distortion of an input signal to compensate for non-linear operation of a power amplifier. According to one aspect, some embodiments provide overlapping spline functions that are defined for two adjacent bins, where any two spline functions overlap in only one bin. Each spline function is computed as a function of one of an input signal envelope and a delayed signal envelope. According to another aspect, a tap weight evaluator includes a least mean squares, LMS, tap correlator updater configured to modulate a step size of an adaptation process to update each tap weight, the step size being modulated based on an approximate logarithm of the average power of the input to a tap weight computation.

First claim

What is claimed is:

1. A method for digital pre-distortion, DPD, of an input signal to compensate for a non-linear operation of a power amplifier, the method comprising:
for each of at least one band of the input signal:
determining a signal envelope, a signal envelope being determined for each of a plurality of successive bins, each bin having a knot at each edge of the corresponding bin;
computing a spline function assigned to each knot, the spline function having a left spline and a right spline, each of the left spline and the right spline having a value of unity at the knot and a value of zero at left and right neighboring knots, respectively, so that each bin is associated with two overlapping successive spline functions that overlap each other in only one bin, each spline function being computed as a function of one of the signal envelope and a delayed signal envelope in the band;
for each bin, delaying one of the signal envelope and a spline function in each of at least one tap delay line to model a pre-selected memory depth;
determining a tap weight for each knot;
multiplying each spline function at each knot by its respective tap weight; and
generating a pre-distorted signal using a sum of tap-weighted products of the spline functions and a delayed input signal.

2. The method of claim 1 , further comprising:
when there are a plurality of bands:
forming cross products of the spline functions of the different bands;
multiplying the cross products at each knot by a respective tap weight; and
wherein generating the pre-distorted signal includes adding a sum of tap-weighted cross products to the sum of tap-weighted products.

3. The method of claim 1 , further comprising:
when there are a plurality of bands:
forming inter-band polar spline functions having a magnitude and phase based on the signal envelopes of the plurality of bands;
multiplying the inter-band polar spline functions at each of a plurality of radial knots by a respective tap weight at each knot; and
wherein generating the pre-distorted signal includes adding a sum of tap-weighted polar spline functions to the sum of tap-weighted products.

4. The method of claim 1 , further comprising, for each band, scaling the signal envelope and computing the spline function as a function of one of the scaled signal envelope and a delayed scaled signal envelope.

5. The method of claim 1 , further comprising, for each band, inputting the signal envelope to a pre-emphasis unit that maps each value of the signal envelope to a scaled value of the signal envelope.

6. The method of claim 5 , wherein the mapping is based on cumulative distribution function of the signal envelope.

7. The method of claim 1 , further comprising selecting, by multiplexers, different ones of the delayed input signals and spline functions to be included in the pre-distorted signal.

8. The method of claim 1 , wherein the generating includes, for each band:
storing each tap weight in a memory;
multiplying the tap weights by spline functions in parallel branches to produce a first set of multiplications;
combining the first set of multiplications; and
multiplying the combined multiplications by a delayed input signal.

9. The method of claim 1 , wherein a spline function is given by:

ψ
2

⁡

(

a
,
u

)

=

{

1
2

⁢

(

u

a
⁢

/

⁢
2

)

2

,

0
≤
u
≤

a
⁢

/

⁢
2

1
-

1
2

⁢

(

u
-
a

a
⁢

/

⁢
2

)

2

,

a
⁢

/

⁢
2

≤
u
≤

3
⁢
a
⁢

/

⁢
2

1
2

⁢

(

2

Assignees

• Ericsson Telefon Ab L M

Inventors

• Rollins Mark
• Fuller Arthur T G

Classifications

• H03F1/3252 Primary

  using multiple parallel paths between input and output (H03F1/3258, H03F1/3282, H03F1/3294 take precedence) · CPC title

• H03F1/3247

  using feedback acting on predistortion circuits (H03F1/3264 takes precedence) · CPC title

• H03F1/3258

  based on polynomial terms · CPC title

• H03F1/3294

  Acting on the real and imaginary components of the input signal · CPC title