Removing effects of gain and phase mismatch in a linear amplification with nonlinear components (LINC) system

• US 6,737,914 B2
• Filed: 01/31/2002
• Issued: 05/18/2004
• Est. Priority Date: 12/07/2001
• Status: Active Grant

• Current Assignee: Jian Gu
• Original Assignee: 4D Connect, Inc.
• Inventors: Gu, Jian
• Primary Examiner(s): Mottola, Steven J.
• Application Number:
  US10/061,810
• Publication Number:
  US 20030107435A1
• Time in Patent Office: 838 Days
• Field of Search: 330/2, 330/124.R, 330/149
• US Class Current: 330/2
• CPC Class Codes: H03F 1/0294 using vector summing of two or more constant amplitude phase-modulated signals
• CPC Class Codes: H03F 3/211 using a combination of several amplifiers H03F3/60 takes precedence

Abstract

A method for removing effects of gain and phase mismatch in amplification branches of a linear amplification using nonlinear components (LINC) system. The method includes receiving an input signal, calculating a relative phase and gain difference in the amplification branches, and generating phasing components. The input signal is then controllably separated into a plurality of branch signals of different but constant envelope. The mismatch between branches may cause each branch signal to have a different envelop. The phases of the branch signals are then appropriately adjusted in a certain amount of corresponding phasing components, such that when the branch signals are recombined, the combined signal substantially replicates the input signal.

Claims

1. A method for removing effects of gain and phase mismatch in amplification branches of a linear amplification using nonlinear components (LINC) system, comprising:

receiving calibration signals;

calculating a relative phase and gain difference in the amplification branches, generating a look-up table for phasing component generation;

receiving an input signal;

generating phasing components; and

controlling separation of the input signal into a plurality of branch signals of different but constant envelopes by appropriately applying the phasing components to the amplification branches, such that when the branch signals are recombined, the combined signal substantially replicates the input signal.

2. The method of claim 1, wherein said calculating a relative phase and gain difference includes

modulating and sending two calibration signals and demodulating the signals using a receiver.

3. The method of claim 2, wherein said modulating and sending two calibration signals and demodulating the signals includes:

generating two sets of branch signals, the first set of branch signals S₁(t)=A and S₂ (t)=Be^jx forming the first transmission signal and the second set of branch signals S₁(t)=A and S²(t)=−

Be^jx forming the second transmission signal, where A and B are the reference vectors and X can be any constant selected in the modulation process; and

,

using a receiver to demodulate the first received signal R₁=G(A+Be^jx) and the second received signal R₂=G (A−

Be^jx) and determine A and B up to a constant complex scale by processing the two received signals with (R₇+R₂)/2 and (R₁−

R₂)/(2e^jx), where G is a complex number representing the gain or attenuation of the receiving path.

4. The method of claim 1, wherein said generating a look-up table for phasing component generation includes

configuring a generalized elliptic curve based on the calculated reference vectors A and B and establishing one-to-one mapping between the points on the curve and possible values of the input signal'"'"'s magnitude.

5. The method of claim 1 wherein said generating phasing components includes

obtaining a point on the generalized elliptic curve corresponding to a vector having same magnitude as the input signal.

6. The method of claim 5, wherein said obtaining a point on the elliptic curve includes searching for an appropriate value for Φ

such that |S(t)|=|(A+B)·

cos Φ

+j(B−

A)·

sin Φ

|, where |·

| stands for the magnitude of the argument, S(t) is the input signal, and A and B are the reference vectors.

7. The method of claim 4, wherein said generating a look-up table for phasing component generation includes

tabulating the parameter Φ

as a function of any possible values of the input signal'"'"'s magnitude |S(t)|, such that |S(t)|=|(A+B)·

cos Φ

+j(B−

A)·

sin Φ

|.

8. The method of claim 5, wherein said obtaining a point on the elliptic curve includes

searching the lookup table for the Φ

related to the point on the elliptic curve that corresponds to a vector having the same magnitude of the input signal.

9. The method of claim 4, wherein said generating a lookup table for phasing component generation includes

calculating a phase skew which is the phase difference between Z₃ and Z₁, where Z₃=·

cos Φ

+j(B−

A)·

sin Φ

is a point on the generalized elliptic curve corresponding to a parameter Φ and

Z₁=(B+A), and

tabulating the phase skew Δ

as a function of any possible values of Φ

or, equivalently, any possible values of the input signal'"'"'s magnitude |S(t)|.

10. The method of claim 1 wherein said generating phasing components includes

searching the look-up table for the corresponding Φ

and Δ

given an input signal'"'"'s magnitude |S(t)|.

11. The method of claim 1, wherein said controlling separation of the input signal into a plurality of branch signals includes

applying appropriate phasing components θ

(t)−

Δ

−

Φ and

θ

(t)−

Δ

Φ

to phase modulators in amplification branches to generate at least following branch signals S₁(t)=A·

e<sup>j(θ

(t)−

Δ

−

Φ

)</sup> and S₂(t)=A·

e<sup>j(θ

(t)−

Δ

+Φ

)</sup>, where S₁(t) is a first branch signal and S₂(t) is a second branch signal, A and B are reference vectors, θ

(t) is phase of the input signal S(t)=|S(t)|e<sup>jθ

(t)</sup>, Δ

is the corresponding phase skew, Δ and

Φ

are associated with a point on the generalized elliptic curve that relates a vector having the same magnitude as the input signal.

12. The method of claim 11, wherein said controlling separation of the input signal into a plurality of branch signals includes

approximating S(t)=|S(t)|e^jθ

by V_min ·

e^jθ

with a corresponding Φ

achieving the minimum when $\left|S<br><br>\left(t\right)\right|<<br><br>V_{\min }=\underset{0\le <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\varphi <br><br>}\le <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}/2}{\min }<br><br>\left\{\left|\left(A+B\right)·<br><br>\cos <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\varphi <br><br>}+j<br><br>\left(B-A\right)·<br><br>\sin <br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\varphi <br><br>}\right|\right\},$

where V_min defines the radius of a dead circle within which an input signal may only be approximated with a two-branch LINC system with branch mismatches.

13. The method of claim 12, where said controlling separation of the input signal into a plurality of branch signals includes

providing at least three branch signals to avoid the dead circle.

14. The method of claim 13, where said providing at least three branch signals includes

providing a third branch signal with a vector whose magnitude is larger than the radius of the dead circle.

15. The method of claim 14, where said providing a third branch signal moves the dead circle away from a null position by forming a new input signal with magnitude larger than the radius of the dead circle.