IP Library › Granted Patent US 10,749,555
Granted Patent B2
US 10,749,555 · App. 16/243,710 · Granted Aug 18, 2020

Time-domain IQ mismatch compensator with frequency-domain observations

Inventors: Elina Nayebi (San Diego, CA); Pranav Dayal (San Diego, CA); Kee-Bong Song (San Diego, CA)
Assignee: Samsung Electronics Co., Ltd.
H04B1/1027H04L27/265
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,749,555
App. No.
16/243,710
Filed
Jan 9, 2019
Granted
Aug 18, 2020
Kind
B2
Examiner
KIM, KEVIN
Art Unit
2632
USPC
375/350
Abstract

A system, method, and electronic device for compensating in-phase (I) and quadrature (Q) mismatch (IQMM) are herein disclosed. The system includes an IQ mismatch compensator (IQMC) configured to compensate for IQMM between a time-domain I signal and a time-domain Q signal using filter weight coefficients, and output a compensated I signal and a compensated Q signal, a fast Fourier transformation (FFT) circuit configured to perform an FFT on the compensated I signal and the compensated Q signal to a frequency-domain compensated signal, and a coefficient updater configured to update the filter weight coefficients based on a frequency-domain observation of the frequency-domain compensated signal.

Claims (39)

1. A system for compensating in-phase (I) and quadrature (Q) mismatch (IQMM), comprising:

an IQ mismatch compensator (IQMC) configured to:

compensate for IQMM in a time domain, including frequency dependent IQMM, between a time-domain I signal and a time-domain Q signal using filter weight coefficients; and

output a compensated I signal and a compensated Q signal;

a fast Fourier transformation (FFT) circuit configured to perform an FFT on the compensated I signal and the compensated Q signal to a frequency-domain compensated signal; and

a coefficient updater configured to update the filter weight coefficients based on a frequency-domain observation of the frequency-domain compensated signal and to determine the updated filter weight coefficients based on a gradient of a cost function of the frequency-domain observation.

2. The system of claim 1 , wherein the IQMC is further configured to receive the updated filter weight coefficients from the coefficient updater and compensate for IQMM using the updated filter weight coefficients.

3. The system of claim 1 , wherein the gradient of the frequency-domain cost function is a function of the filter weight coefficients and the frequency-domain observation.

4. The system of claim 1 , wherein the frequency-domain cost function is based on kurtosis of the frequency-domain observation.

5. The system of claim 1 , wherein the coefficient updater is configured to update the filter weight coefficients based on a gradient descent algorithm applying the gradient of the cost function.

6. The system of claim 1 , wherein the frequency-domain cost function is based on a Euclidean distance between the frequency-domain observation and a constellation, and wherein the constellation is based on a known reference signal and channel response.

7. A method for compensating in-phase (I) and quadrature (Q) mismatch (IQMM), comprising:

compensating, with an IQ mismatch compensator (IQMC), for IQMM in a time domain, including frequency dependent IQMM, between a time-domain I signal and a time-domain Q signal using filter weight coefficients;

outputting, with the IQMC, a compensated I signal and a compensated Q signal;

transforming, with a fast Fourier transformation (FFT) circuit, the compensated I signal and the compensated Q signal to a frequency-domain compensated signal;

updating, with a coefficient updater, the filter weight coefficients based on a frequency-domain observation of the frequency-domain compensated signal; and

determining, with the coefficient updater, the updated filter weight coefficients based on a gradient of a cost function of the frequency-domain observation.

8. The method of claim 7 , further comprising:

receiving, with the IQMC, the updated filter weight coefficients from the coefficient updater; and

compensating, with the IQMC, for IQMM using the updated filter weight coefficients.

9. The method of claim 7 , wherein the gradient of the frequency-domain cost function is a function of the filter weight coefficients and the frequency-domain observation.

10. The method of claim 7 , wherein the frequency-domain cost function is based on kurtosis of the frequency-domain observation.

11. The method of claim 7 , wherein the filter weight coefficients are updated based on a gradient descent algorithm applying the gradient of the cost function.

12. The method of claim 7 , wherein the frequency-domain cost function is based on a Euclidean distance between the frequency-domain observation and a constellation, and wherein the constellation is based on a known reference signal and channel response.

13. An electronic device for compensating in-phase (I) and quadrature (Q) mismatch (IQMM), comprising:

an IQ mismatch compensator (IQMC);

a coefficient updater,

a fast Fourier transformation (FFT) circuit;

a processor; and

a non-transitory computer readable storage medium storing instructions that, when executed, cause the processor to:

compensate, with the IQMC, for IQMM in a time domain, including frequency dependent IQMM, between a time-domain I signal and a time-domain Q signal using filter weight coefficients;

output, with the IQMC, a compensated I signal and a compensated Q signal;

transform, with the FFT circuit, the compensated I signal and the compensated Q signal to a frequency-domain compensated signal;

update, with the coefficient updater, the filter weight coefficients based on a frequency-domain observation of the frequency-domain compensated signal, and

determine, with the coefficient updater. the updated filter weight coefficients based on a gradient of a cost function of the frequency-domain observation.

14. The electronic device of claim 13 , wherein the gradient of the frequency-domain cost function is a function of the filter weight coefficients and the frequency-domain observation.

15. The electronic device of claim 13 , wherein the frequency-domain cost function is based on kurtosis of the frequency-domain observation.

16. The electronic device of claim 13 , wherein the filter weight coefficients are updated based on a gradient descent algorithm applying the gradient of the cost function.

17. The electronic device of claim 13 , wherein the frequency-domain cost function is based on a Euclidean distance between the frequency-domain observation and a constellation, and wherein the constellation is based on a known reference signal and channel response.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 6, 2019
From: NAYEBI, ELINA; DAYAL, PRANAV; SONG, KEE-BONG
To: SAMSUNG ELECTRONICS CO., LTD.
Reel/Frame 048245/0888 →
Continuity (2)
Provisional Application 62736595 · Sep 26, 2018
Related Publication 20200099408A1 · Mar 26, 2020