Compensation method for I/Q channel imbalance errors

• US 5,705,949 A
• Filed: 09/13/1996
• Issued: 01/06/1998
• Est. Priority Date: 09/13/1996
• Status: Expired due to Term

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: US Robotics Access Corp. (US Robotics Incorporated)
• Inventors: Parizhsky, Vladimir, Norrell, Andrew L., Alelyunas, Carl H.
• Primary Examiner(s): MIS, DAVID C
• Application Number:
  US08/712,927
• Time in Patent Office: 480 Days
• Field of Search: 329/304-310, 329/315, 329/318-321, 329/341-343, 375/317, 375/324, 375/346, 375/349, 375/350
• US Class Current: 329/304
• CPC Class Codes: H03D 3/008 Compensating DC offsets
• CPC Class Codes: H03D 3/009 Compensating quadrature phase or amplitude imbalances
• CPC Class Codes: H04L 2027/0016 Stabilisation of local oscillators

Abstract

A method in which correctable I/Q imbalance errors in a complex receiver can be detected and compensated for digitally without the use of special calibration signals. Differential D.C. offset errors are compensated by averaging the incoming I_d and Q_d digital signals and subtracting from them an expected value of differential D.C. offset, for example, computed from the long term average of the I and Q signals to create I'"'"' and Q'"'"' signals. Differential gain imbalance errors are corrected by calculating a root means square average of the I'"'"' and Q'"'"' digital signals and applies to them compensation coefficients K_x and K_y determined from either the RMS average or from a Stochastic Gradient Algorithm. The DSP compensates for the quadrature phase errors by calculating a compensation matrix which is independent of the frequency of the carrier and applies the compensation matrix to the I'"'"' and Q'"'"' digital signals. The compensation matrix for quadrature phase errors is completely independent of the frequency of the input carrier signal supplied to the complex receiver, and is not dependent on the use of a calibration signal. The compensation may be performed as a step in calibration of the complex receiver, or continuously.

Claims

1. In a communication system, a detection and compensation method for I/Q channel imbalance errors for I and Q analog signals generated by a complex receiver, comprising the steps of:

receiving an incoming analog carrier signal at said complex receiver and responsively generating said I and Q analog signals, said I and Q analog signals nominally separated in phase by 90 degrees and having differential offset, differential gain imbalance and quadrature phase errors, said quadrature phase error proportional to a factor ε

related to the phase difference between said I and Q analog signals;

converting said I and Q analog signals to I_d and Q_d digital signals;

compensating for said differential offset errors by averaging the incoming I_d and Q_d digital signals and subtracting from them an expected value values of differential D.C. offset to create offset compensated signals I'"'"' and Q'"'"';

compensating for said differential gain imbalance errors by calculating a root mean square average of the I'"'"' and Q'"'"' digital signals and applying to them compensation coefficients K_x and K_y derived from either said root mean square average or a stochastic gradient algorithm to create gain compensated signals I" and Q";

compensating for said quadrature phase errors by calculating a quadrature compensation matrix independent of the frequency of the carrier and applying said compensation matrix to said I" and Q" signals to generate I'"'"'" and Q'"'"'" output signals,

whereby said I'"'"'" and Q'"'"'" output signals compensate for said differential offset, differential gain and quadrature phase errors without the use of calibration signals, thereby improving the performance of said communication system.

2. The method of claim 1, wherein said incoming analog carrier signal comprises an RF signal.

3. The method of claim 2, wherein said differential offset and differential gain imbalance errors are compensated as part of calibrating said complex receiver using a specific reference carrier as said input analog carrier signal.

4. The method of claim 1, wherein said incoming analog carrier signal comprises an FM signal.

5. The method of claim 4, wherein said step of compensation of quadrature phase errors is performed substantially continuously.

6. The method of claim 1, wherein said compensation of said quadrature phase errors is performed by applying a stochastic gradient algorithm to said I" and Q" signals.

7. In a communication system, a compensation method for I/Q channel quadrature phase errors in I and Q analog signals generated by a complex receiver, comprising the steps of:

receiving an FM analog carrier input signal with said complex receiver and responsively generating I and Q analog signals having a quadrature phase error;

calculating a compensation matrix independent of the frequency of said FM analog carrier input signal and without the use of a calibration signal, said compensation matrix related to said quadrature phase error between said I and Q analog signals; and

applying said compensation matrix to said I and Q signals to compensate for said quadrature phase errors present in said I and Q signals to responsively produce I'"'"'" and Q'"'"'" output signals separated in phase by substantially 90 degrees.

8. The method of claim 7, wherein said I and Q analog signals are converted to I_d and Q_d digital signals and said compensation factor matrix is applied to said I_d and Q_d digital signals.

9. The method of claim 7, wherein the entries in said compensation matrix are calculated according to a stochastic gradient algorithm.

10. The method of claim 7, wherein said compensation matrix entries are determined using a sum-of-squares calculation for I'"'"'" and Q'"'"'".

11. The method of claim 7, wherein said quadrature phase compensation is performed to calibrate said complex receiver.

12. The method of claim 7, wherein said quadrature phase errors are corrected continuously by a digital signal processor.

13. The method of claim 10 or claim 11, wherein said quadrature phase compensation is performed in a cellular telephone system.

14. The method of claim 10 or claim 11, wherein said quadrature phase compensation is performed in a cellular telephone modem.

15. A method for detecting gain variations in a complex receiver output, comprising the steps of:

receiving an incoming analog carrier signal at said complex receiver and responsively generating said I and Q analog signals, said I and Q analog signals nominally separated in phase by 90 degrees and having a differential gain imbalance error;

converting said I and Q analog signals to I_d and Q_d digital signals;

compensating for said differential gain imbalance errors by calculating a root mean square average of the I_d and Q_d digital signals and applying to said I_d and Q_d signals compensation coefficients K_x and K_y derived from either a root mean square average of said I_d and Q_d signals or from a stochastic gradient algorithm.

16. The method of claim 15, wherein said incoming analog carrier signal is an FM signal.

17. The method of claim 15, wherein said step of compensation is performed continuously.

18. The method of claim 15, wherein said step of compensation is performed in a digital signal processor.

19. The method of claim 15, wherein said step of compensation is performed as a calibration of said complex receiver.

20. Communication apparatus, comprising:

a complex receiver for receiving an analog input carrier signal and responsively generating I and Q analog signals;

means for converting said I and Q analog signals into I_d and Q_d digital signals;

an error compensator means receiving said I_d and Q_d digital signals for compensating for differential gain, differential D.C. offset and quadrature phase errors in said I and Q analog signals to produce I'"'"'" and Q'"'"'" compensated signals,

said error compensator means compensating for said differential gain, differential offset and phase errors substantially independent of changes in frequency of said analog input carrier signal and without the use of a calibration signal.

21. The apparatus of claim 20, wherein said error compensator means further comprises a digital signal processor computing the long term averages of said I_d and Q_d digital signals individually and using said long term average to compensate for differential DC offset errors in said I and Q analog signal individually.

22. The apparatus of claim 20, wherein said error compensator means further comprises a digital signal processor computing the long term root mean square averages of said I and Q digital signals individually and using said long term root mean square averages to compensate for differential gain errors in said I_d and Q_d analog signal individually.

23. The apparatus of claim 20, wherein said error compensator means further comprises a digital signal processor that Computes a compensation matrix from a sum-of-squares calculation for said I_d and Q_d signals and applies said compensation matrix to said I_d and Q_d signals to correct for quadrature phase errors, said compensation matrix independent from changes in the frequency of said input carrier signal.

24. The apparatus of claim 23, wherein said correction of quadrature phase errors is performed after compensation for said differential D.C. offset and gain errors has been performed on said I and Q signals.

25. The apparatus of claim 20, wherein said error compensation means comprises a digital signal processor and wherein said compensation for said quadrature phase errors is performed in said digital signal processor continuously by calculating a compensation matrix according to stochastic gradient algorithm and applying said compensation matrix mathematically to said I_d and Q_d signals.

26. The apparatus of claim 20, wherein said error compensation means comprises a digital signal processor and wherein said compensation for said quadrature phase errors is performed in said digital signal processor by

a) shifting the phase of said I_d and Q_d signals into a 180 degree phase relationship and adding said I_d and Q_d signals to thereby determine the magnitude of the phase error in said I and Q signals,

b) using the arcsine of the resultant amplitude error as an argument in a compensation matrix, and

c) applying said compensation matrix to said I_d and Q_d signals.