I/Q Conversion Calibration Using Offset Local Oscillators

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Electronic receiving and transmission devices are vulnerable to I/Q mismatch, resulting in spurious signal components due to mismatched mixing operations, which complicates self-calibration processes.

Innovation Solution

A digital signal processor configures a signal processing circuit to switch to a calibration mode using first and second local oscillator frequencies with a frequency offset, measuring amplitudes of frequency components to determine compensation parameters for I/Q mismatch during up-conversion and down-conversion, allowing for precise calibration without adjusting anti-aliasing bandwidth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If self-calibration is performed using traditional methods (single tone test signal with complex processing), then I/Q mismatch can be eliminated, but the device complexity and calibration process complexity increase significantly

Engineering Contradiction:
ImproveI/Q mismatch compensation precisionVSAvoidcalibration process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the frequency parameter relationship between transmission and reception paths by introducing a frequency offset between the first local oscillator frequency (f_LO1) and the second local oscillator frequency (f_LO2). This parameter change enables the separation of I/Q mismatch error components in the frequency domain, allowing for simplified calibration without complex processing algorithms.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces a frequency offset as an intermediary parameter between the transmission and reception local oscillator signals. This intermediary enables the distinction and separate measurement of I/Q mismatch errors from the transmission path and reception path, respectively, simplifying the calibration process by allowing independent error characterization.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If I/Q mismatch compensation is performed without frequency offset, then the calibration process is simpler, but spurious signal components cannot be distinguished and eliminated effectively

Engineering Contradiction:
Improvecalibration operation simplicityVSAvoidsignal fidelity
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent segments the I/Q mismatch errors from the transmission and reception paths by exploiting the frequency offset. The error components appear at distinct frequency locations (f_offset ± f_signal), allowing separate identification and compensation of transmission path errors and reception path errors, thereby improving signal fidelity while maintaining calibration simplicity.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If frequency offset is introduced for calibration, then I/Q mismatch errors can be distinguished and compensated, but the local oscillator frequency configuration becomes more complex

Engineering Contradiction:
Improveerror component distinction accuracyVSAvoidlocal oscillator frequency configuration
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent implements dynamic frequency configuration where the local oscillators are set to different frequencies (f_LO1 and f_LO2 with offset Δf) during calibration mode, and can be switched to matched frequencies during normal operation. This dynamic adjustment enables precise error distinction during calibration while allowing simplified operation during normal signal processing.

Inventive Principle:
Principle #15Dynamics

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach effectively reduces I/Q mismatch errors by determining and compensating for mismatch parameters, eliminating spurious components and improving signal fidelity in both transmission and reception modes.

Implementation Method 1

a signal is mixed down with versions of a local oscillator signal that are ninety degrees out of phase with each other. This results in an in-phase signal and a quadrature signal

Methodology Applied
Scientific EffectFrequency mixing: Heterodyne

Data Source

PatentEP2229724B1Signal processing circuit and method with frequency up- and down-conversion
Publication Date: 2011.05.18 ST ERICSSON SA
  • EP2229724B1 patent drawingFigure 1~2b
  • EP2229724B1 patent drawingFigure 3a~4c
  • EP2229724B1 patent drawingFigure 4d~5

AI summary

A signal processing circuit comprises a frequency up-conversion circuit (14, 60) for performing up-conversion with a first local oscillator frequency and a frequency down-conversion circuit (16) for performing down-conversion with a second local oscillator frequency. A digital signal processor (10) controls supply first signals representing a first complex signal to the up-conversion circuit, and receives second signals representing a second complex signal. The digital signal processor controls a compensation of I/Q mismatch of results of up-conversion and/or down-conversion. The digital signal processor (10) switches to a calibration mode for selecting a parameter of said compensation. In the calibration mode the first and second local oscillator frequencies have a frequency offset with respect to each other. The digital signal processor (10) measures an amplitude of a frequency component at a frequency corresponding to mismatch in one and not more than one of the results of up-conversion and/or down-conversion, and selects the parameter dependent on the amplitude.