I/Q Baseband Averaging With Round-to-Even Error Cancellation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mobile terminal test devices experience measurement accuracy deterioration due to non-uniform rounding errors in time average amplitude values of I and Q baseband signals, leading to biased calculations.
Innovation Solution
A mobile terminal test device with a frequency conversion unit, A/D converter, orthogonal demodulator, shift unit, rounding unit, and averaging unit that shifts and rounds-to-even the I and Q baseband signals to improve measurement accuracy by canceling out rounding errors, and includes additional units for signal processing and constellation calculation to enhance amplitude measurement accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If rounding off is performed on I and Q baseband signals to reduce logic complexity, then device complexity is reduced, but measurement precision deteriorates due to non-uniform rounding errors accumulating in biased directions
Solution Approach 1:
The patent inverts the conventional rounding approach by using rounding to the nearest even number instead of rounding off. This inversion changes the rounding behavior so that when the digit to be rounded is 5, it rounds to the nearest even number rather than always rounding up or down. This eliminates the systematic bias in rounding errors that occurs with fixed-sign signals, as the rounding errors now cancel out uniformly over time, thereby maintaining measurement precision while still reducing logic complexity.
2Ease of operation
If rounding off is performed on baseband signals, then processing simplicity is improved, but linearity error increases due to accumulated rounding errors
Solution Approach 1:
The patent applies rounding to the nearest even number instead of conventional rounding off. This inverted rounding approach ensures that rounding errors are distributed uniformly rather than accumulating in a biased direction. For signals with fixed signs (all positive or all negative), the rounding errors now cancel out over time, preventing the accumulation that leads to linearity errors, while maintaining processing simplicity.
3Productivity
If conventional rounding off is used for signal processing, then calculation efficiency is improved, but measurement accuracy deteriorates due to non-uniform error distribution
Solution Approach 1:
The patent uses rounding to the nearest even number instead of conventional rounding off. This inverted approach changes the error distribution characteristics: instead of rounding errors accumulating in a biased direction (always positive or always negative for fixed-sign signals), the errors now alternate and cancel out. This maintains calculation efficiency while dramatically improving measurement accuracy for time average amplitude values.
Data Source
AI summary
A first shift unit (14) that shifts an amplitude value of the I baseband signal and an amplitude value of the Q baseband signal in a lower direction by the predetermined number of bits so that the predetermined number of lower bits thereof have a value after a decimal point, a first rounding unit (15) that rounds-to-even the amplitude value of the I baseband signal and the amplitude value of the Q baseband signal shifted by the first shift unit (14) with a rounding width of 1 in decimal notation, and a first averaging unit (16) that respectively calculates a time average of the amplitude value of the I baseband signal and the amplitude value of the Q baseband signal rounded by the first rounding unit (15) are included.