Multi-Antenna DUT Quality Measurement Using Cross-Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing detectors, such as RMS detectors, fail to distinguish between noise originating from a device under test and noise added by the measurement instrument, leading to unreliable noise measurement and inability to detect small signals below the instrument's intrinsic noise level, especially in devices with multiple transmission antennas.
Innovation Solution
A method and measurement system that utilize multiple reception branches with at least two receivers each, calculating quality vectors from complex-valued measurement signals, and performing cross-correlation averaging to reduce noise from the measurement system while preserving noise from the device under test, allowing for a combined average of quality vectors to determine a quality metric.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If RMS detectors are used to determine average power, then noise analysis can be performed, but the detector cannot distinguish between noise from the device under test and noise from the measurement instrument
Solution Approach 1:
The measurement system divides the noise analysis into separate components by using multiple reception branches with independent receivers. Each receiver processes the signal independently, allowing the system to segment the noise contribution from different sources (device under test vs. measurement instrument) and analyze them separately through cross-correlation processing.
Solution Approach 2:
The patent introduces an intermediary processing stage using cross-correlation calculation between multiple receivers. This intermediary operation filters out uncorrelated noise (from the measurement instrument) while preserving correlated signal components (from the device under test), enabling accurate noise measurement without direct contamination from instrument noise.
2Measurement precision
If reference measurement is performed without the device under test, then instrument noise can be determined and subtracted, but the measurement time increases and the instrument may behave differently with or without the device
Solution Approach 1:
Instead of performing a separate reference measurement after the device is disconnected, the system performs preliminary parallel processing during the actual measurement. Multiple receivers simultaneously process the signal, and the cross-correlation operation is executed in advance during the measurement process itself, eliminating the need for subsequent reference measurements and reducing total measurement time.
Solution Approach 2:
The measurement process continues without interruption by performing cross-correlation processing during the same time period as the actual signal measurement. The useful action of noise analysis is performed continuously alongside the primary measurement, rather than requiring separate discrete measurement phases, thereby maintaining measurement continuity and reducing overall time loss.
3Measurement precision
If multiple reception branches with multiple receivers are used, then noise from the measurement system can be reduced through cross-correlation, but the device complexity increases
Solution Approach 1:
The patent merges multiple receivers and their processing results into a unified measurement system that leverages cross-correlation to achieve noise reduction. By combining the outputs of multiple receivers and applying cross-correlation processing, the system achieves superior signal-to-noise ratio while managing the complexity through integrated processing rather than separate independent analysis paths.
Data Source
AI summary
The present disclosure relates to a method of determining a quality metric for a device under test having multiple transmission antennas for transmitting multiple transmission signals. Each of the multiple transmission signals is received by a reception branch having a first receiver and a second receiver, which provide a first complex-valued measurement signal and a second complex-valued measurement signal. A first quality vector for the first receiver is calculated based on the first complex-valued measurement signal and a second quality vector for the second receiver is calculated based on the second complex-valued measurement signal. A combined average of the first quality vector and of a complex conjugate of the second quality vector is determined over a predetermined number of samples, thereby obtaining a complex-valued average signal. A quality metric based on the complex-valued average signal is determined. A measurement system for determining a quality metric is also described.