Sliding-Window Phase Calibration for Wireless Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional phase calibration methods in wireless communication systems suffer from inaccuracies due to non-linear phase-frequency responses, leading to errors in phase measurement calibration across the entire frequency band.
Innovation Solution
The method involves generating a frequency-domain calibration sequence, performing frequency-domain transformation, determining phase differences, and using a sliding-window approach with a window function for phase fitting, which divides the frequency band into sub-bands for more accurate phase calibration by fitting phases within each sub-band.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If linear fitting is performed to the phase of the whole frequency band, then the phase calibration factor for the whole frequency band can be determined, but the accuracy of phase calibration is reduced due to non-linear phase-frequency response characteristics
Solution Approach 1:
The frequency band is divided into multiple sub-frequency bands, and phase fitting is performed separately for each sub-band. This segmentation allows the non-linear phase-frequency response to be approximated as linear within each smaller sub-band, thereby improving phase calibration accuracy while managing complexity through modular processing
Solution Approach 2:
Different phase calibration characteristics are applied to different frequency sub-bands. By treating each sub-band locally with its own linear fitting, the method adapts to the local phase-frequency response characteristics, improving overall accuracy compared to a single global fitting approach
2Ease of manufacture
If the phase is assumed to be completely linear within the whole frequency band, then the phase calibration process is simplified, but errors are introduced due to the non-ideal phase-frequency response characteristics of analog devices
Solution Approach 1:
The frequency band is divided into multiple sub-frequency bands, and phase fitting is performed separately for each sub-band. This segmentation allows the non-linear phase-frequency response to be approximated as linear within each smaller sub-band, thereby improving phase calibration accuracy while managing complexity through modular processing
Solution Approach 2:
The method changes the parameter scope by transitioning from a global frequency band assumption to local sub-band assumptions. By adjusting the frequency range parameter for each sub-band, the linear approximation becomes more accurate while maintaining computational feasibility
Data Source
Figure 1
Figure 2
Figure 3a~3c
AI summary
A method and device for increasing an accuracy of phase measurement, wherein the method includes: receiving a measurement signal (101); performing frequency-domain transformation to the measurement signal to obtain a frequency-domain measurement sequence (102); determining phases that correspond to the frequency-domain measurement signals, and determining a phase difference between frequency-domain measurement signals that correspond to two neighboring specified frequency points (103); according to the phases, the phase difference and a window function, performing sliding-window-type phase fitting to the frequency-domain measurement sequence, to obtain phase-fitting data that correspond to sliding windows (104); and according to the phase-fitting data of the sliding windows, determining phase-calibration data that correspond to the sliding windows, and, by using the phase-calibration data of the sliding windows, forming phase-calibration data within the specified frequency band (105). The method reduces the error of fitting, and increases the accuracy of the phase calibration.