Capacitive Touch Noise Reduction via Bidirectional Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Capacitive touch screens face performance degradation due to noise accumulation during signal integration, particularly in differential sensing techniques, leading to reduced signal-to-noise ratio and poor touch interface performance.
Innovation Solution
A two-stage noise reduction approach is implemented, involving median and exponential filtering of differential touch data, followed by integration from both ends of the touch panel and calculating a weighted average using location-dependent weights to minimize noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If differential sensing technique is used in capacitive touch screens, then touch sensitivity and multi-touch capability are improved, but noise accumulation during signal integration increases
Solution Approach 1:
The patent segments the integration process into multiple discrete stages: differential sensing stage, filtering stage (median and exponential), integration stage, and weighted averaging stage. Each stage processes the signal separately, allowing noise reduction techniques to be applied at specific points without compromising the overall touch detection accuracy.
Solution Approach 2:
The patent introduces intermediary processing steps between differential sensing and final integration. Median filtering and exponential filtering act as intermediaries that clean the signal before integration, while weighted averaging serves as a final intermediary that reduces accumulated noise. These intermediary processes preserve the benefits of differential sensing while mitigating noise accumulation.
2Reliability
If noise filtering is applied to differential touch data, then signal-to-noise ratio is improved, but processing complexity increases
Solution Approach 1:
The patent changes the parameters of the filtering process by using different filtering methods (median filtering with specific kernel sizes, exponential filtering with different time constants) at different stages. Weighted averaging with location-dependent weights further adjusts parameters based on spatial position. These parameter changes optimize noise reduction while managing processing complexity through systematic variation rather than uniform complex processing.
Solution Approach 2:
The patent applies preliminary filtering actions (median filtering followed by exponential filtering) before the main integration process. This preliminary noise reduction prepares the data for subsequent integration, reducing the burden on later processing stages and overall system complexity while maintaining reliable signal-to-noise ratio.
3Object-affected harmful factors
If integration is performed from both ends of the touch panel, then noise reduction is improved, but computational overhead increases
Solution Approach 1:
The patent introduces asymmetry through location-dependent weights in the weighted averaging process. Instead of treating all integrated values equally, the system applies different weights based on spatial location, allowing more efficient processing of regions with different noise characteristics. This asymmetric approach reduces computational overhead compared to uniform high-precision processing while maintaining effective noise reduction.
Solution Approach 2:
The patent performs integration from both ends (excessive action) but applies partial weighting in the final averaging step. Not all integrated values receive equal computational attention; instead, weights are applied selectively based on location and noise characteristics. This partial processing approach reduces computational overhead while maintaining the benefits of bidirectional integration for noise reduction.
Data Source
AI summary
A method for controlling a capacitive touch panel includes filtering differential capacitance data. The filtered data is integrated in a first direction of the touch panel to generate a first set of integrated data, and integrated in a second direction of the touch panel to generate a second set of integrated data. The first and second sets of integrated data are averaged, and a touch on the touch panel is detected based on the average of the first and second sets of integrated touch data.