Touch Sensing Method Using Reference Profile Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touch sensing technologies face challenges in accurately detecting multiple touches on a touch panel, especially when touches are close or noisy, leading to difficulties in distinguishing touch positions due to interference and overlapping signals.
Innovation Solution
A touch sensing method that uses a reference touch profile to match and identify touch positions by correlating or calculating differences with sensing signals, employing techniques like SAD or MSE algorithms to find optimal centroid positions, even in noisy conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a centroid formula is used to calculate touch position, then the calculation is simple, but the calculated position drifts due to noises
Solution Approach 1:
The patent applies preliminary action by pre-storing reference touch profiles that represent ideal touch patterns in advance. These reference profiles are prepared beforehand and stored in memory, allowing the system to quickly compare actual sensing signals against known good patterns without performing complex calculations during the touch detection moment, thus maintaining simplicity while improving accuracy.
Solution Approach 2:
The patent uses copying by creating reference copies of ideal touch profiles and storing them for comparison. Instead of calculating positions from scratch using centroid formulas that are sensitive to noise, the system creates and stores template copies of what a clean touch signal should look like, then matches actual signals against these templates to identify touch positions more accurately.
2Ease of operation
If waveform peaks are used to identify touch positions, then the method is straightforward, but it fails when touches are close or noisy causing peaks to overlap or disappear
Solution Approach 1:
The system performs preliminary action by pre-preparing reference touch profiles that capture the characteristic shape of valid touch signals. These reference profiles are stored in advance and include information about expected peak patterns, valleys, and overall signal characteristics. During operation, actual signals are matched against these pre-prepared templates, allowing reliable detection even when peaks overlap or are obscured by noise.
Solution Approach 2:
The patent introduces an intermediary element - the reference touch profile - that mediates between the raw sensing signal and the touch position determination. Instead of directly interpreting potentially ambiguous waveform peaks, the system uses the reference profile as an intermediary template to guide the matching process, making the detection more robust against noise and overlapping touches.
3Device complexity
If traditional touch sensing methods are used, then the system is simple, but it cannot accurately distinguish multiple close touches due to signal overlapping
Solution Approach 1:
The patent applies segmentation by dividing the touch detection process into distinct stages: signal acquisition, reference profile matching, and position identification. The reference touch profiles themselves are segmented into characteristic features (peaks, valleys, transition regions) that can be independently evaluated. This segmentation allows the system to handle multiple overlapping touches by analyzing different segments of the composite signal against corresponding segments of reference profiles.
Solution Approach 2:
The system uses parameter changes by varying the matching criteria and evaluation parameters during the comparison process. Instead of relying on a single fixed method, the system adjusts matching parameters such as correlation thresholds, peak detection sensitivity, and profile weighting based on the specific signal characteristics. This allows accurate discrimination of multiple touches even when they are close together.
Data Source
AI summary
A touch sensing method for detecting positions of touching is disclosed. In the present invention, a reference touch profile is provided. A variable ideal combination profile indicating multiple touches is obtained by synthesizing a plurality of reference touch profiles at different positions. The variable ideal combination profile is matched with a sensing signal. Actual touch positions can be identified from the sensing signal when the matching of the variable ideal combination profile with the sensing signal is optimal.