Multi-Touch Noise Detection via Frequency Variance Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional touch and proximity-based input systems face challenges in effectively rejecting noise, especially when multiple simultaneous or near-simultaneous touch events are detected, as existing noise rejection techniques can be impaired by noise on one or more channels, leading to degraded signal-to-noise ratios.
Innovation Solution
The proposed solution involves a two-clean-frequency and one-clean-frequency noise detection method, which determines mismatches in sample values obtained at different frequencies, identifies noisy frequencies by comparing variances to a predetermined threshold, and computes touch values excluding noisy frequencies to improve noise rejection. Additionally, a combined method is used to adapt to varying noise levels, incorporating statistical analyses like F-tests and unsharp mask filtering to differentiate between touch and noise signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple separate stimulus waveforms with unique frequencies are applied to sense points, then noise rejection capability is improved, but device complexity increases due to multiple frequency channels requiring processing
Solution Approach 1:
The patent segments the noise detection process into distinct frequency channel analyses. Each stimulus frequency is processed independently to compute variances and identify noisy channels, allowing targeted noise rejection without requiring complex full-spectrum processing. This segmentation enables reliable noise rejection by analyzing each frequency component separately and combining results.
2Reliability
If statistical measures like mean, weighted mean, or median are used to combine sample values, then resistance to noise degradation is improved, but measurement precision decreases when noise is present on one or more channels
Solution Approach 1:
The patent applies local quality by treating each frequency channel differently based on its noise characteristics. Instead of uniformly applying statistical measures to all channels, the system identifies noisy channels through variance analysis and applies noise rejection specifically to those channels while preserving clean channels. This localized approach maintains measurement precision by avoiding unnecessary processing of already-clean data while targeting problematic channels.
Solution Approach 2:
The patent changes the parameter of signal combination by moving from simple statistical averaging to a selective combination process. The system dynamically adjusts which frequency channels contribute to the final touch detection based on real-time noise assessment. This parameter change enables the system to maintain high measurement precision by excluding noisy channels from the combination process rather than relying solely on statistical robustness.
3Measurement precision
If variance comparison and analysis of variance are performed on sample values, then accurate identification of noisy frequencies is improved, but computational load and processing time increase
Solution Approach 1:
The patent applies preliminary action by computing variances for each frequency channel before performing the full analysis of variance. This preliminary variance computation provides a quick assessment of noise levels in each channel, allowing the system to prioritize or skip detailed ANOVA processing for certain channels. This preliminary step reduces overall processing time while maintaining accurate noise identification through subsequent targeted analysis.
Data Source
AI summary
Multi-touch touch-sensing devices and methods are described herein. The touch sensing devices can include multiple sense points, each of which can be stimulated with a plurality of periodic waveforms having different frequencies to measure a touch value at the sense point. Noise at one or more of the frequencies can interfere with this measurement. Therefore, various noise detection (and rejection) techniques are described. The noise detection techniques include two-clean-frequency noise detection, one-clean-frequency noise rejection, and combined two-clean-frequency/one-clean-frequency noise detection. Each of the noise detection techniques can include statistical analyses of the sample values obtained. The touch sensing methods and devices can be incorporated into interfaces for a variety of electronic devices such as a desktop, tablet, notebook, and handheld computers, personal digital assistants, media players, and mobile telephones.