Capacitive Touch Filtering in the Sequency Domain for Gloved Input
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional capacitive touch systems face challenges in reliably detecting touch events by gloved fingers due to smaller changes in capacitance, leading to false negatives or false positives from noise and contaminants, and require tuning for specific scenarios.
Innovation Solution
Transforming raw capacitance signals from the time domain to the sequency domain using a Walsh-Hadamard transform allows for improved signal processing with better signal-to-noise ratio and reduced computational complexity, enabling more accurate detection of touch and release events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional capacitive touch systems use raw capacitance threshold comparison to detect touch events, then the system is simple to implement, but the system produces false negatives for gloved fingers and false positives from noise and contaminants
Solution Approach 1:
The patent replaces conventional time-domain signal processing with a frequency-domain approach using Fast Fourier Transform (FFT) to convert raw capacitance samples into frequency spectrum data. This substitution enables more reliable touch detection by analyzing frequency characteristics rather than raw amplitude values, effectively distinguishing genuine touch events from noise and contaminants while maintaining computational efficiency through optimized FFT algorithms.
2Adaptability or versatility
If the system is tuned to detect touch by a bare finger, then bare finger touch detection is reliable, but the system produces false negatives for gloved finger touch
Solution Approach 1:
The patent transforms the detection parameter from raw capacitance amplitude to frequency spectrum characteristics through FFT analysis. This parameter change enables the system to detect touch events based on their frequency signature rather than absolute capacitance values, making the detection consistent across different finger conditions (bare finger, gloved finger, wet finger) while maintaining high reliability. The frequency-domain representation captures the dynamic characteristics of touch events that are independent of the finger's capacitive properties.
3Measurement precision
If the system uses lower threshold to detect gloved finger touch, then gloved finger detection sensitivity increases, but false positives from noise and contaminants increase
Solution Approach 1:
The patent moves the detection problem from one dimension (raw capacitance amplitude) to another dimension (frequency spectrum). By analyzing the frequency characteristics of capacitance changes rather than their magnitude, the system can distinguish between genuine touch events and spurious signals from noise or contaminants. This dimensional transformation allows the system to maintain high sensitivity for detecting subtle gloved finger touches while achieving high specificity by rejecting false positives based on their frequency signatures.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The Walsh-Hadamard transform enhances the system's ability to distinguish between touch and proximity events, particularly for gloved fingers, and reduces false detections by improving sensitivity and specificity, while requiring less computational power compared to Fourier-based methods.
Implementation Method 1
The control circuit provides an excitation voltage to, and thereby generates an electric field about, the sensor electrode. This electric field establishes a capacitance, sometimes referred to as a parasitic capacitance, from the sensor electrode to ground or another reference potential
Implementation Method 2
Introduction of a stimulus, for example, a user's finger, to or near the sensor electrode or corresponding touch surface establishes an additional capacitance, sometimes referred to as a finger capacitance, from the electrode, through the finger, to ground, thereby changing the overall sensor electrode-to-ground capacitance
Data Source
AI summary
A transform is used to transform raw sensor data from the time domain to the frequency or sequency domain. The transformed data falls into several signal bins. The transformed data in at least one of the signal bins is analyzed to determine whether a touch event or release event has occurred.