Touch Sensor Noise Suppression Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Touch screens face noise interference from components like display circuitry, which affects the accuracy of touch data, particularly in portable devices where adding shielding materials is undesirable.
Innovation Solution
Implementing machine learning techniques, such as gated recurrent units and convolutional neural networks, within the electronic device's display and touch chips to estimate and remove noise from touch data using image data as input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If shielding materials are added to reduce noise in touch data, then noise reduction effectiveness is improved, but device size and weight increase
Solution Approach 1:
The patent replaces physical shielding materials with a software-based machine learning solution. A neural network model processes touch data to identify and remove noise components caused by display circuitry, substituting mechanical/physical noise reduction methods with computational algorithms that achieve the same goal without adding physical mass to the device.
Solution Approach 2:
The patent changes the approach from physical parameter modification (adding shielding materials) to data parameter processing. By analyzing touch data parameters and display circuitry characteristics through machine learning, the system dynamically adjusts noise reduction strategies based on operational conditions, achieving effective noise suppression without physical modifications.
2Object-affected harmful factors
If shielding materials are added to reduce noise in touch data, then noise reduction effectiveness is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex physical shielding structures with a computational approach. Instead of designing and integrating physical barriers between display and touch components, the system uses machine learning algorithms to differentiate between valid touch signals and noise from display circuitry, simplifying the overall device architecture.
Solution Approach 2:
The patent makes existing components multi-functional. The display driver circuitry not only drives the display but also provides characteristic data to the machine learning model for noise identification. The touch controller simultaneously performs touch detection and noise reduction processing, reducing the need for separate dedicated shielding components.
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
This approach effectively reduces noise in touch data, enhancing the accuracy of touch detection and user experience without increasing device size by using data processing techniques instead of additional shielding materials.
Implementation Method 1
one or more signals transmitted to the display circuitry of an electronic device can become capacitively coupled to the touch circuitry of the device and cause noise in the touch data
Data Source
AI summary
In some examples, touch data can include noise. The noise can be generated by a component of an electronic device that includes a touch screen. For example, one or more signals transmitted to the display circuitry of an electronic device can become capacitively coupled to the touch circuitry of the device and cause noise in the touch data. Machine learning techniques, such as gated recurrent units and/or convolutional neural networks can estimate and reduce or remove noise from touch data when provided data or information about the displayed image as input. In some examples, the algorithm includes one or more of a gated recurrent unit stage and a convolutional neural network stage. In some examples, a gated recurrent unit stage and a convolutional neural network stage can be arranged in series, such as by providing the output of the gated recurrent unit as input to the convolutional neural network.