Neural Network Sensor Controller for Proximity and Touch Input Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing display devices struggle to simultaneously detect proximity and touch inputs accurately, leading to potential misinterpretation of user interactions.
Innovation Solution
A sensor controller equipped with a neural network processing unit is used to receive sensing signals from an input sensor, employing a prediction model to distinguish between proximity and touch inputs, and a display device with a sensor controller that includes a neural network processing unit to process sensing signals and determine the type of input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a conventional sensor controller is used to detect input signals, then the device complexity is low, but the measurement precision for distinguishing proximity and touch inputs is insufficient
Solution Approach 1:
The patent replaces the conventional mechanical sensor controller with a neural network processing unit that uses machine learning algorithms to analyze sensing signals. This substitution enables the system to automatically distinguish between proximity and touch inputs through pattern recognition, significantly improving measurement precision while the integrated NN processor keeps the overall device complexity manageable.
Solution Approach 2:
The patent changes the operational parameters of the sensor controller by introducing a neural network processing unit that processes sensing signals through multiple layers of neurons. The NN processor dynamically adjusts detection parameters based on learned patterns, enabling accurate differentiation between proximity and touch inputs without requiring separate dedicated circuits for each function.
2Measurement precision
If separate processing units are used for proximity and touch detection, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent merges the proximity detection and touch detection functions into a single neural network processing unit. The NN processor receives sensing signals and uses a unified prediction model to simultaneously determine both proximity presence and touch coordinates, eliminating the need for separate processing units while maintaining high detection accuracy for both input types.
Solution Approach 2:
The neural network processing unit is designed as a universal processor that can handle multiple input detection functions. By training the neural network with diverse input patterns, a single processor unit achieves the capability to detect proximity, detect touch, and differentiate between them, providing multi-functionality without increasing device complexity.
3Reliability
If a prediction model is implemented to distinguish input types, then the reliability of input detection is improved, but the use of energy increases
Solution Approach 1:
The patent implements preliminary action by training the neural network model offline before deployment. The prediction model is pre-trained with large datasets to learn the characteristics of proximity and touch inputs, enabling it to make accurate classifications during operation with minimal real-time computational energy. This preliminary training phase transfers the energy-intensive learning process to a separate setup stage.
Solution Approach 2:
The neural network processing unit operates autonomously to classify input signals without requiring external intervention or complex decision-making logic. The NN processor self-manages the detection process by automatically analyzing sensing signals, determining input types, and outputting results, thereby reducing the energy overhead associated with external control mechanisms.
Data Source
AI summary
A sensor controller includes a neural network processing unit to receive a sensing signal from an input sensor and to output one of a first sensing result and a second sensing result using a prediction model. The sensor controller further includes a first input processing unit and a second input processing unit. The first input processing unit is to receive the first sensing result and to determine whether a first input is present based on the first sensing result. The second input processing unit is to receive the second sensing result and to calculate a coordinate signal for a second input based on the second sensing result.