ENHANCEMENT OF SENSOR PANEL TACTILE TOUCH INTERFACE

Abstract

Introduction. The article explores the possibility of utilizing a touch-sensitive button for the identification of two types of signals – a button press signal and a finger swipe signal across the button. This concept has the potential to revolutionize the way we interact with electronic devices, making them more intuitive and user-friendly. The ability to detect different types of gestures using a single button can also lead to a more compact and cost-effective design.

Materials and Methods. To address the classification task, an algorithm has been proposed for the identification of signal characteristics upon which recognition will be conducted. The algorithm is based on a neural network architecture, which is trained on a dataset of signals collected from a touch-sensitive button. The dataset includes a variety of signals, each corresponding to a specific gesture, such as a button press or a swipe in different directions. The neural network is designed to learn the patterns and characteristics of the signals, allowing it to accurately classify new, unseen signals. The algorithm is optimized to minimize processing time and computational resources, making it suitable for real-time applications.

Results. Using neural networks to solve the recognition task allows for the easy determination of optimal classification algorithm parameters for a specific type of touch button. The results show that the proposed algorithm achieves high accuracy in identifying two types of gestures, with a minimal error rate.

Conclusions. The proposed classification algorithm exhibits satisfactory accuracy in identifying the two signals within a minimal timeframe and requires minimal computational resources. Therefore, it can be employed cost-effectively to enhance the functionality of touch panels. The algorithm's ability to detect gestures using a single button makes it an attractive solution for applications where space is limited, such as in wearable devices or mobile phones. Additionally, the algorithm's low computational requirements make it suitable for use in low-power devices, such as those powered by batteries or energy harvesting systems. Future work can focus on improving the algorithm's accuracy and robustness, as well as exploring its application in different domains.

Keywords: gesture detection, neural network, capacitive sensor, signal profile.