Abstract: In particular for a spelling programme application, this research offers an optimisation technique for steady state visual evoked potential (SSVEP)-based brain computer interface (BCI). Stimulator, signal processing, and application (spelling programme) are the minimum number of components required to construct a comprehensive BCI application in this application. To achieve the best performance, the three components should ideally operate on independent processing units. However, combining those three parts into a single computer system has other benefits, such as improving concentration and simplifying system setup.
The spelling system and the jerky animation are the two key components that require optimisation. We will concentrate on the display driver technology and programming factors while optimising the flickering animation. The layout and representation of the letter matrix will be the main areas of emphasis for spelling system optimisation. We put our program's frequency range, frequency resolution, and frequency stability to the test across many computers.
Conclusion: Using a computer monitor as the stimulator, it can be concluded that, regardless of the software technology used (DirectX or OpenGL), the maximum synthesizable stimulator frequency is always half of its minimum refresh-rate. With a frequency resolution of 0.11 Hz, the highest synthesizable frequency of up to 30 Hz is attained. 106 people participated in our system testing at CeBIT 2008 in Hannover, Germany. The spelling system's average accuracy is 92.5%. Therefore, without substantial expert assistance or pricy hardware, the optimisation approach outlined here resulted in a stable and dependable system that performed well across the majority of subjects.
|
DOI:
10.17148/IJARCCE.2023.125157
