Huang, YufeiCorley, Isaac2024-02-092024-02-0920199781392353905https://hdl.handle.net/20.500.12588/3092This item is available only to currently enrolled UTSA students, faculty or staff. To download, navigate to Log In in the top right-hand corner of this screen, then select Log in with my UTSA ID.The advancement and commercialization of Brain Computer Interface (BCI) research utilizing Electroencephalography (EEG) sensors is hindered by high cost and algorithm compatibility issues. Simply put, an algorithm requiring a specific number of EEG channels is incompatible with any headset with a lesser number of sensors. Current spatial interpolation methods are only commonly used for recreation of a single or few malfunctioning sensors and not known to perform well when recreating numerous channels. We prove through experiments that generative Deep Neural Network (DNN) based models can be trained to intelligently upsample the spatial resolution of a headset by recreating the necessary missing signals solely from the known low spatial resolution signals.49 pagesapplication/pdfbrain computer interfacedeep learningelectroencephalographygenerative adversarial networksneural networkssuper resolutionElectrical engineeringBioinformaticsComputer scienceThesis