Vol.11 (2021) No. 2 ISSN: 2088-5334 Analysis of the Information Transfer Rate-ITR in Linear and Non-linear Feature Extraction Methods for SSVEP Signals Danni De la Cruz a,b,* , Wilfredo Alfonso b , Eduardo Caicedo b a Electrical and Electronics Department, Universidad de la Fuerzas Armadas-ESPE, Sangolqui, 170501, Ecuador b School of Electrical and Electronics Engineering, Universidad del Valle, Cali, 76001, Valle del Cauca, Colombia Corresponding author: * drde@espe.edu.ec; danni.cruz@correounivalle.edu.co Abstract— The most popular paradigm in BCIs is the steady-state visually evoked potential (SSVEP) due to their advantages, such as the high information transfer rate (ITR), the time spent on users in the training phase, and the capacity to discriminate each stimulus. One of the most influential factors in the ITR evaluation is the feature extraction methods since these can increase the accuracy. Here, we compare nine methods for the extraction of features from SSVEP signals to identify those with better performance, according to the time window (TW), its technology (equipment and number of nodes), and the value of ITR. The study identifies two groups: the first one is characterized by presenting variations of correlated component analysis (CCA), which is highly used to increase the ITR due to its efficiency in classification and its capacity of response to reduction (TW), such as MsetCCA, IT-CCA, FBCCA; the second one are the representation special based methods that consider the non-linear nature of the electroencephalogram (EEG) signal such as TRCA, CORRCA, EMD, and VMD. The results show a considerable difference between these groups. The maximum ITR value for FBCCA was 117.75 [bits/min] in a TW of 1.25s, while the VMD method achieved 3120 [bits/min] in a TW of 1s, respectively. The comparison covers signals between 0.55 and 8 seconds, taking into account visual strain, the experimental environment, and other artifacts. Keywords— Steady-state visually evoked potential; brain-computer interfaces; information transfer rate; canonical correlation analysis; empirical mode decomposition. Manuscript received 10 Nov. 2020; revised 15 Jan. 2021; accepted 25 Feb. 2021. Date of publication 30 Apr. 2021. IJASEIT is licensed under a Creative Commons Attribution-Share Alike 4.0 International License. I. INTRODUCTION Controlling a device, robot, or another machine using only thoughts has been a fantastic notion that has long captured humanity's imagination and interest. In the last decade, this has become a proven reality by avoiding the conventional communication channels (i.e., muscles or speech) between the brain and a computer. A "Brain-Computer Interface (BCI)" gives the users an alternative communication channel linking their brains and external devices. The BCI allows the control of applications using brain signals without the requirement of using the peripheral nervous system, benefiting access to people with limited motor skills, and developing alternative access methods for healthy users [1]. In a BCI system, the user must generate mental activities to produce voluntary changes in brain signals. These activities can be exogenous or endogenous: the exogenous one depends on the electrophysiological activity evoked by external stimuli (for example, the P300, Visually Evoked Potential (VEP), and the Steady-State Visually Evoked Potentials (SSVEP)); the endogenous one depends on the capacity of the user to control his electrophysiological activity without the need for external stimulation [2]. The paradigm of SSVEP stands out for its minimal training capacity, robustness, high Information Transfer Rate (ITR), and high Signal to Noise Ratio (SNR) [3]. The SSVEP paradigm is a spontaneous response to visual stimuli with specific frequencies through the retina, which emits stimuli to the brain, generating a response with the same spectrum [4]. The stimulus normally appears in the occipital and parietal brain lobes, where it is possible to gather much evidence in a relatively short time [4]. BCI systems based on VEP or SSVEP stimuli have demonstrated successful integration from single or multi- frequency coding. The user transmits different commands by switching their attention to different coded targets [5]. According to Scopus, In the last three years, from 14,016 published works about Brain-Computer Interface, 1,088 761