Brain-computer interface (BCI) technologies allow direct connection between the activity of the brain and the activity of an external device. Communication is established as a result of neural responses that occur within the brain. Muscular and non-muscular acts are included in the definition of means of communication in this context. These signals from the brain are applied to a physical device to fulfill a certain job. They are extremely valuable and have the potential to enhance or completely replace human peripheral working capabilities, with applications in various fields. Nonetheless, despite the effectiveness of BCIs, their uptake has been minimal. Recently, there is a significant increase in research on BCI. This review paper examines past and present research investigations on BCI as well as the implications of these findings. The BCI research work begins with the acquisition of brain signals and proceeds through a pipeline that includes pre-processing, extracting important features from processed signals, classification of signals using emerging technologies such as machine learning and deep learning, evaluation of the designed model, and finally interface with devices in the real world. This review, first, examines the different signal acquisition methods, each monitor’s different functional brain activity, including electrical and magnetic to identify brain activity, and shows the various types of signal acquisition methods. Second, it includes the procedures that can be applied to raw signals to pre-process them. Third, it describes the feature extraction algorithm that was used to extract information from the BCI signal. Next, the classification method and evaluation techniques are discussed in detail. Finally, explain how BCI can be applied in several real-world situations. This review will provide a better understanding to the researchers who are willing to start the work on BCI.

Brain, Signal, Transform, Accuracy, Features, Pre-processing.