Multi-dimensional signal processing and adaptive learning-based hybrid brain-computer interface system

Abstract

The invention discloses a multi-dimensional signal processing and adaptive learning-based hybrid brain-computer interface system. The system comprises a multi-dimensional signal acquisition module, adecoding unit, an execution unit and an adaptive learning module, wherein the multi-dimensional signal acquisition module is used for acquiring physiological signals in a plurality of dimensionalitiesand carrying out amplification and analog-digital conversion on the physiological signals; the decoding unit is used for respectively extracting useful information from the multi-dimensional signals,namely, carrying out multi-dimensional signal feature extraction, carrying out fusion calculation on features, and finally carrying out task decision making according to the feature fusion result; the execution unit is used for executing corresponding functions according to tasks given by the decoding unit; and the adaptive learning module is used for carrying out learning by users before using the system for the first time, executing set behavior tasks according to an operation guide, acquiring physiological activity signals in the task execution process and carrying out calculation to obtain best parameters of feature extraction and task classification. The system has an adaptive learning function and is capable of carrying out parameter adjustment according to different environments, time and using individuals, so that the system control correctness is ensured to the greatest extent.

Description

technical field [0001] The invention relates to the technical field of brain-computer interface, in particular to a hybrid brain-computer interface system. Background technique [0002] The brain generates certain characteristic signals to realize communication and control with electronic devices (such as computers), which is called brain-computer interface (Brain Computer Interface, BCI). for the correct command to control the device. The physiological activity of the brain can be measured by means of electroencephalogram (EEG), magnetic resonance imaging (fMRI), near-infrared optical imaging (fNIR), or electrode invasion into the scalp. There are three types of brain-computer interface systems based on EEG signals: P300 events, steady-state visual evoked potentials (SSVEP), and imagined movements. [0003] At present, the biggest application of brain-computer interface is the neurorehabilitation of people with severe movement disorders (such as stroke, disability), and i...

AI Technical Summary

Problems solved by technology

Multi-task type is a prerequisite for complex control (for example, the movement control of a robotic arm needs to give control signals in three dimensions), and the brain-computer interface based on EEG signals needs to be improved in this respect

Images