A storm-based p300 real-time distributed computing method

Abstract

The present invention relates to a P300 real-time distributed computing method based on Storm, which mainly includes the following steps: first, the Spout worker in Storm receives the original EEG data block from the front desk in real time, repackages it and transmits it to the Bolt worker; Then, Bolt workers are responsible for a series of signal processing, and improve the efficiency of data processing through parallel computing. After completing the P300 classification work, select the EEG fragments containing P300 components, and record the corresponding blinking row and column numbers; finally , the Bolt worker sends the recorded row number back to the front desk, and the front desk determines the feedback result based on the returned row number. The performance improvement of the P300 Speller brain-computer interface requires the introduction of various algorithms, but the requirements of the brain-computer interface on the feedback time limit the range of algorithms used. The invention enables more algorithms to play a role in the P300 Speller brain-computer interface in a distributed computing manner.

Description

technical field [0001] The invention belongs to the combined application of the field of cognitive neuroscience and the field of information technology, and relates to the computing mechanism of the P300Speller brain-computer interface, specifically a Storm-based P300 real-time distributed computing method. Background technique [0002] Brain-computer interface is a new technology of human-computer interaction, which can provide assisted living ability for patients with motor function loss but intact brain function, and can also expand the communication ability of normal people. P300 Speller is a way of brain-computer interface. Its function is to identify the characters that the user wants to output by analyzing the user's EEG signal, so as to help the user communicate with the outside world. Currently, a series of processing in P300 Speller is based on serial calculation. When some algorithms are introduced to improve performance, serial computing requires a long computin...

AI Technical Summary

Problems solved by technology

When some algorithms are introduced to improve performance, serial computing requires a long computing time, which is not conducive to the timely response of the brain-computer interface, resulting in performance degradation on the other hand.

Images