一种多码元时分编码SSVEP-BCI少训练检测算法  

A SSVEP-BCI Less-Training Detection Algorithm Based on Multi-Symbol Time Division Coding

作　　者：张舒玲 杨晨[1] 张洪欣[1,2] 叶晓晨 ZHANG Shuling;YANG Chen;ZHANG Hongxin;YE Xiaochen(School of Electronic Engineering,Beijing University of Posts and Telecommunications,Beijing 100876,China;Beijing Key Laboratory of Work Safety Intelligent Monitoring,Beijing University of Posts and Telecommunications,Beijing 100876,China)

机构地区：[1]北京邮电大学电子工程学院,北京100876 [2]北京邮电大学安全生产智能监控北京市重点实验室,北京100876

出　　处：《北京邮电大学学报》2022年第6期40-45,共6页Journal of Beijing University of Posts and Telecommunications

基　　金：国家自然科学基金项目(62006024);航空科学基金项目(2019ZG073001)。

摘　　要：稳态视觉诱发电位(SSVEP)脑-机接口(BCI)是无创脑-机接口研究领域的3种主流范式之一。使用传统SSVEP-BCI有训练算法需预先采集大量被试者的脑电数据用于训练模型,不利于脑-机接口技术的推广。为减少训练数据量,提出了一种基于多码元时分编码的SSVEP-BCI少训练检测算法,利用相同码元刺激在不同码字刺激之间的可复用性,在保证较高的信号识别准确率和信息传输速率的前提下,以少量脑电训练数据即可识别出大量备选目标,有望提升SSVEP-BCI的实际应用价值。Steady-state visual-evoked potential(SSVEP) brain computer interface(BCI) is one of the three mainstream paradigms in the field of non-invasive brain-computer interface research. The traditional SSVEP-BCI training algorithm has to collect a large number of electroencephalogram data before training, which greatly increases the cost of SSVEP-BCI. To reduce the training time of SSVEP-BCI, we propose a less-training detection algorithm based on multi-symbol time-division coding. This algorithm takes advantage of the transferable symbol response between code responses and uses a small amount of data to identify a large number of targets with high recognition accuracy and information transmission rate, which can hopefully enhance the overall performance of SSVEP-BCI.

关 键 词：稳态视觉诱发电位 脑-机接口 多码元 时分编码 少训练 

分 类 号：TP335[自动化与计算机技术—计算机系统结构]