A conventional subject-dependent (SD) brain-computer interface (BCI) requires a complete data-gathering, training, and calibration phase for each user before it can be used. In recent years, a number of subject-independent (SI) BCIs have been developed. However, there are many problems preventing them from being used in real-world BCI applications. A weaker performance compared to the subject-dependent (SD) approach, and a relatively large model requiring high computational power are the most important ones. Therefore, a potential real-world BCI would greatly benefit from a compact low-power subject-independent BCI framework, ready to be used immediately after the user puts it on. To move towards this goal, we propose a novel subject-independent BCI framework named CCSPNet (Convolutional Common Spatial Pattern Network) trained on the motor imagery (MI) paradigm of a large-scale electroencephalography (EEG) signals database consisting of 21600 trials for 54 subjects performing two-class hand-movement MI tasks. The proposed framework applies a wavelet kernel convolutional neural network (WKCNN) and a temporal convolutional neural network (TCNN) in order to represent and extract the diverse spectral features of EEG signals. The outputs of the convolutional layers go through a common spatial pattern (CSP) algorithm for spatial feature extraction. The number of CSP features is reduced by a dense neural network, and the final class label is determined by a linear discriminative analysis (LDA) classifier. The CCSPNet framework evaluation results show that it is possible to have a low-power compact BCI that achieves both SD and SI performance comparable to complex and computationally expensive.

翻译：常规的基于主题(SD)的大脑-计算机界面(BCI)要求每个用户在使用之前有一个完整的数据收集、培训和校准阶段。近年来,已经开发出一些基于主题(SI)的BCI(SI) BCI(BCI)系统,但是,有许多问题使它们无法在现实世界的BCI应用中使用。与基于主题(SD)的方法相比,性能较弱,而要求高计算力的模型相对较大。因此,潜在的真实世界BCI将大大受益于一个低功率、依赖BCI(BCI)的低能主题(BCI)框架,在用户进行特性分析后立即使用。为了实现这一目标,我们提议了一个名为CCCPNet(CVAL 通用空间模式网络)的新主题(CCPNet(CVI 通用空间光学模型) 模式(EEEEG) 信号数据库,为54个执行双级手动任务的人进行2600次测试。拟议的框架将一个波内层(SLI) 级的螺旋内径(WCNNS) 内径(OL) 内径(C) 内径(C) 直径) 网络的轨(C) 和直径(CLLLLLL) 显示常规) 常规(C(C) 常规) 常规) 和直线(C(C(C) 常规) 常规) 常规) 运行) 的运行(C) 向) 的系统(C) 的运行(C) 的运行) 的运行(C) 向) 的运行(CLDLDLD) 的功能(通过常规) 的功能(C) 的功能(C) 和直径(C) 向(C) 向(C) 向) 的) 的常规) 和直径(C) 流) 的运行(C) 的运行(C) 的运行) 向) 向) 和直径流) 的运行(C) 的运行(C) 的运行(C) 的) 向) 的功能(显示的) 平流) 的常规(C) 向(C) 的(C) 的(C) 的(C) 的(C)