Scholarlink

<p>Brain-computer interface (BCI) is a modern way of human–machine interaction. With the use of electroencephalograph (EEG) data, this connection allows the human brain to directly operate or control other devices. A variety of studies have so far attempted to establish EEG-based BCI program to enhance the lifestyle of people with disabilities including paralysis and brain stroke, etc. The proposed home appliance management system (using BCI) is intended to give participants with disabilities an additional communication channel to help them maintain their quality of life. The proposed work aims to develop a speller system for home environment control utilizing steady-state visual evoked potential (SSVEP). The proposed system is tested for participants without disability and participants with motor impairments. EEG signals from seven participants without disability and seven participants with motor impairments were recorded using a 14-channel portable acquisition (Emotive EPOC) system. Six frequencies in the lower range (i.e. 6.67, 7.5, 8.57, 10, 12, and 14&#xa0;Hz) were chosen as stimulation frequencies. The existing SSVEP-based method commonly utilizes detection algorithms like frequency-selective filter banks and canonical correlation analysis (CCA). However, detecting the target frequencies using SSVEP signals is still a major challenge because of the low signal-to-noise ratio (SNR). Therefore, in this article, a deep learning method of EEGNet with ensemble learning is proposed for frequency discrimination and SSVEP identification. The existing CCA and multiway CCA (MCCA) methods were also employed for comparing the performance of the proposed system. The experimental results validate that the proposed method performed better than other methods with respect to bits per second (BPC), mean detection time (MDT) metrics, and detection accuracy (DA). Furthermore, the detected SSVEP was translated into commands to control and operate several home appliances.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Home appliances control using SSVEP-based brain-machine interface and deep learning

  • 期刊论文

  • 发表日期：

    2025年6月17日

查看全文
  • Praveen Kumar Shukla,
  • Rahul Kumar Chaurasiya

摘要

Brain-computer interface (BCI) is a modern way of human–machine interaction. With the use of electroencephalograph (EEG) data, this connection allows the human brain to directly operate or control other devices. A variety of studies have so far attempted to establish EEG-based BCI program to enhance the lifestyle of people with disabilities including paralysis and brain stroke, etc. The proposed home appliance management system (using BCI) is intended to give participants with disabilities an additional communication channel to help them maintain their quality of life. The proposed work aims to develop a speller system for home environment control utilizing steady-state visual evoked potential (SSVEP). The proposed system is tested for participants without disability and participants with motor impairments. EEG signals from seven participants without disability and seven participants with motor impairments were recorded using a 14-channel portable acquisition (Emotive EPOC) system. Six frequencies in the lower range (i.e. 6.67, 7.5, 8.57, 10, 12, and 14 Hz) were chosen as stimulation frequencies. The existing SSVEP-based method commonly utilizes detection algorithms like frequency-selective filter banks and canonical correlation analysis (CCA). However, detecting the target frequencies using SSVEP signals is still a major challenge because of the low signal-to-noise ratio (SNR). Therefore, in this article, a deep learning method of EEGNet with ensemble learning is proposed for frequency discrimination and SSVEP identification. The existing CCA and multiway CCA (MCCA) methods were also employed for comparing the performance of the proposed system. The experimental results validate that the proposed method performed better than other methods with respect to bits per second (BPC), mean detection time (MDT) metrics, and detection accuracy (DA). Furthermore, the detected SSVEP was translated into commands to control and operate several home appliances.