便携式听觉脑—机接口研究与实现
发布时间：2019-06-15 17:11
【摘要】：脑-机接口技术是一种不依赖于外周神经和肌肉系统的人机交互技术。近年来，推动该技术的实用化已经成为研究的热点。该技术主要应用在临床和消费电子领域，便携性是其走向实用化的重要指标。听觉脑-机接口系统可以使用耳机呈现听觉刺激，适于搭建便携性能更好的脑-机接口系统。本论文使用听觉空间注意和主动认知脑-机接口范式，使用无线脑电放大器和智能手机，设计并实现了一个高度便携的脑-机接口系统。 为了论证基于听觉空间注意和主动认知脑-机接口范式具备搭建便携式在线系统的潜力，本论文系统地比较了引入听觉空间注意前后，，听觉主动认知脑-机接口系统的性能。本论文使用基于头相关传递函数的虚拟声场技术生成具有空间属性的听觉刺激，处理后的听觉刺激可通过立体声耳机呈现。本论文从事件相关电位时空模式，平均事件相关电位波形，单次事件相关电位波形和目标识别性能四个方面对空间范式和非空间范式进行了对比分析。分析表明，“空间范式”的性能优于“非空间范式”。为了增强系统的便携性，本论文在目标识别性能分析时只使用了两个固定导联。 基于上述研究，本论文选用了无线脑电放大器和智能手机作为便携式听觉脑-机接口系统的脑电放大模块和信号处理模块。本论文根据范式要求完成了系统软件的设计与实现，测试了系统的性能并使用该系统进行了在线实验。使用事件相关电位面积法对来自Cz和Pz两个导联的脑电信号进行分类，5位受试者的平均在线五分类正确率为77±4%。系统还提供了一个单键拨号应用，实验证明用户完全可以使用该系统实现在线听觉脑电拨号。 本文所实现系统大大提高了听觉脑-机接口系统便携性，同时保持了较高的性能，对于面向残疾人的临床使用和面向健康人的人机交互等具有广阔的应用前景。
[Abstract]:Brain-computer interface technology is a kind of human-computer interaction technology which does not depend on peripheral nerve and muscle system. In recent years, promoting the practicality of this technology has become a hot research topic. This technology is mainly used in clinical and consumer electronics, and portability is an important index for its practicality. Auditory brain-computer interface system can use headphones to present auditory stimulation, which is suitable for building a better portable brain-computer interface system. In this paper, a highly portable brain-computer interface system is designed and implemented by using auditory spatial attention and active cognitive brain-computer interface paradigm, wireless EEG amplifier and smartphone. In order to demonstrate the potential of building portable online system based on auditory spatial attention and active cognitive brain-computer interface paradigm, this paper systematically compares the performance of auditory active cognitive brain-computer interface system before and after the introduction of auditory spatial attention. In this paper, the virtual sound field technique based on head correlation transfer function is used to generate auditory stimuli with spatial attributes, and the processed auditory stimuli can be presented through stereo headphones. In this paper, the spatial paradigm and non-spatial paradigm are compared and analyzed from four aspects: temporal and spatial pattern of event-related potential, average event-related potential waveform, single event-related potential waveform and target recognition performance. The analysis shows that the performance of "spatial paradigm" is better than that of "non-spatial paradigm". In order to enhance the portability of the system, only two fixed leads are used in the performance analysis of target recognition. Based on the above research, this paper selects wireless EEG amplifier and smart phone as EEG amplification module and signal processing module of portable auditory brain-computer interface system. In this paper, according to the requirements of paradigm, the design and implementation of the system software are completed, the performance of the system is tested and the online experiment is carried out by using the system. Event-related potential area method was used to classify EEG signals from Cz and Pz leads. The average online five-classification accuracy of the five subjects was 77 卤4%. The system also provides a single-key dialing application, and the experimental results show that the system can be used to realize online auditory EEG dialing. The system realized in this paper greatly improves the portability of auditory brain-computer interface system and maintains high performance. It has a broad application prospect for clinical use for disabled people and human-computer interaction for healthy people.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318.0
本文编号：2500376
[Abstract]:Brain-computer interface technology is a kind of human-computer interaction technology which does not depend on peripheral nerve and muscle system. In recent years, promoting the practicality of this technology has become a hot research topic. This technology is mainly used in clinical and consumer electronics, and portability is an important index for its practicality. Auditory brain-computer interface system can use headphones to present auditory stimulation, which is suitable for building a better portable brain-computer interface system. In this paper, a highly portable brain-computer interface system is designed and implemented by using auditory spatial attention and active cognitive brain-computer interface paradigm, wireless EEG amplifier and smartphone. In order to demonstrate the potential of building portable online system based on auditory spatial attention and active cognitive brain-computer interface paradigm, this paper systematically compares the performance of auditory active cognitive brain-computer interface system before and after the introduction of auditory spatial attention. In this paper, the virtual sound field technique based on head correlation transfer function is used to generate auditory stimuli with spatial attributes, and the processed auditory stimuli can be presented through stereo headphones. In this paper, the spatial paradigm and non-spatial paradigm are compared and analyzed from four aspects: temporal and spatial pattern of event-related potential, average event-related potential waveform, single event-related potential waveform and target recognition performance. The analysis shows that the performance of "spatial paradigm" is better than that of "non-spatial paradigm". In order to enhance the portability of the system, only two fixed leads are used in the performance analysis of target recognition. Based on the above research, this paper selects wireless EEG amplifier and smart phone as EEG amplification module and signal processing module of portable auditory brain-computer interface system. In this paper, according to the requirements of paradigm, the design and implementation of the system software are completed, the performance of the system is tested and the online experiment is carried out by using the system. Event-related potential area method was used to classify EEG signals from Cz and Pz leads. The average online five-classification accuracy of the five subjects was 77 卤4%. The system also provides a single-key dialing application, and the experimental results show that the system can be used to realize online auditory EEG dialing. The system realized in this paper greatly improves the portability of auditory brain-computer interface system and maintains high performance. It has a broad application prospect for clinical use for disabled people and human-computer interaction for healthy people.
【学位授予单位】：清华大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：R318.0
【参考文献】
相关硕士学位论文 前2条
1 郭婧;基于听觉注意的认知脑—机接口研究[D];清华大学;2010年
2 刘涛;基于运动起始视觉响应的在线实用化脑—机接口[D];清华大学;2010年
本文编号：2500376
本文链接：https://www.wllwen.com/yixuelunwen/swyx/2500376.html
