用于神经电信号采集的模拟前端处理

发布时间：2018-06-15 21:33

本文选题：模拟放大器 + ADC　；参考：《北京理工大学》2015年硕士论文

【摘要】：神经科学和电子科学的学科交叉是目前的前沿课题，科学家们试图通过电子与信息的技术来重构神经网路，通过微电极将电子设备植入大脑，从而直接探测单独或者群体神经元细胞的电活动。对人类大脑和神经系统的研究是人类目前科技发展的一个重要突破方向，通过电子科学技术去发现其工作规律并通过电信号去激励甚至取代神经信号传输，为人类运动和感官功能的神经修复开辟了一条道路。然而，实现以上目标，一个基本前提是能够很准确、实时、高校的得到神经元细胞的电信号，尤其是安全可以保证的前提下。本文的研究目标是以植入式脑机接口神经信号采集和修复系统为应用背景，采用CMOS集成电路工艺技术实现神经信号模拟前端的信号采集、放大和模数转换，为后续数字处理做基础。尽管国际上，对于神经信号采集芯片的研究已经有一定的技术和历史，，且有众多成果。但真正用于人类日常生活的还基本没有。对于神经电信号的采集中的等效输入噪声、直流偏执抑制范围、信号的失真度等性能参数的优化研究仍然需要学者投入大量的研究精力，最终实现高效、有效、实时的神经信号采集。本文对植入式神经信号采集系统的神经电极接口模块和信号采集处理模块进行了重点阐述。且对于神经电路接口模块做了详尽的讨论，尤其是对神经信号及其特征、电极以及组织接口电学模型做了综述，为模拟前端设计提供理论支撑。此外，本文的核心内容就是介绍模拟前端的设计。具体分析和介绍了模拟低噪声放大器的设计要点，分别从结构级和器件级分析了放大器的噪声和功耗性能，给出了基于理论分析进行的折衷优化设计。介绍了模数转换器的结构选择，基于选择的结构进行了原理分析和行为级建模，根据行为级建模的扫描仿真结果进行了电路结构选取和设计，给出了模数转换器的前后仿真结果。最后，给出整体四通道模拟前端设计的流片测试结果，以及结果分析。
[Abstract]:The intersection of neuroscience and electronics is the frontier of the day, with scientists trying to reconstruct neural networks using electronic and information technology, and microelectrodes to implant electronic devices into the brain. This directly detects the electrical activity of individual or colony neuron cells. The study of human brain and nervous system is an important breakthrough direction in the development of human science and technology at present. Through electronic science and technology to discover its working rules and electrical signals to stimulate or even replace neural signal transmission. It opens the way for nerve repair of human motor and sensory functions. However, to achieve the above goal, one of the basic prerequisites is to be able to be very accurate, real-time, college to get neuronal cell electrical signals, especially under the premise that security can be guaranteed. The research goal of this paper is to realize the signal acquisition, amplification and analog-to-digital conversion of the neural signal analogue front end using CMOS integrated circuit technology based on the neural signal acquisition and repair system of the implanted brain-computer interface. To provide the basis for subsequent digital processing. Although the research of neural signal acquisition chip has a certain technology and history in the world, and has a lot of achievements. But the real use of human daily life is basically not. The optimization of the performance parameters such as equivalent input noise, DC paranoid suppression range, signal distortion and so on in the acquisition of neural signals still needs a lot of research effort from scholars to achieve high efficiency and effectiveness. Real-time neural signal acquisition. In this paper, the neural electrode interface module and signal acquisition and processing module of implanted neural signal acquisition system are described. The interface module of neural circuit is discussed in detail, especially the electrical model of neural signal and its characteristics, electrode and tissue interface, which provides theoretical support for the design of analog front-end. In addition, the core of this paper is to introduce the design of analog front-end. The design points of analog low noise amplifier are analyzed and introduced in detail. The noise and power performance of the amplifier are analyzed from the structure level and the device level, and the tradeoff optimization design based on the theoretical analysis is given. The structure selection of A / D converter is introduced. The principle analysis and behavior level modeling are carried out based on the selected structure, and the circuit structure is selected and designed according to the scan simulation results of behavioral modeling. The simulation results before and after the analog-to-digital converter are given. Finally, the test results of the whole four-channel simulation front end are given and the results are analyzed.
【学位授予单位】：北京理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN911.7

【共引文献】