指纹识别传感器电路系统设计

发布时间：2018-06-29 15:43

本文选题：指纹识别 + 反馈电容式　；参考：《电子科技大学》2016年硕士论文

【摘要】：随着集成电路的发展和物联网的兴起,生物识别技术在个人认证中的应用变得越来越重要、越来越广泛了。其中指纹识别是众多生物识别技术中发展最为成熟的技术,应用也是最广泛的现在。超声波式指纹识别还处在研发阶段,而射频式只有少数厂家掌握,所以市场上的半导体指纹传感器主要还是电容式的。电容式指纹传感器兼具体积小、灵敏度高、仿假手指等优点。反馈电容式指纹识别相对于普通的电容式具有更高的灵敏度的优点。本文主要是对反馈电容式指纹识别的模拟电路部分进行了设计。该传感器的数字输出为8位,将手指与芯片的距离量化为256个灰度值。文章首先对指纹识别的种类和发展历史作了简要的介绍。紧接着,详细地介绍了反馈电容式指纹识别的工作原理。在对电路模块进行设计前,我们必须有一个问题需要解决。由于在电路模型中,手指和芯片形成的电容中存在边缘电场电容,而该电容很难通过手工计算获得。所以文章先利用ANSYS Maxwell软件对手指芯片距离和电容的关系进行了模拟。传感器模拟部分主要分为采集单元、可变增益放大器、8位Pipeline ADC、环形振荡器和其他一些辅助电路。考虑到面积原因,采集单元没有采用运算放大器,而是运用共源共栅放大器结构,增益在70dB以上。可变增益放大器采用反馈电阻式,通过改变接入电路的电阻大小,从而改变电路增益,可变增益的范围为1到2.4,步长为0.2。8位ADC采用Pipeline结构,一共分为两级,第一级4位,第二级5位,多的一位做数字校正用,采样频率为1MHz。环形振荡器主要为系统提供时钟,有三种频率提供选择,分别为12MHz、16MHz、22MHz。其他的一些辅助电路包括跟随器、寄存器和选择电路等等。本文采用0.5?m工艺,利用Spectre对电路进行了仿真。通过整体仿真获得手指与芯片距离的关系和灰度值的关系图。最后经过流片测试验证,各个电路模块能正常工作,指纹图像清晰。
[Abstract]:With the development of integrated circuits and the rise of Internet of things, the application of biometrics in personal authentication has become more and more important and extensive. Fingerprint identification is the most mature technology in many biometric technologies, and its application is the most extensive now. Ultrasonic fingerprint identification is still in the research and development stage, but only a few manufacturers control the RF type, so the semiconductor fingerprint sensor in the market is mainly capacitive. Capacitive fingerprint sensor has the advantages of small size, high sensitivity and fake finger. Feedback capacitive fingerprint recognition has higher sensitivity than conventional capacitance. In this paper, the analog circuit of feedback capacitive fingerprint identification is designed. The digital output of the sensor is 8 bits and the distance between finger and chip is quantified to 256 gray values. First of all, this paper briefly introduces the types and development history of fingerprint recognition. Then, the working principle of feedback capacitive fingerprint identification is introduced in detail. Before we design the circuit module, we must have a problem to solve. In the circuit model, there is an edge electric field capacitance in the capacitance formed by finger and chip, which is difficult to be obtained by manual calculation. So the relationship between distance and capacitance of finger chip is simulated by ANSYS Maxwell software. The sensor simulation is mainly divided into acquisition unit, variable gain amplifier (VGA) and 8-bit pipeline ADCC, ring oscillator and some other auxiliary circuits. Considering the area reason, the acquisition unit does not use an operational amplifier, but uses a common-gate amplifier structure with a gain of more than 70dB. The variable gain amplifier adopts feedback resistor type, and changes the circuit gain by changing the resistance of the access circuit. The variable gain range is 1 to 2.4, the step size is 0.2.8 bits ADC adopts pipeline structure, which is divided into two levels, the first stage is 4 bits. The second level 5 bit, more than one digit for digital correction, sampling frequency of 1 MHz. The ring oscillator mainly provides the clock for the system, and there are three frequencies to choose from, which are 12MHz / 16MHz / 22MHz. Other auxiliary circuits include followers, registers, and selection circuits. In this paper, the circuit is simulated by Spectre using 0.5m process. The relationship between finger and chip distance and the relationship between gray value and finger are obtained by whole simulation. Finally, the circuit module works well and the fingerprint image is clear.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP212;TN402

【参考文献】