Individual interface states and their implications for low-frequency noise in MOSFETs

We show that the wide distribution of time constants required to explain 1/ƒ noise in MOSFETs arises as a natural consequence of the multi-phonon model of carrier trapping into individual Si-SiO₂ interface states. A new class of random telegraph signal found in the drain current of small-area silicon MOSFETs is described. These signals are a result of defect metastability and are shown to be a source of non-Gaussian noise.     

Analysis of noise properties of an optocoupler device

In the paper, localization of a source of random telegraph signal noise (RTS noise) in optocoupler devices of CNY 17 type was defined. The equivalent noise circuit in low frequency noise for these types of optocouplers was proposed.     

20MPixel/s高速红外焦平面读出电路设计

针对中大规模红外焦平面对高速读出的需求,研究并设计了一款20MPixel/s红外焦平面高速读出电路。读出电路单元电路由电容负反馈运放输入级、相关双采样、源随输出级电路组成,总线输出级采用基于低功耗推挽运放的跟随器结构。研究了输出级运放像元信号建立时间和负载电容的关系,给出了20 MPixel/s高速读出的负载电容适用范围。采用0.5μm Mixed Signal CMOS工艺研制了一款红外焦平面高速读出电路芯片,和InGaAs光敏芯片耦合后实测读出速率达到20MPixel/s,像元信号之间最大上升时间为17ns。     

A 128 × 128 Pixel Complementary Metal Oxide Semiconductor Image Sensor with an Improved Pixel Architecture for Detecting Modulated Light Signals

A complementary metal oxide semiconductor (CMOS) image sensor for the detection of modulated light under background illumination has been developed. When an object is illuminated by a modulated light source under background illumination the sensor enables the object alone to be captured. This paper describes improvements in pixel architecture for reducing fixed pattern noise (FPN) and improving the sensitivity of the image sensor. The improved 128 × 128 pixel CMOS image sensor with a column parallel analog-to-digital converter (ADC) circuit was fabricated using 0.35-mm CMOS technology. The resulting captured images are shown and the properties of improved pixel architecture are described. The image sensor has FPN of 1/28 that of the previous image sensor and an improved pixel architecture comprising a common in-pixel amp and a correlated double sampling (CDS) circuit. The use of a split photogate increases the sensitivity of the image sensor to 1.3 times that of the previous image sensor.     

天文观测相机低噪声CCD视频处理电路的分析与设计

在讨论天文观测相机低噪声视频处理电路的基础上,对前置放大、差分放大、可编程增益放大、相关双采样及16bit模数转换等关键组成部分进行了分析,给出各组成部分的设计方法。对所设计的低噪声视频处理电路进行了PSPICE仿真和实验测试,仿真和实验结果表明,该视频处理电路在本身引入噪声较小的同时,有效地抑制了CCD的复位噪声和1/f等低频噪声,使天文观测相机的整机噪声<15e-,达到了指标要求。     

Theoretical analysis and experimental application of CDS CMOS integrated circuit for uncooled infrared focal plane arrays

In this paper, a typical correlated double sampling (CDS) complementary metal oxide semiconductor (CMOS) circuit for uncooled infrared focal plane array (IRFPA) is theoretically analyzed, the key factor of CDS CMOS integrated circuit is pointed out, and a new CDS integrated circuit which is high correlative for low-frequency noise is applied in an experimental readout chip for uncooled IRFPA. Theoretical analysis indicates that the sample transfer function of a noise source acted on by CDS processing is related to noise frequency and sampling time interval and the key factor of CDS circuit for reducing or eliminating noise in readout integrated circuit is the sampling time interval. The experimental readout chip with high noise-correlative CDS integrated circuit is fabricated to verify the theoretical analysis, which can be applied to uncooled IRFPAs.     

A percolation study of RTS noise in deep sub-micron MOSFET by Monte Carlo simulation

Based on percolation theory and random telegraph signal (RTS) noise generation mechanism, a numerical model for RTS in deep submicron metal-oxide-semiconductor field-effect transistor (MOSFET) was presented, with which the dependence of Tc/Te (where Tc=capture time, Te=emission period ) on energy levels and trap depth with respect to the interface of traps can be simulated. Compared with experimental results, the simulated ones showed a good qualitative agreement.     

Modulation bandwidth and noise limit of photoconductive gates

The modulation bandwidth and noise limit of a photoconductive sampling gate are studied by reducing the parasitic capacitance and leakage current of the sampling circuit using an integrated junction field-effect transistor (JFET) source follower. The modulation bandwidth of the photoconductive sampling gate is limited by the external parasitic capacitance, and its efficiency is found to saturate at a laser gating power of about 1 mW. It is determined that the noise of the photoconductive sampling gate is dominated by the photovoltaic current due to the gating laser amplitude fluctuation. A minimum noise level of 4 nV Hz^–1/2 has been measured, and an enhancement in signal-to-noise ratio by a factor of >45 has been achieved after the integration of the source follower with the photoconductive sampling gate. The JFET source follower serves to increase the modulation bandwidth of the photoconductive sampling gate by about 15 times and buffer the charge of the measured signal using its extremely high gate input impedance. The performance of the photoconductive sampling gate in regard to invasiveness and gating efficiency has been optimized, while a picosecond temporal resolution has been maintained and the signal-to-noise performance has been enhanced using a gating laser power as low as 10 W.     

Microwave irradiation effects on random telegraph signal in a MOSFET

We report on the change of the characteristic times of the random telegraph signal (RTS) in a MOSFET operated under microwave irradiation up to 40 GHz as the microwave field power is raised. The effect is explained by considering the time dependency of the transition probabilities due to a harmonic voltage generated by the microwave field that couples with the wires connecting the MOSFET. The RTS experimental data are in agreement with the prediction obtained with our model.     

Random telegraph signal noise simulation of decanano MOSFETs subject to atomic scale structure variation

As MOSFETs shrink into the decanano regime it is predicted that random telegraph signals (RTS), resulting from trapping events in defect states near the Si/SiO₂ interface, will significantly affect analogue and digital circuit performance. At these same scales, intrinsic parameter fluctuations introduced by atomic differences between devices will also be significant. In this work, a methodology based on 3D simulation is developed which can correctly model RTS noise in the time and frequency domain in the presence of random discrete dopants. The approach is illustrated with results obtained for 30×30  nm devices. We find that atomicity can significantly increase RTS magnitude in devices with particular doping configurations, and ensemble average RTS effects vary markedly from those predicted on an assumption of continuous doping.     

面阵CCD信号采集系统的噪声抑制

分析了CCD输出噪声及其一般抑制方法,提出了一种基于面阵CCD信号采集系统的噪声抑制方法。设计了CCD信号采集系统的噪声抑制电路和处理电路,应用于CCD442A型面阵CCD;并使用积分球对采集系统进行辐射定标,计算得到系统的信噪比。仿真和辐射定标实验表明,该面阵CCD信号采集系统具有相关双采样和暗电平校正功能,抑制了CCD输出信号的复位噪声和暗电流噪声;在中等照度条件下,系统信噪比达到40dB。     

32 × 32 SOI CMOS image sensor with pinned photodiode on handle wafer

We have fabricated a 32 × 32 silicon on insulator (SOI) complementary metal oxide semiconductor (CMOS) active pixel image sensor with a pinned photodiode on a handle wafer. The structure of one pixel is a four-transistor type active pixel sensor (APS) which consists of a reset and a source follower transistor on a seed wafer, and is comprised of a photodiode, a transfer gate, and a floating diffusion on the handle wafer. The photodiode could be optimized for better quantum efficiency and low dark currents because its process on the handle wafer is independent of that of transistors on a seed wafer. Most of the wavelengths are absorbed within the visible range, because the optimized photodiode is located on the handle wafer. The image has been captured by the fabricated 32 × 32 SOI CMOS image sensor with array pixels, vertical scanner, horizontal scanner, and delta-difference sampling circuit.     

可见光CCD的光致过饱和现象

 报道了可见光CCD的过饱和状态——在强光辐照下,CCD输出图像中的白色饱和区域内出现黑色盲区。介绍了相机成像系统的信号处理过程和CCD的4种基本输出波形,给出了这4种波形的信号在经过相关双采样电路处理之后的结果和对应的CCD的4种基本输出图像效果,即无光照的黑区、弱光照的灰区、光照饱和的白区和强光致过饱和的黑色盲区（即过饱和状态）。过饱和状态CCD的输出信号波形中,复位电平发生改变而与处于饱和的数据电平持平,使该信号在经过相关双采样电路处理后变为零输出,造成CCD输出图像中的白色饱和光斑中间出现黑色盲区。     

非制冷红外探测器读出电路的非均匀性研究

对于长线列的非制冷红外探测器组件, 不同探测元之间的非均匀性是衡量电路设计的关键指标. 为了实现长线列非制冷红外探测器的高性能读出, 本文设计了一种基于电流镜方式的非制冷红外探测器160线列读出电路, 电路由电流镜输入模块、电容负反馈互导放大器模块及相关双采样输出模块组成. 电路采用0.5 μm工艺制作完成. 通过合理设置电路中MOS管的参数和布局电流镜版图, 电路的非均匀性有了明显地改善. 通过测试, 电路的非均匀性小于1%, 器件总功耗约为100 mW, 并具有良好的低噪声特性, 输出噪声小于1 mV, 输出摆幅大于2 V. 该电路与160线列非制冷红外探测器互连后, 能较好地完成红外信号的读出, 在积分时间为20 μups的情况下, 器件的响应为0.294 mV/Ω, 整体性能良好. 该电路的研制对超长线列的非制冷红外冷探测器读出电路研制奠定了重要的技术基础.     

基于辐照前1/f噪声的金属-氧化物-半导体场效应晶体管辐照退化模型

基于金属-氧化物-半导体场效应晶体管(MOSFET)噪声的载流子数涨落和迁移率涨落理论,建立了MOSFET辐照前1/f噪声参量与辐照后分别由氧化层陷阱和界面陷阱诱使阈值电压漂移之间的定量数学模型,并通过实验予以验证.研究结果表明,辐照诱生的氧化层陷阱通过俘获和发射过程与沟道交换载流子,在引起载流子数涨落的同时也通过库仑散射导致沟道迁移率的涨落,因此辐照前的1/f噪声幅值正比于辐照诱生的氧化层陷阱数.利用该模型对MOSFET辐照前1/f噪声与辐照退化的相关性从理论上 关键词： f噪声')" href="#">1/f噪声 辐照 金属-氧化物-半导体场效应晶体管 陷阱     

基于像素置乱技术的多重双随机相位加密法

提出一种基于像素置乱技术的多重双随机相位加密法,对该加密法中像素置乱操作的原理进行了阐述,并且提出在光学上实现像素置乱操作和解置乱操作的途径.在计算机上模拟实现了该加密法,并且得到很好的加密解密结果.仿真结果证实仅用部分加密图像来解密也能够得到原图像,并且得到随着待解密的加密图像像素的增加,解密图像的信号能量、噪声以及信噪比的变化曲线.最后分析比较了该加密法与双随机相位加密法,得到该加密法与双随机相位加密法相比具有更高的保密性,而且解密图像的信噪比也不会因为引入像素置乱操作而降低.     

0.5 μm工艺CMOS有源像素传感器的总剂量辐射效应

采用⁶⁰Co-γ射线对某国产0.5 μm CMOS N阱工艺CMOS有源像素传感器(APS)的整体电路和像素单元结构进行了电离总剂量辐射效应研究,重点考察了器件的饱和输出信号、像素单元输出信号、暗信号等参数的变化规律.随着辐射剂量的增大,饱和输出信号逐渐减小且与像素单元饱和输出信号变化基本一致;暗信号随总剂量的增大而显著增大.研究结果表明,0.5 μm工艺CMOS APS电离总剂量辐射效应引起参数退化的主要原因是光敏二极管周围的整个LOCOS(Local oxidation of silicon)隔离氧化层产生了大量的辐射感生电荷.     

基于随机共振原理的分段线性模型的理论分析与实验研究

提出了一种分段线性双稳态模型, 推导了模型的解析关系及其输出信噪比, 通过对该模型与连续双稳态模型的对比分析和仿真实验, 证明了该模型的优越性.该模型具有参数之间相互独立、易于调节的特点. 在对模型分析与数值仿真的基础上, 通过电路对强噪声背景下的微弱周期信号检测进行了实验研究. 结果表明分段线性随机共振模型能够有效实现对微弱周期信号的检测, 并能显著增强输出信噪比.     

Double image encryption using double pixel scrambling and random phase encoding

A novel double image encryption method is proposed by utilizing double pixel scrambling technique and random fractional Fourier domain encoding. One of the two original images is encoded into the phase of a complex signal after being scrambled by one matrix, and the other original image encoded into its amplitude after being scrambled by another matrix. The complex signal is then encrypted into stationary white noise by utilizing double random phase encoding in fractional Fourier domain. By applying the correct keys with fractional orders, the random phase masks and the pixel scrambling operation, the two original images can be retrieved without cross-talk. Numerical simulations have been done to prove the validity and the security of the proposed encryption method.     

Generation of burst noise by nonlinear autoregression

Nonlinear generalization of autoregressive sequence of first order is used as basis for construction of mathematical model of bistable burst noise. It is proved that the noise generated using such a model has Lorentzian spectrum after clipping the white noise component. In passing it is demonstrated that random telegraph signal is obtainable by simple iteration. The proposed model suggests that one should expect to observe temperature dependence of frequency of transitions between the two levels of burst noise waveform.