The wavelet transform based processing means for 1/f noise in millimeter wave focal plane array

After briefly introducing the characteristics of 1/f noise in millimeter wave focal plane array detectors, the paper analyses the relation of wavelet transform and 1/f noise in detail, suggests the fashion of decorrelating 1/f noise using the wavelet transform and deduces the relative expressions. The results of computer simulation show good effectiveness.     

Subthreshold 1/f noise measurements in MOS transistors aimed at optimizing focal plane array signal processing

In focal plane signal processing applications it is imperative to minimize noise in the associated input circuit. The latter usually consists of MOS transistor preamplifiers and multiplexers. This paper concentrates on the 1/f noise generation in NMOS input signal processing circuits. A 1/f noise model for the subthreshold region is derived and correlated with 1/f noise measurements of NMOS transistors. This model is unique in that slow and fast surface states can be modeled separately. Typical values for slow and fast surface state densities from measured data and model correlation were found to be N _{ss s}=1×10¹¹ cm^-2 eV^-1 and N _{ss f}=3×10¹⁰ cm^-2 eV^-1, respectively. The order of magnitude of these results correlate well with published data [1–3]. The importance of these results is that process changes to minimize the 1/f noise can be monitored and the required minimization of 1/f noise for focal plane signal processing circuits can be achieved.     

Detection of known signals in 1/f fractal noise

Based on the whitening property of wavelet transformation for 1/f noise, this paper addresses the problem of detecting deterministic signals in the presence of 1/f fractal noise. The transfer function of whitening filter is provided as well as the condition for whitening. The receiver structure based on Karhunen-Loeve expansion and the decision rule are also given. Finally performance of the detector is analyzed.     

Impact of gate oxide nitridation process on 1/f noise in 0.18 μm CMOS

The aim of this paper is to study the impact of the nitridation techniques on the 1/f noise performances of dual gate 0.18 μm CMOS transistors. Nitrogen is often introduced to prevent boron penetration in ultrathin oxides especially when BF₂ is used for the PMOS junction implantation, but as a result the MOS transistor exhibits higher 1/f noise because of the increased fixed trap density. We show how the nitridation process can be improved in terms of 1/f noise characteristics, in a fully integrated technology. Projections of the 1/f noise behaviour for different technologies are also shown, to emphasise how the 1/f noise becomes an issue when other downscaling properties are considered for analog/RF CMOS applications.     

毫米波焦平面阵列的1/f噪声的小波处理方法

文中在概要介绍毫米波焦平面阵列成象探测器中存在的1/f 噪声的特性的基础上,深入分析了小波变换与1/f 噪声的关系,提出用小波分解的方法对毫米波焦平面探测器的1/f 噪声进行去相关处理,推导了相应的表达式。仿真结果表明可得到满意的效果。     

1/f Noise separated from white noise with wavelet denoising

The low-frequency noise in the devices often appears to be a superposition of 1/f noise and white noise, which are induced by different physical mechanisms or microscopic defects. The separation of the two noises is necessary especially in the characterization of quality and reliability by means of noise measurement. The traditional separation method is usually used in frequency domain with limited frequency range, and the estimated value of exponent index γ for 1/f^γ noise has considerable error when the ratio of 1/f noise and white noise is smaller than 20 dB. In this paper, based on wavelet denoising theory, a new method is proposed to separate 1/f^γ noise from white noise in time domain and can be used to estimate γ values with much better accuracy even if the ratio of 1/f noise to white noise in low enough.     

A wavelet analysis of 1/f and white noise in microwave transistors

The results of a computer-controlled experiment devoted to on-line measurements of the wavelet transform (WT) of low-frequency electronic noise voltage in bipolar microwave transistors are presented. We measure the average recurrence time of fluctuations of the WT coefficients occurring selectively in limited time and at frequency intervals. We detect two quite different behaviors in such recurrence times in the 1/f and white noise regimes.     

Voltage step-up filters not using transformers

A network of resistors and capacitors is to serve as a filter for an arbitrary signal voltagef(t). It consists of twin parts connected in parallel across the input terminals. One part is a low pass ladder filter; the rungs of the ladder are capacitors, and the other branches are resistors. The twin part is connected to the input terminals in theopposite direction. The output signal is the voltageg(t) measured between the end nodes of the ladders. Kirchhoff's current law gives a differential equation relatingf and (g−f). IfF is the Fourier transform off, then the transform of the differential equation is an algebraic equation relating the transformsF and (G−F). This equation shows that ⋎G/F⋎>1. Then by Parseval's theorem the norm off is greater than the norm ofg. Thus there is a stepup; the norm of the output always exceeds the norm of the input. Such a filter is here termed anRCformer. This paper is dedicated to Abraham Charnes, mathematician (1917–1992)     

基于多尺度Wiener滤波器的分形噪声滤波

针对淹没在1/f噪声中的有用信号恢复问题,本文提出了一套基于双正交小波变换与Wiener滤波的多尺度滤波算法,并设计出多尺度Wiener滤波器.首先,利用双正交小波变换将带有1/f噪声的信号分解成多尺度的子带信号,通过小波变换对1/f噪声的白化作用,消除了1/f噪声的非平稳性、自相似性和长程相关性.其次,在小波域内,利用Wiener滤波,实现了噪声和有用信号的分离,估计出了各子带中的有用信号.最后,利用双正交小波的精确重构性,较好地恢复出淹没在1/f噪声中的有用信号.仿真实验表明,该滤波器能有效的抑制分形噪声,显著地提高信噪比.     

小波变换在心电信号处理中的应用

在心脏病诊断过程中,心电信号的检测是重要的环节,然而心电信号的噪声很强,为了能够较好地滤除信号中的噪声,对信号的特点进行准确标定,利用基于小波变换的阈值去噪算法和基于小波的模极大值-极小值的算法进行心电信号的处理.采用MIT/BIH中的数据进行仿真调试验证,实验结果表明,被引入的几种噪声能被很好地去除,而且心电信号能较完整地保留下来,特征点能被准确地检测到,从而提高了诊断心脏等疾病的诊断效率.     

1/f noise as a reliability indicator for subsurface Zener diodes

It is shown from accelerated life tests and noise measurements that the degradation in reference voltage for subsurface Zener diodes is strongly correlated with 1/f noise in the devices. The larger the initial 1/f noise of a diode is, the earlier its degradation occurs. Compared with dc parameters, 1/f noise is more sensitive to the slight change in the structure of the devices subjected to operation or test stresses. In the mechanism analysis, the degradation and the 1/f noise are attributed to similar physical origin, and both are related to dislocations in the space-charge region of p–n junction. Based on the results, a 1/f noise screening approach is proposed for the high reliability application of the devices.     

傅里叶变换与小波变换在信号去噪中的应用

对于高频信号和高频噪声干扰相混叠的信号,采用小波变换去除噪声可以避免用傅里叶变换去噪带来的信号折损。对于噪声频率固定的平稳信号,在对信号进行傅里叶变换后使用滤波器滤除噪声。对高频含噪信号则采用正交小波函数sym4对信号分解到第4层,利用极大极小值原则选择合适的阈值进行软阈值处理,最后利用处理后的小波系数进行重构。实验结果表明,对于高频含噪信号傅里叶去噪会出现严重的信号丢失现象,使用极大极小值原则选择阈值进行小波去噪可以有效地保留高频部分的有用信号。     

小波变换在心音奇异信号处理中的应用

针对如何剔除心音信号奇异点处的噪音信号的问题,该文利用小波变换方法,通过PhysioBank数据库中的相关数据,运用Matlab对含有噪音信号的心音奇异信号进行处理。其结果表明,该方法可在保留奇异信号的前提下,有效滤除噪音信号。说明利用小波变换技术分析与处理心音奇异信号是一种新的有效和实用的方法。     

一种混合噪声图像的自适应滤除方法

基于小波变换的阈值去噪方法仅适用于去除高斯白噪声,对于脉冲噪声得不到好的去噪效果,正交小波变换由于缺乏平移不变性,在去噪过程中会产生人为的振荡现象,使图像边缘失真,甚至图像模糊,提出了基于平稳小波域自适应阈值算法同中值滤波相结合的去噪方法,该方法能有效地滤除图像中的高斯白噪声和脉冲噪声组成的混合噪声,并验证了该方法的有效性。     

一种基于自适应滤波的语音降噪方法研究

分析和研究自适应滤波和小波变换法的原理及方法,提出了一种新的综合使用自适应滤波和小波变换法的语音降噪方法。该方法首先用仿生小波变换法对带噪声的语音信号进行小波分解,将小渡变换法分离出来的噪声信号作为自适应滤波器的输入。最后选择用最小均方误差（LMS）的自适应算法对带噪声语音信号进行降噪处理,实现了信噪分离,去除语音信号中的噪声信号。实验结果表明,该方法对语音信号有较为明显的降噪效果。     

On the intrinsic origin of 1/f noise

The problem of the intrinsic origin of 1/f noise is considered. Currents and signals consisting of a sequence of pulses are analyzed. It is shown that the intrinsic origin of 1/f noise is a random walk of the average time between subsequent pulses of the pulse sequence, or the interevent time. This results in the long-memory process for the pulse occurrence time and in 1/f type power spectrum of the signal.     

Numerical solutions for orthogonal wavelet filters by Newton method

The wavelet transform has recently generated much interest in applied mathematics, signal processing and image coding. Mallat (1989) used the concept of the function space as a bridge to link the wavelet transform and multiresolution analysis. Daubechies (1990) added regularity conditions to find 2N, 2N10, tap coefficients for orthogonal wavelet filters. Owing to the difficulty of finding their closed solutions for large N a numerical method called the Newton method is proposed. We constructed the orthogonal wavelet filter with 2N-tap coefficients by N linear equations and N nonlinear equations. The 2N-tap, 2N10, coefficients we found are very consistent with those of Daubechies. Also, the method can be used to find the orthogonal wavelet filter with N-tap coefficients for N>10.     

基于中值滤波和小波变换的红外图像去噪方法

针对红外图像对比度差、信噪比低、红外目标易被噪声干扰的缺点,提出一种自适应中值滤波和小波变换相结合的去噪方法.首先对图像进行小波变换,得到各频带子图像,对不同子图像采用不同的滤波模版进行自适应中值滤波,最后对图像进行小波重构,实现图像的去噪.将该方法与目前常用的去噪方法进行对比实验,结果表明,该算法简单有效,突出目标图像细节,可获得更好的图像视觉效果.     

Hooge parameter of InGaAs bulk material and InGaAs 2DEG quantum well structures based on InP substrates

The 1/f noise of various InGaAs layers lattice matched to InP is investigated systematically at room temperature. To elucidate the origin of the 1/f noise in this type of III–V compound semiconductor, thick n-type doped InGaAs layers, typical InP based InAlAs/InGaAs or InP/InGaAs heterostrucuture field-effect transistor (HFET) structures, respectively, as well as InP based InAlAs/InGaAs quantum well structures with doped InGaAs two-dimensional electron gas (2DEG) channel are analysed. From the experiments it is found that mobility fluctuation is the only origin for the 1/f noise observed both in InGaAs bulk material and in 2DEG heterostructures of high quality. A Hooge parameter attributed to phonon scattering α_Hphon in the bulk material is found to be about 7×10⁻⁶ and agrees with those obtained from HFET structures with the highest mobilities (α_H≈1.5×10⁻⁵). Furthermore, the Hooge parameter of 2DEG structures strongly depends on the channel design and on the doping concentration in the n-type doped 2DEG channels.