COLORED NOISE PREDICTION BASED ON NEURAL NETWORK

A method for predicting colored noise by introducing prediction of nonlinear time series is presented. By adopting three kinds of neural networks prediction models, the colored noise prediction is studied through changing the filter bandwidth for stochastic noise and the sampling rate for colored noise. The results show that colored noise can be predicted. The prediction error decreases with the increasing of the sampling rate or the narrowing of the filter bandwidth. If the parameters are selected properly, the prediction precision can meet the requirement of engineering implementation. The results offer a new reference way for increasing the ability for detecting weak signal in signal processing system.     

低相噪锁相介质振荡器研究

针对高的相位噪声指标要求,对取样锁相介质振荡器进行了研究.通过相位噪声分析,明晰了采用介质振荡器与取样锁相技术降低相位噪声的机理,并分别对介质振荡器与锁相环路进行了设计.设计中,应用HFSS与ADS对介质振荡器进行了联合仿真,体现了计算机辅助设计的优势.最终研制出17 GHz锁相介质振荡器,测试结果为:输出功率13.1 dBm;杂波抑制>70 dB;谐波抑制>25 dB; 相位噪声为-105 dBc/Hz@1 kHz,-106 dBc/Hz@10 kHz,-111 dBc/Hz@100 kHz,-129 dBc/Hz@1 MHz.     

正态随机噪声环境下遗传算法的动态适应度评价

在大量实际应用问题进化优化的适应度评价都受到了噪声的干扰.本文研究了正态随机噪声对适应度评价的影响机理,对比分析了正态随机噪声环境下不同适应度评价方法的性能.本文提出了一种多次评价一次采样的动态适应度评价方法,该方法在种群更新过程中,重新评价上一代种群,可以降低伪优个体（劣个体）的生存周期,抑制随机噪声对优胜劣汰机制的影响.实验结果也表明总采样次数相同的情况下,新方法的性能优于一次评价一次采样和一次评价多次采样方法.     

大幅度CCD输出信号降噪的相关双取样电路

电荷耦合器件(CCD)的输出信号构成复杂,包含有典型的KTC、1/f等类型的噪声,需要进行专门处理后才能获得与入射光信号相对应的高信噪比信号。文章针对具有较大幅度的CCD输出信号,采用宽电压工作的独立运放满足幅度较大的信号处理要求;通过在同一个运算放大器上实现噪声保持及信号采样的形式,消除了不同通道增益差异对信号的影响,获得了较高线性度的信号处理效果;同时结合CCD驱动器的设计,获取相关双取样技术所需的采样及保持脉冲信号,增强了采样与CCD输出信号间的关联程度,从而进一步提高了相关双取样技术消除CCD噪声的效果。采用这种信号处理电路后,将原来噪声处理的水平从约22 mV提高到了约1 mV,并且在一种精密的位移测量系统中得到应用,最后就具体电路设计的难点及注意事项进行了阐述。     

噪声密度不敏感的随机采样椒盐噪声滤波算法

中值滤波系列算法在处理被不同密度椒盐噪声污染的细节图像和平坦图像时,降噪性能不一致。本文借鉴开关中值滤波和压缩感知的思想,提出了随机采样滤波算法去除椒盐噪声。算法以噪声检测为基础,将被椒盐噪声污染的图像分为疑似噪声像素和信号像素,随机采样仅对信号像素采样。然后,利用正交匹配追踪算法重构出被污染前的图像,替代了中值滤波对噪声像素的估计。由于随机采样滤波基于压缩感知理论,对稀疏信号的重构具有最少测量次数的条件,因此随机采样点的数量具有一定的浮动空间,表现为对噪声密度不敏感。以被不同噪声密度污染图像的纹理、平坦局部区域进行验证,实验表明,当噪声密度在一定范围内变化时,算法可以实现对噪声密度不敏感。在高密度噪声污染的情况下,相较于中值滤波系列算法,随机采样滤波算法具有更好的细节保留能力和滤波能力。对标准测试图像进行了全局滤波,不同噪声密度具有一致的滤波效果,与自适应滤波算法相比,随机采样滤波算法在处理包含密集边缘特征的区域时更具备优势。      

BOLD contrast and noise characteristics of densely sampled multi-echo fMRI data

Blood oxygenation level dependent (BOLD) contrast in functional magnetic resonance imaging (fMRI) can be enhanced using multi-echo imaging and postprocessing techniques that combine the echoes in weighted summation. Here, existing echo-weighting methods are reassessed in the context of an explicit physiological noise model, and a new method is introduced: weights that scale linearly with echo time. Additionally, a method using data-driven weights defined using principal component analysis (PCA) is included for comparison. Differences in BOLD contrast enhancement between methods were compared analytically where possible, and using Monte Carlo simulations for different noise conditions and different combinations of acquisition parameters. The comparisons were also validated through densely sampled (256-echo) multi-echo fMRI experimental data acquired at 1.5T and 3.0T. Results indicated that the contrast-to-noise ratio (CNR) of the studied weighting methods have a strong dependence on the physiological noise, echo spacing and the width of the sampling window. With low noise correlations between echoes, contrast gain for all weighting methods was shown to have a square root dependence on the echo sampling density, and in typical experimental noise conditions, increasing the sampling window beyond 3·T2* produced marginal additional benefit. Simulations and experiments also emphasized that noise correlations between echoes are likely the main factor limiting the potential CNR gains achievable by densely sampled multi-echo fMRI.     

Achievable information rates on digital subscriber loops: limitinginformation rates with crosstalk noise

The capacity and cutoff rates for channels with linear intersymbol interference, power dependent crosstalk noise, and additive white noise are examined, focusing on high speed digital subscriber line data transmission. The effects of varying the level of additive white noise, crosstalk coupling gain, sampling rate, and input power levels are studied in detail for a set of simulated two-wire local loops. A closed-form expression for the shell constrained Gaussian cutoff rate on the crosstalk limited channel is developed and related to the capacity, showing that the relationship between these two rates is the same as on a channel without crosstalk noise. The study also projects achievable rates on a digital subscriber line, inside and outside of a carrier serving area, with a sophisticated but realizable receiver     

Ka频段上变频模块设计

论述一种Ka频段上变频模块设计,对模块的杂散和相位噪声性能做了分析,根据分析结果,设计出合理的变频方案。滤波器采用Ansoft Designer和HFSS软件协同仿真,薄膜工艺制作,获得足够的杂散抑制度从而实现低杂散。应用取样锁相技术合成了相位噪声极低的27 GHz本振。对波导—微带过渡结构进行仿真,并给出仿真结果。从测试结果表明,模块设计实现了低杂散和低相位噪声。     

低采样率高分辨率压缩功率谱估计方法的仿真研究

低采样率的宽带功率谱估计在很多领域具有应用价值.采用压缩多核采样结构得到信号的压缩测量值, 然后建立测量值相关函数与信号相关函数之间的关系, 用最小二乘法实现相关函数估计, 最后实现功率谱的估计.该压缩采样方法的等效采样率为M/N·f_s, 可在没有任何对时域或频域稀疏性的假设条件下降低采样率.仿真分析表明, 该方法的系统噪声与加性噪声性能比周期图法略有降低, 但只要系统设计合理, 对于一定信噪比的信号, 系统噪声与加性噪声基本可以忽略, 并给出了对应的理论分析.估计分辨率与周期图法相比, 等效长度相同时略有提高; 由于本文方法降低了测量值的数目, 对于一定长度的数据来说, 估计分辨率得到了极大的提高.本文方法适用于低信噪比信号的低采样率高分辨率功率谱估计.     

基于取样平均的UWB雷达时域去噪原理及仿真

分析了取样平均器在无载波超宽带运动目标检测雷达中的应用原理，对一种取样平均器的工作过程作了推导，建立了取样平均器的数学模型。并进行了仿真，结果表明，基于取样平均的时域去噪方法能有效滤除噪声和高频干扰。     

Optimum sampling vectors for Wiener filter noise reduction

Sampling is a very important and basic technique for signal processing. In the case that noise is added to a signal in the sampling process, we may use a reconstruction and noise reduction filter such as the Wiener filter. The Wiener filter provides a restored signal of which the mean square error is minimized. However, the mean square error by the Wiener filter depends on the sampling vectors. We may have the freedom to construct sampling vectors. We provide optimum sampling vectors under the condition that the Wiener filter is used for noise reduction for two cases wherein the noise is added before/after sampling. The sampling vectors provided in this paper may not be practical since they are very complicated. However, the minimum mean square error, which we provide theoretically, can be used for evaluating other sampling vectors. We provide all proofs of the theorems and lemmas. Furthermore, by experimental results, we show their advantages     

Noise analysis of DC-AC random PWM schemes

Random pulsewidth modulation (RPWM) techniques for DC-AC power converters usually work well with high-sampling frequency. When RPWM schemes are implemented in digital signal processors (DSPs), the sampling frequency is limited by the speed of the processors. This paper analyzes the noise spectrum of various DSP-implemented RPWM techniques using a statistical approach. The relationship of the noise components and the sampling frequency for both the standard RPWM and the weighted RPWM methods is generalized. The dependency of the noise characteristics on various factors is calculated theoretically and verified experimentally. The generalized noise theory for randomized PWM switching provides useful practical guidelines to the choice of sampling frequency     

Accurate on-chip interconnect evaluation: a time-domain technique

This paper describes an on-chip sampling and measurement technique for accurate (<15 ps) evaluation of interconnect delays and coupled noise. We have used this nonintrusive time-domain technique to extract in situ driver/receiver waveforms, propagation delays, and coupled noise in 120 interconnect structures. The effects studied include multiple AC returns through active devices, gridded planes on adjacent layers, via impedances, variable driver impedances, and noise in bus structures. The results provide a comprehensive evaluation of interconnect delays and noise in a 1.8 V, 0.25 μm process     

On the reconstruction aspects of moment descriptors

The problem of reconstruction of an image from discrete and noisy data by the method of moments is examined. The set of orthogonal moments based on Legendre polynomials is employed. A general class of signal-dependent noise models is taken into account. An asymptotic expansion for the global reconstruction error is established. This reveals mutual relationships between a number of moments, the image smoothness, sampling rate, and noise model characteristics. The problem of an automatic (data-driven) section of an optimal number of moments is studied. This is accomplished with the help of cross-validation techniques     

A 2.4 GHz CMOS sub-sampling mixer with integrated filtering

A sub-sampling mixer that incorporates sampling switches and hold capacitors into the parallel resonant LC load of an LNA is proposed. The noise figure of the proposed sub-sampling mixer is lower than that of a standard sampling circuit because the proposed mixer has narrow-band gain and input noise filtering properties. A novel level-shifting clock buffer with fast rise and fall times to drive the mixer sampling switches is presented. The mixer was fabricated in a 0.18 /spl mu/m CMOS process and measured results are presented for an RF input frequency of 2.42 GHz and a sampling frequency of 100 MHz. With a measured noise figure of 21.8 dB, the proposed circuit shows improved performance compared to other published sub-sampling mixers.     

Polarity coincidence correlation detection of a weak noise source

One application of sample polarity coincidence correlation to the detection of a weak noise source in background noise is briefly described. Assuming an input SNR much less than one, and Gaussian input signals and noise with identical normalized power spectra, expressions for the output SNR are derived for the analog and the polarity coincidence correlator, with and without sampling. The loss in attainable SNR due to clipping and sampling is computed for three different input spectra, viz.; white noise which is passed through an RC low-pass filter, a single-tuned band-pass filter or a rectangular filter. The resulting loss is given in three diagrams, as a function of relative bandwidth of the input signal and sampling frequency. For broad-band input signals the loss is between10and1db, and between4and1db for narrow-band signals.     

FTIR光谱仪动镜速度和采样抖动的噪声分析

分析了傅里叶变换光谱仪由动镜速度和采样抖动的随机误差引起的光谱噪声,建立了基于噪声等效辐亮度差(NERD)和噪声等效温差(NETD)的噪声模型。分析表明,动镜速度的波动与抗混叠滤波器幅频响应、时间延迟的综合作用,以及采样触发的短时抖动,都会引起系统的噪声。通过MATLAB数值模拟,对动镜速度和采样抖动的噪声影响进行了仿真分析与模型计算,验证了理论模型,评估了实际仪器的噪声水平,为动镜控制、抗混叠滤波器的设计以及傅里叶变换光谱仪的噪声定量化计算和改善提供了依据。     

Parallel correlated double sampling technique for pipelined analogue-to-digital converters

A new correlated double sampling technique that avoids the additional thermal noise penalty is presented. The new technique employs a low- gain two-stage opamp with the second stage made up of multiple gain stages in parallel. The superior noise performance of the proposed technique to correlated double sampling is shown.     

A sampling algorithm for digitally controlled boost PFC converters

Digital control of a boost power factor correction (PFC) converter requires sampling of the input current. As the input current contains a considerable amount of switching ripple and high frequency switching noise, the choice of the sampling instant is very important. To avoid aliasing without employing a (very) high sampling frequency, the sampling is synchronized with the pulse width modulation (PWM). Sampling algorithms employing this technique successfully reject the input current ripple but are not immune to the high frequency switching noise present on all sampled signals. Therefore, a new sampling algorithm, called alternating-edge-sampling and intended for center-based or symmetric PWM, is deduced with as most important features: switching noise immunity, straightforwardness, accurate measurement of the averaged input current and the need for only few processor cycles. The operating principle, design issues and a theoretical study of the input current error induced by the sampling algorithm due to sampling instant timing errors are derived. All theoretical results are validated experimentally for a digitally controlled boost PFC converter switching at 50 kHz.     

Optical transient waveform monitoring based on cross-gain modulation in semiconductor optical amplifier

An all-optical transient waveform equivalent time sampling system based on cross-gain modulation(XGM) in semiconductor optical amplifier(SOA) is presented.A noisy SOA dynamic model and PIN equivalent circuit function are employed for system evaluating.The results show this SOA-XGM sampler with subtracting postprocessing can achieve picosecond sampling window.The shape of sampling window can be adjusted by SOA bias current and amplitude of control pulse.Compared with amplified spontaneous emission(ASE) noise of SOA and thermal noise in PIN,the jitter of sampling control pulse causes most sampling error.Simulations show that this SOA-XGM sampler can be a capable candidate for transient waveform sampling.