A Note on Dynamic Range of Quantizers

Based on the relationship between the inputand output correlation functions for a symmetric2N step optimum quantizer withGaussian noise inputs a computer programwas generated to obtain plots of the inputnumber of levels versus dynamic range of thequantizer. For it to be physically useful thedynamic range should be related to the distortionin the fundamental band and thedesired harmonic suppression. A definitionbased upon these considerations is suggestedand curves are plotted for two different fundamentaldistortion levels and three differentlevels of harmonic suppression. It is foundthat, for low levels of harmonic suppression(0-20 dB range), the dynamic range can beincreased by tolerating a higher level of fundamentaldistortion. In the medium range ofharmonic suppression (20-35 dB), any increasein the dynamic range due to higherfundamental distortion levels disappearsfor high level quantizers (typically 4 bitsor more). For still higher harmonic suppression(40 dB or higher), the dynamic range ofthe quantizer is independent of the acceptablefundamental distortion.     

具有量化器的网络控制系统稳定性分析

数字设备硬件性能的提高和网络控制系统的发展使得量化问题成为目前研究的热点。由于网络控制系统具有带宽约束的特点,数据的传输和处理是零传输延迟和无限高精度的这一假定不再合适,有必要对量化对于控制系统性能的影响进行研究。针对考虑量化器的状态反馈镇定系统的模型从非线性的角度研究了量化器对于控制对象状态动态特性的影响。进而推广到考虑量化器的周期通信的网络控制系统模型上。用Matlab／Simulink仿真算倒,验证了结论的有效性。     

Data Quantization for Narrowband Signal Detection

In automatic radar detection, digital integration of the envelope detector outputs is often used as a good approximation to the optimum. This requires quantizing the envelope detector outputs. In this paper, quantizer structures for narrowband signal detection are considered. Quantizer characteristics are derived to optimize performance as measured by the detector efficacy?an asymptotic performance measure. Asymptotic and finite sample performance results are presented. The results obtained are not limited in their application to Gaussian noise only, although this important case is given specific consideration.     

图C4k∪pn的优美性

证明了圈C4k是优美交错图，证明了路Pn是优美交错图。证明了由圈C4k与路Pn粘接而成的图C4k∪Pn是优美图，也是交错图，并给出了图C4k∪Pn的优美标号。     

Data quantization effects in CFAR signal detection

In this paper, we investigate data quantization effects in constant false alarm rate (CFAR) signal detection. Exponential distribution for the input data and uniform quantization are assumed for the CFAR detector analysis. Such assumptions are valid in the case of radar for a Swerling I target in Gaussian clutter plus noise and a receiver with analog square-law detection followed by analog-to-digital (A/D) conversion. False alarm and detection probabilities of the cell averaging (CA) and order statistic (OS) CFAR detectors operating on quantized observations are analytically determined. In homogeneous backgrounds with 15 dB clutter power fluctuations, we show analytically that a 12-bit uniform quantizer is sufficient to achieve false alarm rate invariance. Detector performance characteristics in nonhomogeneous backgrounds, due to regions of clutter power transitions and multiple interfering targets, are also presented and detailed comparisons are given     

Postfiltering in Digital Beamformers

Assuming a sinusoidal signal superimposed on a narrow-band Gaussian noise as the input to a receiving array, the output power and signal-to-noise ratio of a digital beamformer with postfiltering were formulated so that subsequent calculations could be made without an analysis in the frequency domain. The formulation utilized the quantizer functions previously given by the author and certain spectral power distribution factors originally attributed to Davenport but more rigorously derived and discussed in the present work. A numerical study based on this formulation for a DIMUS array in a correlated noise field reveals that except for certain rare circumstances, postfiltering generally improves the output SNR or array gain. It is demonstrated that the amount of postfiltering gain not only varies with array input SNR but also depends strongly upon the spacing-to-wavelength ratio, and its meaningful interpretation can only be made in conjunction with both the clipping and noise correlation losses. In particular, balancing postfiltering gain against the two losses suggests that receiving arrays with element spacings smaller than one-half of the operating wavelength may be used to the advantage of system design under certain conditions.     

Performance of a Mean Level Detector Processing M-Correlated Sweeps

The performance of a mean level detector processing M-correlated consecutive sweeps is derived. Performance when sweeps are independent can be obtained simply as a special case. The background noise is assumed stationary Gaussian and pulses are fluctuating according to the Swerling I model. Results are obtained for both finite and infinite reference noise samples. It is shown that for fixed M the relative improvement over the single hit case increases when the correlation between sweeps decreases and as the probability of false alarm is kept at lower rates.     

Detection and Time-of-Arrival Estimation of Received Pulses in Radar Beacon Systems

This paper is devoted to the analysis of the detection and time-of-arrival estimation process of the signalling pulses in a sweep time of a radar beacon system. A simplified model of the detector/estimator is worked out to evaluate pulse detection and false-alarm generation. Finally the performances of a previously proposed video-pulse quantizer are derived in the absence of interference, whose effects are examined in some critical conditions.     

Data-Adaptive Detection of a Weak Signal

Motivated by a form of the likelihood-ratio-tesf statistic for detection of a rank-one Gaussian signal in colored Gaussian noise, we apply our earlier technique for estimation of a low-rank signal to the problem of estimating and subtracting the waveform of a strong sinusoidal interference prior to detection of a weak sinusoidal signal. We consider the difficult case in which samples of data are taken over a short interval of time or space and the frequencies of the sinusoidal signal and sinusoidal interference are more closely spaced than the reciprocal of the extent of the aperture. The method can be applied to cases of nonsinusoidal and/or random signals and interference. The most important assumption is that when the samples of the interference are arranged in matrix form the matrix is approximately of low rank in the sense that, with high probability, the interference-only matrix can be well approximated by a matrix of low rank.     

Post Mixer Spectra of Periodic FM Altimeters with Area Target Returns

A periodic FM altimeter is postulated and its illuminated area target modeled as an array of elemental scatterers. A computational form is then developed expressing the desired post mixer spectrum as a sum of spectral components corresponding to the individual scatterer returns. Application of the model is made to the case of a triangular linear sweep FM altimeter and samples of the calculated post mixer spectrum presented, illustrating the effects of antenna beam geometry, target surface properties, and modulation parameters.     

A Quantitative Analysis of Sea Clutter Decorrelation with Frequency Agility

A brief statement of the sea clutter problem in surface-search radar operation illustrates the need for some form of signal-to-clutter enhancement. Post-detection integration used in the simpler radars is limited by the pulse-to-pulse correlation of the clutter. Analysis of the effect of changing frequency from pulse to pulse leads to an expression for the correlation between pulses in the sequence. Knowing this correlation, the reduction in the fluctuating clutter component produced by integration can be determined. This is described by an equivalent number of independent pulses, Nc. For the particular case of sinusoidal modulation of the transmitted frequency, N6 is computed. The critical dependecne of Nc upon the modulating frequency fm is illustrated by spectrum photographs. Choice of an optimum fm is discussed. The results of computations of N4 for optimum fm are presented as a family of normalized curves. These data permit the tradeoff of the radar parameters against their quantitative effect on radar performance.     

Reduction of Radar Tracking Errors with Frequency Agility

Frequency agility with random frequency in each pulse gives an improvement in radar angle tracking with a monopulse radar. With a conical-scan tracking radar, the glint error is reduced but fading error can be increased, and the net result must be studied in each case. A theory, usable for calculating angle tracking errors with a frequency agile radar, is given, and two examples showing the error reduction are presented. According to the theory, one part of the glint or fading spectra is ``smeared out' to half the pulse repetition frequency. Another part, the size of which depends on the degree of correlation between pulses, keeps the form of the original spectrum.     

Collaborative image compression and classification with multi-task learning for visual Internet of Things

Widespread deployment of the Internet of Things(Io T) has changed the way that network services are developed, deployed, and operated. Most onboard advanced Io T devices are equipped with visual sensors that form the so-called visual Io T. Typically, the sender would compress images, and then through the communication network, the receiver would decode images, and then analyze the images for applications. However, image compression and semantic inference are generally conducted separately, and t...     

Practical Fusion of Quantized Measurements via Particle Filtering

Most treatments of decentralized estimation rely on some form of track fusion, in which local track estimates and their associated covariances are communicated. This implies a great deal of communication; and it was recently proposed that by an intelligent quantization directly of measurements, the communication needs could be considerably cut. However, several issues were not discussed. The first of these is that estimation with quantized measurements requires an update with a non-Gaussian distribution, reflecting the uncertainty within the quantization "bin."; In general this would be a difficult task for dynamic estimation, but Markov-chain Monte-Carlo (MCMC, and specifically here particle filtering) techniques appear quite appropriate since the resulting system is, in essence, a nonlinear filter. The second issue is that in a realistic sensor network it is to be expected that measurements should arrive out-of-sequence. Again, a particle filter is appropriate, since the recent literature has reported particle filter modifications that accommodate nonlinear-filter updates based on new past measurements, with the need to refilter obviated. We show results that indicate a compander/particle-filter combination is a natural fit, and specifically that quite good performance is achievable with only 2-3 bits per dimension per observation. The third issue is that intelligent quantization requires that both sensor and fuser share an understanding of the quantization rule. In dynamic estimation this is a problem since both quantizer and fuser are working with only partial information; if measurements arrive out-of-sequence the problem is compounded. We therefore suggest architectures, and comment on their suitabilities for the task. A scheme based on delta-modulation appears to be promising.     

M-correlated sweeps performance analysis of mean-level CFARprocessors in multiple target environments

This paper is devoted to the detection performance evaluation of the mean-level (ML) constant false-alarm rate (CFAR) detectors processing M-correlated sweeps in the presence of interfering targets. The consecutive pulses are assumed to be fluctuating according to the Swerling I model. Exact expressions are derived for the detection probability of the conventional mean-level detector (MLD) and its modified versions under Rayleigh fluctuating target model. Performance for independent sweeps can be easily obtained by setting the sweep-to-sweep correlation coefficient equal to zero. Results are obtained for both homogeneous and nonhomogeneous background environments. It is shown that for fixed M, the relative improvement over the single sweep case increases as the correlation between sweeps decreases. For the same parameter values, the minimum MLD has the best performance in the presence of extraneous target returns among the reference noise samples     

发动机故障诊断的MONTE CARLO通解算法

给出了发动机故障诊断的 Monte Carlo通解算法,该算法可以有效地解决发动机故障诊断算法中由于故障方程存在多重共线性所引起的误诊、漏诊与多解问题。首先给出求解故障方程的 Monte Carlo算法,该算法能够保证得到满足故障方程的全部合理解;能够简单地用于各种故障相关性准则 (例如各种残差向量范数准则 )以及亚定故障方程的求解问题;并且算法简单易行,无需为每个特定情况专门编制计算程序。其次提出了基本解与通解的概念,并且给出了利用主成分分析与利用主因子模型求基本解的两种有效算法。利用基本解与通解算法可以将由于多重共线性引起的复杂故障诊断结果表示为简单明了的形式,有利于对诊断结果作出正确决策。用计算机模拟方法对算法的有效性进行了分析,对于 JT9D发动机气路方面的 2 4个实际故障样本,所给出的算法的确诊率为 86%～ 92 %     

Statistically Optimum Optical Data Processing with Automatic Focus Estimation

The problem of optimally processing data with unknown focus is investigated. Optimum data processors are found by the method of maximum likelihood under a variety of assumptions that apply to most of the situations arising in practice. The unknown focus may be either an unknown parameter or an unknown random variable; the signal may be of known form or a random function; it is further assumed that the signal is received in additive, white, Gaussian noise. The problems of jointly estimating other unknown parameters and, in the case of a random signal, jointly estimating the signal, are also treated. The asymptotic variance and correlation of the estimators is discussed. Electrooptical realizations of the maximum likelihood computers are given. An iterative method of solution of the likelihood equation is also discussed. The discussion and results are directly applicable to the processing of synthetic aperture radar data.     

量化对GPS接收机捕获性能影响的研究

由于接收机输入端的信噪比对接收机的捕获性能具有很重要的意义,所以要尽量减少接收机输入端信噪比的损失。本文首先给出了含高斯白噪声的GPS信号经过不同量化位数和量化电平的量化器后的输出信噪比计算公式;然后,在此基础上分析其对接收机捕获性能的影响;最后,仿真实验表明,接收机可以在不影响捕获性能的情况下,尽量选择合适的量化位数和量化门限。实验结果也为量化器的设计提供了重要的技术参考。     

相关系数ARMA(p,q)序列分析方法

本文提出相关系数ARMA(p,q)序列分析方法。相关系数ARMA(p,q)序列是从非平稳序列中分离出的一类工程上常见且便于研究的时间序列,在模式识别、故障诊断、信号处理、自动控制和结构响应分析等领域有着广泛的应用。传统的相关函数ARMA(p,q)序列仅是它的一个特例。文中建立了相关系数ARMA(p,q)序列的条件极大似然估计和精确极大似然估计,前者在样本较大时简单便于工程应用,后者则在样本较小时仍具有较高的精度,它们通过时域的全程分析,充分利用样本信息确定相关系数ARMA(p,q)序列的均值函数、方差函数和相关系数函数。在此基础上可进行高精度的频谱分析。     

A noncoherent adaptive detection technique

An adaptive detection technique suitable for both stationary and nonstationary noise environments based upon a generalized likelihood ratio test (GLRT) formulation is presented. The detector, which is statistically equivalent to a special form of the Wilks's lambda test, noncoherently combines the information contained in a pulse train of arbitrary length for decision-making purposes. The probability density function of the test under the noise only hypothesis is shown to be central χ². Under the signal plus noise hypothesis, an exact statistical characterization of the test cannot be obtained, and, therefore, a Chernoff bound is derived. Results in terms of the probability of detection versus signal-to-noise ratio (SNR) obtained from Monte Carlo simulation, the Chernoff bound, and the optimal matched filter case are examined. The performance of the noncoherent detector is shown to be a function of the covariance matrix estimate and the number of data samples