Multiresolution FIR neural-network-based learning algorithm applied to network traffic prediction

In this paper, a multiresolution finite-impulse-response (FIR) neural-network-based learning algorithm using the maximal overlap discrete wavelet transform (MODWT) is proposed. The multiresolution learning algorithm employs the analysis framework of wavelet theory, which decomposes a signal into wavelet coefficients and scaling coefficients. The translation-invariant property of the MODWT allows alignment of events in a multiresolution analysis with respect to the original time series and, therefore, preserving the integrity of some transient events. A learning algorithm is also derived for adapting the gain of the activation functions at each level of resolution. The proposed multiresolution FIR neural-network-based learning algorithm is applied to network traffic prediction (real-world aggregate Ethernet traffic data) with comparable results. These results indicate that the generalization ability of the FIR neural network is improved by the proposed multiresolution learning algorithm.     

Multiresolution Markov models for signal and image processing

Reviews a significant component of the rich field of statistical multiresolution (MR) modeling and processing. These MR methods have found application and permeated the literature of a widely scattered set of disciplines, and one of our principal objectives is to present a single, coherent picture of this framework. A second goal is to describe how this topic fits into the even larger field of MR methods and concepts-in particular, making ties to topics such as wavelets and multigrid methods. A third goal is to provide several alternate viewpoints for this body of work, as the methods and concepts we describe intersect with a number of other fields. The principle focus of our presentation is the class of MR Markov processes defined on pyramidally organized trees. The attractiveness of these models stems from both the very efficient algorithms they admit and their expressive power and broad applicability. We show how a variety of methods and models relate to this framework including models for self-similar and 1/f processes. We also illustrate how these methods have been used in practice.     

脑电信号成分稀疏分析范式及其可行证明

稀疏分解技术作为一种可靠的信号处理与传输方法,在包括EEG的多种时变信号分析和处理领域得到了广泛的应用。在EEG信号的成分分析中,现有算法（ICA,EMD）等都存在分解结果与真实成分显著不符的情况,难以对实际成分的波形进行估计。本文在稀疏分解算法基础上,通过对样本稀疏分布情况进行度量,给出了一个经过改良的稀疏性能评价指标（SPI）并以此建立了一个新的成分分析范式和相应的优化函数,经过理论和实际证明,该范式在成分分析领域能比传统方法更有效地使分解结果趋向于真实成分,对EEG信号、乃至其他时变信号的成分解析都具有相当的积极意义。      

A wreath product group approach to signal and image processing .I.Multiresolution analysis

We propose the use of spectral analysis on certain noncommutative finite groups in digital signal processing and, in particular, image processing. We pay significant attention to groups constructed as wreath products of cyclic groups. Within this large class of groups, our approach recovers the discrete Fourier transform (DFT), the Haar wavelet transform, various multichannel pyramid filter banks, and other aspects of multiresolution analysis as special cases of a more general phenomenon. In addition, the group structure provides a rich algebraic structure that can be exploited for the analysis and manipulation of signals. Our approach relies on a synthesis of ideas found in the early work of Holmes (1987, 1990), Karpovsky and Trachtenberg (1985), and others on noncommutative filtering, as well as Diaconis's (1989) spectral analysis approach to understanding data     

Adaptive machine learning algorithm employed statistical signal processing for classification of ECG signal and myoelectric signal

Multidimensional Systems and Signal Processing - In this research paper we present designing and evaluating the electrocardiography (ECG) and Myoelectric signal (EMG) pattern recognition methods...     

Automotive signal diagnostics using wavelets and machine learning

In this paper, we describe an intelligent signal analysis system employing the wavelet transformation in the solution of vehicle engine diagnosis problems. Vehicle engine diagnosis often involves multiple signal analysis. The developed system first partitions a leading signal into small segments representing physical events or states based on wavelet multi-resolution analysis. Second, by applying the segmentation result of the leading signal to the other signals, the detailed properties of each segment, including inter-signal relationships, are extracted to form a feature vector. Finally, a fuzzy intelligent system is used to learn diagnostic features from a training set containing feature vectors extracted from signal segments at various vehicle states. The fuzzy system applies its diagnostic knowledge to classify signals as abnormal or normal. The implementation of the system is described and experiment results are presented     

Active and cooperative learning in signal processing courses

This work describes positive effects of using active and cooperative learning (ACL) methods to improve signal processing instruction. It provides examples, references, and assessment data that encourage other instructors to consider this approach. Conclusions are based on impressions gathered through conversations with students during office hours as well as on responses from anonymous student opinion surveys. In addition to these subjective assessments, preliminary quantitative data measured with the signals and systems concept inventory (SSCI) support the benefits of ACL techniques in signal processing courses.     

Multiresolution ESPRIT algorithm

Multiresolution ESPRIT is an extension of the ESPRIT direction finding algorithm to antenna arrays with multiple baselines. A short (half wavelength) baseline is necessary to avoid aliasing, and a long baseline is preferred for accuracy. The MR-ESPRIT algorithm allows the combination of both estimates. The ratio of the longest baseline to the shortest one is a measure of the gain in accuracy. Because of various factors, including noise, signal bandwidth, and measurement error, the achievable gain in accuracy is bounded     

Multiresolution image modelling

A novel approach to the statistical modelling of images based on widely used multiresolution image representations, such as pyramids and wavelets, is presented. It is shown how models which are causal in the scale dimension provide an effective way of expressing the concept of successive approximation, or coarse-fine refinement, by the addition of details to the coarse structure of images. The ideas are illustrated using applications including image enhancement, data compression and segmentation     

样条多分辨分析

本文根据平移不变子空间二尺度分解理论，系统，简结地构造了各类样条多分辨分析，其中涉及的运算只有序列的卷积，求逆，开平方和抽取以及信号的内积。     

Machine learning for multimodality genomic signal processing

This paper discusses how machine learning can be applied to genomic signal processing, particularly via fusion of multiple biological or algorithmic modalities, to improve prediction performance.     

A methodology for microcontroller signal frequency stress prediction

This paper proposes a methodology for analysis and prediction of microcontroller failures due to signal frequency stress. Microcontrollers are constantly under signal frequency stress when controlling external devices. However, signal frequency stress has received relatively little attention in comparison with other sources of stress such as temperature, humidity, and voltage. This study involved the following steps: (1) identifying the highest stress point before failure; (2) dividing signal frequency into different stress levels; (3) characterizing the impact of signal frequency stress on IC functionality; (4) constructing a thermal profile of a microcontroller under signal frequency stress over time; (5) predicting stress levels using regression and neural network methods; and (6) comparing and contrasting performance differences for each method. Results indicated that the average prediction error is about 7.9% for the neural network approach and about 23.8% for the statistical regression approach. This may be due to the neural network modeling approach’s inherent ability to tolerate noise in the data due to factors such as variation in quality due to variations in the manufacturing process. This general methodology has also been utilized with low error rates in failure analysis and stress prediction in operational/power amplifiers (8% error rate), timer oscillator chips (25%) and resistors (30%).     

基于信道借用和信号预测的切换方法

随着业务需求的日益增长,蜂窝也越变越小,移动蜂窝通信环境中的切换成为日益重要的问题。在CDMA蜂窝系统中使用软切换,为了解决切换时的信道短缺问题,提出了许多解决方法。该文中提出了称为基于信道借用和信号预测的切换算法。当切换请求到达蜂窝时,如果没有空闲信道,就将从参与软切换的静止呼叫借用一信道,并将借用的信道分配给移动呼叫的切换请求.如果没有信道可借用,就将切换请求放入队列中,使用信号预测的方法来确定队列中的优先级。并将此切换方式的性能与其它切换方式进行了比较。     

基于无监督学习的目标轨迹预测

针对情报与侦察监视领域中目标轨迹预测问题,提出了一种基于无监督学习的预测方法。首先,根据历史信息分析目标历史活动规律;其次,构建隐马尔科夫模型,通过无监督学习自动实现预测目标在栅格网中的运动方向;最后,根据学习得到的运动方向和目标运动速度信息,计算未来短期内的目标轨迹。数值仿真验证了该方法能够有效地预测目标在未来短时刻内（通常为5 min）的运动轨迹。     

基于机器学习的PCB串扰预测

随着电子系统中逻辑和时钟频率的迅速提高以及信号边沿的不断变抖,串扰成为印刷电路板(PCB)设计人员必须关心的问题。高速电路仿真软件帮助设计人员降低了一定的设计成本,但对串扰的仿真预测仍需花费大量时间。为提高PCB串扰预测的效率,提出一种用于描述PCB的统一数据结构,全面分析了PCB产生串扰的因素,选用自然语言处理(NLP)模型构建了用于PCB串扰预测的系统,成功将PCB串扰预测的时间降至秒级,并拥有73.2%的准确率。     

基于改进型深度学习的流量预测

为了解决无线网络中流量的预测精度不高的问题,提出了一种自适应分组的栈式自编码( AG-SAEs)深度学习预测方法。在数据的预处理过程中,首先使用最大最小方式对数据进行归一化处理,并提出一种新型的自适应分组方法,把归一化后的链路数据进行关联性分组;然后,基于深度学习方法建立了一个多输入多输出的预测模型,并将分组后的数据输入到预测模型中,对该模型进行训练来建立输入和输出流量之间的映射关系;最后,为了进一步提高预测精度,在模型的训练过程中,使用改进型的牛顿法来进行权值参数更新。仿真实验以及和其他算法对比的结果证实了所提方案具有更小的预测相对误差。     

基于对比度的多分辨图像融合

本文给出一种基于对比度的多分辨图像融合算法.首先,利用小波变换得到待融合图像的多分辨分析,同时得到图像的多分辨对比度序列;然后以对比度为判据,在图像的多分辨分析的相应各级上进行融合,得到融合图像的多分辨分析;最后,利用小波逆变换重构融合图像.算法使用同一场景可见光图像和红外图像进行了验证,结果表明,融合图像完好地显示了源图像各自的信息.