基于机器学习的恶意URL识别

网络攻击成为日益重要的安全问题,而多种网络攻击手段多以恶意URL为途径。基于黑名单的恶意URL识别方法存在查全率低、时效性差等问题,而基于机器学习的恶意URL识别方法仍在发展中。对多种机器学习模型特别是集成学习模型在恶意URL识别问题上的效果进行研究,结果表明,集成学习方法在召回率、准确率、正确率、F1值、AUC值等多项指标上整体优于传统机器学习模型,其中随机森林算法表现最优。可见,集成学习模型在恶意URL识别问题上具有应用价值。     

Robust maximum energy adaptive matched field processing

Adaptive array processing algorithms have achieved widespread use because they are very effective at rejecting unwanted signals (i.e. controlling sidelobe levels) and in general have very good resolution (i.e. have narrow mainlobes). However, many adaptive high-resolution array processing algorithms suffer a significant degradation in performance in the presence of environmental mismatch. This sensitivity to environmental mismatch is of particular concern in problems such as acoustic array processing in the ocean where the array processor's knowledge of the propagation characteristics of the ocean is usually imperfect. A maximum energy matched field processor is formulated which “tunes” the replica vector used by a minimum variance distortionless response (MVDR) matched field processor to adjust for fluctuations in the environment. The resulting processor has the same structure as the two-stage MVDR processor interpretation of the adaptive minmax matched field processor proposed previously. However, the criteria which the two processors use to “tune” replica vector differ. The result is that the MEMV processor achieves a level of performance superior to that of the adaptive minmax processor and close to that of traditional adaptive processors while limiting the sensitivity of the processor to environmental mismatch. An algorithm for implementing the professor is developed, and the performance of the processor is analyzed     

Vectorized transforms in scalar processors

We disclose a generalized approach to creating efficient implementations of linear, orthogonal transforms, with specific examples discussed for the 8 x 8 DCT used in image compression. We connect this with a method for performing signed, parallel processing in scalar, off-the-shelf processors for integer transforms. Uniform data precision may be used, but is not required for the method. The coefficients resulting from the new algorithm converge more quickly than the approximation made to the coefficients. Furthermore, the new algorithm allows more control of the specific representation chosen for the coefficients, as is detailed below. The methods described were designed for addressing this need with two's-complement arithmetic. Data that can be processed in parallel, because of the algorithm structure, are packed in a "vector" format, described, into registers. Many signed arithmetic operations can be performed on these vectors, including addition, subtraction, multiplication by scalars, shifting, and others. When the parallel processing is completed, the vectors can be unpacked into scalar values for storage or subsequent processing. The importance of these methods lies in their handling of carries and borrows in the packed vector format. The generalized method is described. Notation is given at the beginning to establish consistency through the article. We discuss a generalized approach to integer transforms, using the DCT as a specific example. Then we detail the vector format, which allows vector computation in scalar processors of parallelizable algorithms. The IDCT is used as a numerical example in the discussion of the vector format. The results were developed for high-end printers (e.g., more than 100 pages per minute), where image compression and decompression must be performed in real time, either in FPGAs, or in embedded processors; however, the methods are applicable to a broad range of signal processing systems     

双离轴衍射透镜的矢量设计算法

设计特征尺寸可与照射光波长相比拟的衍射光学器件时,必须采用严格的矢量设计算法.为克服传统搜索算法计算次数多、收敛速度慢的缺点,同时结合时域有限差分算法进行光场与衍射器件相互作用的矢量计算,提出了一种高效的迭代算法.利用该算法,设计了双离轴透镜,结果表明,该算法计算次数少、收敛效果好.     

基于单特显点目标回波的阵列3D SAR 幅相误差一致性校正方法研究

阵列3D SAR 技术以实孔径和合成孔径相结合的方式,实现对观测场景目标的3 维分辨。实际阵列SAR系统接收回波中不可避免地存在多通道幅相误差问题,直接成像处理会造成重建的雷达图像质量降级。该文建立了考虑幅相误差的阵列3D SAR 回波模型,提出了一种基于单特显点目标回波进行阵列幅相误差估计的方法,并针对阵列3D SAR 应用,提出了对幅相误差进行补偿的数据处理流程,最后通过仿真和实际数据处理对提出的模型和方法进行了验证。      

Very Fast Signal Processors As a Result of the Coupling of Surface Acoustic Wave and Digital Technologies

The recent progress of the digital and the surface acoustic wave (SAW) technologies have made them compatible; and it is now possible to design signal-processing modules which benefit from the flexibility of the digital techniques and the very high computation speed of the SAW techniques. Very fast signal processors can now be built which are able to process several tens of megasamples per second and whose volume and power consumption are limited. This paper shows the compatibility of these technologies and the advantages yielded by their joint use. Several examples are described which relate to one- and two-dimensional Fourier and correlation processors.     

A Generalized Mixed-Radix Algorithm for Memory-Based FFT Processors

In this brief, a generalized mixed-radix (GMR) algorithm is proposed for memory-based fast Fourier transform (FFT) processors to support prime-sized and traditional $2^{n}$ -point FFTs simultaneously. It transforms the index to a multidimensional vector for efficient computation. By controlling the index vector to satisfy the “vector reverse” behavior, the GMR algorithm can support not only in-place policy for both computation and I/O data for continuous data flow to minimize the memory size but also multibank memory structures to increase the maximum throughput without memory conflict. Finally, a low-complexity implementation of an index vector generator is also proposed for our algorithm.      

Bootstrap Particle Filtering

This article provides an overview of nonlinear statistical signal processing based on the Bayesian paradigm. The next-generation processors are well founded on MC simulation-based sampling techniques. The development of the sequential Bayesian processor is reviewed using the state-space models. The popular bootstrap algorithm was outlined and applied to an ocean acoustic synthetic aperture towed array target tracking problem to test the performance of a particle filtering technique.     

Recursive estimation of images using non-Gaussian autoregressivemodels

We consider recursive estimation of images modeled by non-Gaussian autoregressive (AR) models and corrupted by spatially white Gaussian noise. The goal is to find a recursive algorithm to compute a near minimum mean square error (MMSE) estimate of each pixel of the scene using a fixed lookahead of D rows and D columns of the observations. Our method is based on a simple approximation that makes possible the development of a useful suboptimal nonlinear estimator. The algorithm is first developed for a non-Gaussian AR time-series and then generalized to two dimensions. In the process, we draw on the well-known reduced update Kalman filter (KF) technique of Woods and Radewan (1977) to circumvent computational load problems. Several examples demonstrate the non-Gaussian nature of residuals for AR image models and that our algorithm compares favorably with the Kalman filtering techniques in such cases.     

基于DA算法的二维DCT的FPGA实现

研究了一种采用现场可编程门阵列(FPGA)实现超高性能二维离散余弦变换(DCT)的方法。在DCT算法结构上利用了变换的可分离性和行列的可分解性采用行列分解的方法将二维DCT转换为2个串行的一维DCT实现,同时采用了基于分布算法(Distributed Arithmetic)的乘法累加结构,从而极大地减少了硬件资源需求,提高了运算速度,使图像处理的实时性得到了大幅提高。最后还给出了FPGA的实现和仿真结果。     

3-D Kalman filter for image motion estimation

This paper presents a new three-dimensional (3-D) Markov model for motion vector fields. The three dimensions consist of the two space dimensions plus a scale dimension. We use a compound signal model to handle motion discontinuity in this 3-D Markov random field (MRF). For motion estimation, we use an extended Kalman filter as a pel-recursive estimator. Since a single observation can be sensitive to local image characteristics, especially when the model is not accurate, we employ windowed multiple observations at each pixel to increase accuracy. These multiple observations employ different weighting values for each observation, since the uncertainty in each observation is different. Finally, we compare this 3-D model with earlier proposed one-dimensional (1-D) (coarse-to-fine scale) and two-dimensional (2D) spatial compound models, in terms of motion estimation performance on a synthetic and a real image sequence.     

Mapping of Trellises Associated with General Encoders onto High-Performance VLSI Architectures

A rate 1/n binary generic convolutional encoder is a shift-register circuit where the inputs are information bits and the outputs are blocks of n bits generated as linear combinations on the appropriate shift register contents. The decoding of the outputs of a convolutional encoder can be carried out by the well-known Viterbi algorithm. The communication pattern of the Viterbi Algorithm is given as a graph, called trellis, associated to the state diagram of the corresponding encoder. In this paper we present a methodology that permits the efficient mapping of the Viterbi algorithm onto a column of an arbitrary number of processors. This is done through the representation of the data flow by using mathematical operators which present an inmediate hardware projection. A single operator string has been obtained to represent a generic encoder through the study of the data flow of free-forward encoders and feed-back encoders. The formal model developed is employed for the partitioning of the computations among an arbitrary number of processors in such a way that the data are recirculated opimizing the use of the processors and the communications. As a result, we obtain a highly regular and modular architecture suitable for VLSI implementation.     

基于线性校正总体最小二乘准则的三维说话人定位算法

针对到达时间差(TDOA)度量误差和麦克风阵列误差,提出了一种基于线性校正总体最小二乘准则的三维说话人定位算法.该定位算法通过对球面误差准则进行总体最小二乘优化来控制阵列误差的影响,并通过对优化过程引入方向参数的二次约束来控制算法的方差.仿真实验表明,本算法可以同时兼顾定位的方差性能和偏差性能,给出稳健的定位结果.     

DReAC:一种新型动态可重构协处理器

本文提出了一种应用于数据并行和高密度计算任务的新型动态可重构协处理器——DReAC.DReAC可以独立地以并行或流水工作模式重构协处理器内部数据路径,完成主处理器分配的任务.DReAC由全局控制器、计算阵列和阵列数据缓冲区三部分组成.文中简要介绍了DReAC系统模型,并使用该模型模拟了部份典型算法在DReAC中的实现.仿真结果表明,在典型的多媒体和信号处理应用中,DReAC能够达到通用处理器的10倍以上的速度,甚至在某些应用中远优于其他可重构处理器的性能.     

3D EM/MPM Image Segmentation Using an FPGA Embedded Design Implementation

This paper presents a Field Programmable Gate Array (FPGA) based embedded system which is used to achieve high speed segmentation of 3D images. Segmentation is performed using Expectation-Maximization (EM) with Maximization of Posterior Marginals (MPM) Bayesian algorithm. This algorithm segments the 3D image using neighboring pixels based on a Markov Random Field (MRF) model. In this system, the embedded processor controls a custom circuit which performs the MPM and portions of the EM algorithm. The embedded processor completes the EM algorithm and also controls image data transmission between host computer and on-board memory. The whole system has been implemented on Xilinx Virtex 6 FPGA and achieved over 100 times processing improvement compared to standard desktop computer. Three new techniques were the key to achieve this speed: Pipelined computational cores, sixteen parallel data paths and a novel memory interface for maximizing the external memory bandwidth.

     

一种基于模式挖掘的网络通信安全方法研究

针对子项集时间序列,提出了一种模式挖掘的数学模型。首先设置实时频率矢量;针对每个到达子项集中的项,计算并更新实时频率矢量中记录项的平均频率;以模式考察时间阈值为周期,每周期计算当前实时频率矢量和模式集中现有实时频率矢量的皮尔松相关性;如果相关系数大,则说明当前模式已经存在于模式集中;如果相关系数小,则说明当前模式是一个新模式,从而加入模式集。实验中,运用此模型到Web安全上,对新浪门户网站进行考察和检验。结果表明,该模型对于防御Web非常有效。     

Fourier domain reconstruction of synthetic focus acoustic imaging system

This letter presents a new numerical reconstruction procedure for the synthetic focus system which improves the transversal resolution in acoustic B-scope imaging. It involves calculations in Fourier domain and could be made especially efficient by the use of now commonly available hardwired FFT processors. Computer simulation demonstrates the feasibility of the procedure.     

Soft computing-based synthesis model for equilateral triangular ring printed antenna

A soft computing-based synthesis model is proposed for the design of equilateral triangular ring printed antennas (ETRPAs) that operate at ultrahigh band applications. The synthesis of printed antennas having irregular shapes needs great efforts and time. Soft computing models eliminates complex, lengthy and time consuming mathematical procedures for synthesis and analysis of the printed antennas. In this paper, a soft computing model based on artificial neural network (ANN) is constructed for synthesis of ETRPAs. To generate training and testing data set, number of 100 ETRPAs with various geometrical and electrical parameters are simulated in terms of resonant frequency with the aid of a 3D full wave simulator. Soft computing model based ANN is designed to obtain the patch physical dimensions of ETRPAs simultaneously. The soft computing model is trained with simulated data set of 75 ETRPAs and tested with remainders 25 ETRPAs. A prototype of ETRPA is then fabricated to verify the proposed model. The testing results of the soft computing model are found in close agreement with simulated and experimental data. The presented model is simply/fast predicting the patch dimensions of ETRPAs and it is extremely useful to antenna engineers for the design of the ETRPAs.     

Time-domain finite difference and EMC study of hot carrier transport in GaAs on a picosecond scale

It is now possible to use optoelectronic techniques to generate and measure electrical waveforms with subpicosecond risetimes. Though experimental methods have become relatively advanced, simulation techniques for analyzing and modelling such ultrafast transients remain inexact and oversimplified. The simplifications commonly made while developing equivalent circuit models for the experimental structures, prevent accurate simulation of the electrical response during the picosecond regime. In order to obtain a better physical picture, it is essential to develop a more physical model for the microstrip circuits routinely used in such experiments. Furthermore, non-Ohmic transport behavior for the photogenerated carriers within the optoelectronic device also needs to be correctly incorporated. We address both the circuit and device response issues through a novel scheme which combines direct time domain solutions to Maxwell's equations with the ensemble Monte Carlo model for carrier transport. By coupling the electromagnetics with the EMC, we avoid having to make assumptions whose validity breaks down for picosecond transport.     

Real-Time Calculation of Switching Angles Minimizing THD for Multilevel Inverters With Step Modulation

Multilevel inverters have been widely applied in industries. A family of optimal pulsewidth modulation (PWM) methods for multilevel inverters, such as step modulation, can generate output voltage with less harmonic distortion than popular modulation strategies, such as the carrier-based sinusoidal PWM or the space vector PWM. However, some drawbacks limit the application of optimal PWM. One of such crucial drawback is that the optimal switching angles could not be calculated in real-time and one has to rely on lookup tables with precalculated angles. We propose a novel real-time algorithm for calculating switching angles that minimizes total harmonic distortion (THD) for step modulation. We give a mathematical proof that the output voltage has the minimum THD. We implemented the algorithm on a digital signal processor and provide experimental results that verify the performance of the proposed algorithm.