Phylogenetic methods are being increasingly used to help understand the transmission dynamics of measurably evolving viruses, including HIV. Clusters highly similar sequences often observed, which appear follow a 'power law' behaviour, with small number very large clusters. These clusters may identify subpopulations in an epidemic, and inform where intervention strategies should be implemented. However, clustering samples does not necessarily imply presence subpopulation high rates, as...

In modern intelligent transportation system, traffic control and congestion management are main components of it. Real-time accurate short-term flow prediction is the precondition key to realizing management. order improve accuracy prediction, a combined model method proposed in this paper which based on Xgboost LightGBM algorithms. First, data preprocessed features sampled. Then, according different characteristics, models constructed by using Finally, these merged generate final model....

This paper proposes a joint acoustic echo cancellation (AEC) and speech enhancement method with adaptive filter deep neural network (DNN) model. A partitioned block is adopted for linear AEC followed by convolutional transformer based model to suppress the residual echo, noise, reverberation. The DNN has three modules: encoder, dual-path (DPT) decoder. encoder explore potential relationships of far-end near-end signals attention mechanism transformer. DPT module further used context...

This paper presents a Bayesian inference-based dynamic linear model (DLM) with switching based on three-phase traffic flow theory to predict online short-term travel time plate recognition data. The proposed method combines the DLM Hidden Markov Model (HMM) capture probability of breakdown and delays associated congestion. By viewing fluctuations as time-varying stochastic process due unforeseen events (e.g., incidents, accidents, or bad weather), (SDLM) employs HMM determine optimal state...

Most of current best performing Acoustic Scene Classification (ASC) systems utilize Mel scale spectrograms with Convolutional Neural Networks (CNNs). is a common way to suit frequency warping human ears, strict decreasing resolution on low high range. However, we find that significant bins are located at mid range for some acoustic scenes, such as travelling by bus, tram or train. In this paper, show better ASC can be automatically learned from raw spectrograms, using Kullback-Leibler (KL)...

Conflict management is a hot research topic in Dempster–Shafer theory which used to avoid the counterintuition problem of combination results. In present conflict methods, conflicting evidence assigned smaller weight reduce its influence on result. However, these methods will be disabled when similarity collision occurs. this article, new method based and Basic Probability Assignment-Matrix proposed; proposed method, diminished by computing index each element evidence, then, more reasonable...

In many machine learning or patter recognition tasks such as classification, datasets with a large number of features are involved. Feature selection aims at eliminating the redundant and irrelevant which would bring computational burden degrade performance algorithms. Particle swarm optimization (PSO) has been widely used in feature due to its global search ability efficiency. However, PSO was originally designed for continuous problems discretization is still problem needs further...

Traditionally, speech quality evaluation relies on subjective assessments or intrusive methods that require reference signals additional equipment. However, over recent years, non-intrusive assessment has emerged as a promising alternative, capturing much attention from researchers and industry professionals. This article presents deep learning-based method exploits large-scale simulated data to improve the accuracy generalization of methods. The major contributions this are follows. First,...

Power MOSFETs are important satellite components. In this paper, a radiation-hardened trench power MOSFET is designed. order to meet the requirements of aerospace uses, anti-total-dose radiation effect performance improved by changing process sequences, gate oxide growing at low temperature and controlling growth atmosphere strictly; Anti-single-event increasing injection dose P-type base region, reducing doping concentration source region shortening width. The fabricated using...

Sparse matrix-vector multiplication (SpMV) is an elementary kernel of many high-performance computing (HPC) applications, and it often one the performance bottlenecks them. Accelerating SpMV on vector processors usually faces several issues including irregular data accesses, memory bandwidth limitation, short problem. Based a detailed analysis effects interactions various technologies introduced by state-of-the-art studies (ALBUS, CVR, CSR5, SELL-C-σ etc.), we propose Regu2D, comprehensive...

A compact module for long-pulse power generator, based on Blumlein pulse forming network (PFN), was designed. Two PFNs with L-type configuration and 20 Ω characteristic impedance were connected symmetrically to the primary coil of linear transformer driver (LTD) driven by an identical high voltage spark switch ensure two synchronizing operation. The output 10 water load is about 135 kV in amplitude 200 ns duration a rise time ∼50 flat top ∼100 ns. On this basis, repetitive generator PFN-LTD...

为提高在低频、宽带、高强度及大振幅振动环境下的可靠性,提出一种换向激励式压电振动俘能器,它由拾振器和换能器组成。换能器的振动方向垂直于拾振器振动方向(环境振动方向),振动方向的转换通过磁力换向结构实现,换向结构使响应振幅不随外部激励的增加而一直增加,从而提高可靠性。建立了俘能器的动力学模型,通过数值仿真和实验获得了相关参数对其输出特性的影响。仿真和实验结果表明:激励频率小于20 Hz时,该两自由度系统存在两阶谐振频率使输出电压达到峰值,一阶为拾振器谐振频率,二阶为换能器谐振频率。随着拾振簧片长度和拾振质量的增加,一阶谐振频率升高,所对应的输出电压基本不变;二阶谐振频率基本不变,所对应的输出电压逐渐升高;工作频带变宽。当外部激励振幅达到阈值时,换能器的响应振幅被有效控制,输出电压不再随之增加,最佳负载电阻为540 kΩ,此时输出功率最大为0.4 mW。实际应用中可通过改变俘能器的结构参数调整谐振频率及输出电压,将响应振幅控制在安全区域内,以适应低频、宽带、高强度及大振幅的工作环境。

This article explores the coupling relationship between modified nodal analysis (MNA), time-domain finite difference (TDFD), and particle-in-cell (PIC) simulation methods. It proposes a tight method for road-field-particle collaboration simulation. Subsequently, this is used to simulate pulse electron beam generator (EBG) from linear transformer driver (LTD) cavity transmission line diode load collaboratively. The vertical distance anode cathode of optimized by considering its effects on...

Abstract Data augmentation is required for the implementation of many Markov chain Monte Carlo (MCMC) algorithms. The inclusion augmented data can often lead to conditional distributions from well‐known probability some parameters in model. In such cases, collapsing (integrating out parameters) has been shown improve performance MCMC We show how integrating infection rate parameter epidemic models leads efficient algorithms two very different scenarios, final outcome a multitype SIR and...

A fully integrated Ku-band voltage controlled oscillator (VCO) is presented in an InGaP/GaAs heterojunction bipolar transistor (HBT) technology. To achieve the wide tuning range (TR), VCO employs a Colpitts configuration, and simultaneously achieves high output power. The implemented demonstrates oscillation frequency from 12.82 to 14.97 GHz, TR of 15.47%, power 0.31 6.46 dBm, phase noise −94.9 dBc/Hz at 1 MHz offset 13.9 GHz center frequency. consumes 52.75 mW 5 V supply occupies area 0.81...

A repetitive electron beam driver based on 6-stage Linear Transformer Driver (LTD) cavities is presented. Each cavity consists of two Blumlein pulse forming networks (BPFNs) sharing a laser trigger switch. Voltage adding obtained by means (LT) technology. The LTD BPFNs connected in parallel, an LT with ratio 1:1, and laser-triggered spark energy efficiency single was investigated. results show that the L-type BPFN 94.3% 92.8%, respectively. An consisting 6 such developed output voltage 740...

The realization technique of modularization on the long pulse high power generator was expounded. relation between ferromagnetic dissipation linear transformer and exciting magnetic current analyzed. main influential factors wave distortion from primary to system output waveforms were discussed. Finally, through engineering experiments, a electron beam up several gigawatt with repetition frequency dozen hertz rise time 40—50 ns obtained. This paper gives new idea for research higher rate.

In virtualized cloud computing environments, the memory resource is becoming major bottleneck for limiting server consolidation degree of virtual machines. The fact that great amount redundant cache pages exist in consolidated machines' disk cache, which we refer to as multi-caching problem, give us a chance improve efficiency utilization. Previous scanning-based deduplication well-known technique eliminate with identical content. But periodically scanning cost expensive, and it hard detect...

In order to accommodate huge number of antennas in a limited antenna size, large scale array is expected have three dimensional (3D) structure.By using the Active Antenna Systems (AAS), weights elements arranged vertically could be configured adaptively.Then, degree freedom (DOF) vertical plane provided for system design.So three-dimension MIMO (3D MIMO) realized solve actual implementation problem massive MIMO.However, 3D systems, pilot contamination studied 2D systems and inter-cell...

Currently, there is limited research on the influence of gas ionization pulse formation process in power source-driven loads. This paper introduces a road-field-Particle-In-Cell (PIC)/Monte Carlo Collision (MCC) collaborative simulation method that can accurately simulate Linear Transformer Driver (LTD) electron beam generation (EBG). The couples electromagnetic field and charged particle simulated through PIC/MCC with circuit modules, load's voltammetry characteristics real-time feedback to...

The stability condition of intense annular electron beam transmission is analyzed and deduced. Based on the static electric-magnetic field simulation, position configuration guide magnetic-field its magnitude as well power supply are designed. Then experiments repetitive rate inte nse current electron-beam diode carried out CHP01 generator, ima ge measured too. results show that t he designed 1-s magnetic satisfies to transmission. It can stably transmit elect ron-beam pulse 800kV/8kA, width...

Abstract Phylogenetic methods are being increasingly used to help understand the transmission dynamics of measurably evolving viruses, including HIV. Clusters highly similar sequences often observed, which appear follow a ‘power law’ behaviour, with small number very large clusters. These clusters may identify subpopulations in an epidemic, and inform where intervention strategies should be implemented. However, clustering samples does not necessarily imply presence subpopulation high rates,...