Speeding up the MATLAB complex networks package using graphic processors

The availability of computers and communication networks allows us to gather and analyse data on a far larger scale than previously. At present, it is believed that statistics is a suitable method to analyse networks with millions, or more, of vertices. The MATLAB language, with its mass of statistical functions, is a good choice to rapidly realize an algorithm prototype of complex networks. The performance of the MATLAB codes can be further improved by using graphic processor units (GPU). This paper presents the strategies and performance of the GPU implementation of a complex networks package, and the Jacket toolbox of MATLAB is used. Compared with some commercially available CPU implementations, GPU can achieve a speedup of, on average, 11.3×. The experimental result proves that the GPU platform combined with the MATLAB language is a good combination for complex network research.     

Quantum machines

We discuss quantum information processing machines. We start with single purpose machines that either redistribute quantum information or identify quantum states. We then move on to machines that can perform a number of functions, with the function they perform being determined by a program, which is itself a quantum state. Examples of both deterministic and probabilistic programmable machines are given, and we conclude with a discussion of the utility of quantum programs.     

A polynomial algorithm for some preemptive multiprocessor task scheduling problems

In this paper we consider a problem of preemptive scheduling of multiprocessor tasks on dedicated processors in order to minimize the sum of completion times. Using a standard notation, our problem can be denoted as P ∣ fix_j, pmtn ∣ ∑C_j. We give a polynomial-time algorithm to solve P ∣ fix_j, G = {P₄, dart}-free, pmtn ∣ ∑C_j problem. This result generalizes the following problems: P2 ∣ fix_j, pmtn ∣ ∑C_j, P ∣ ∣fix_j∣ ∈ {1, m}, pmtn ∣ ∑C_j and P4 ∣ fix_j = 2, pmtn ∣ ∑C_j.     

Ultrasound phase rotation beamforming on multi-core DSP

Phase rotation beamforming (PRBF) is a commonly-used digital receive beamforming technique. However, due to its high computational requirement, it has traditionally been supported by hardwired architectures, e.g., application-specific integrated circuits (ASICs) or more recently field-programmable gate arrays (FPGAs). In this study, we investigated the feasibility of supporting software-based PRBF on a multi-core DSP. To alleviate the high computing requirement, the analog front-end (AFE) chips integrating quadrature demodulation in addition to analog-to-digital conversion were defined and used. With these new AFE chips, only delay alignment and phase rotation need to be performed by DSP, substantially reducing the computational load. We implemented the delay alignment and phase rotation modules on a Texas Instruments C6678 DSP with 8 cores. We found it takes 200 μs to beamform 2048 samples from 64 channels using 2 cores. With 4 cores, 20 million samples can be beamformed in one second. Therefore, ADC frequencies up to 40 MHz with 2:1 decimation in AFE chips or up to 20 MHz with no decimation can be supported as long as the ADC-to-DSP I/O requirement can be met. The remaining 4 cores can work on back-end processing tasks and applications, e.g., color Doppler or ultrasound elastography. One DSP being able to handle both beamforming and back-end processing could lead to low-power and low-cost ultrasound machines, benefiting ultrasound imaging in general, particularly portable ultrasound machines.     

Almost nonpreemptive schedules

Graph-theoretical models are described for solving preemptive and nonpreemptive scheduling problems with renewable resources. Conditions are obtained for nonpreemptive schedules to exist. These results may be applied for reducing the preemptions in the schedules obtained by the two-phase method developed for preemptive scheduling on unrelated processors.     

基于多核点共享树的多源光组播路由方法

通过网络编码方法优化多核点选择和组播信息传输,本文提出一种基于多核点共享树和网络编码的光组播路由构造和波长分配方法、减少波长资源消耗和提高网络的负载平衡性能.首先,删除产生源点迂回回路的网络编码备选核点集合,采用启发式矩阵运算方法确定多源共享树的网络编码核点,实现多源共享树以最少的核点覆盖最多的源节点;然后,为减少波长信道消耗数目,在确定的核点到目的节点间加入网络编码方法传输信息;最后,讨论了多核点共享树的波长分配方法和目的节点成功解码的边分离路径方法.仿真结果表明:与单核共享树、基于网络编码的单核共享树相比,基于网络编码的多核点共享树组播路由方法需求最少的波长数目和获得最好的网络负载平衡性能.     

OFDM水声通信实时处理系统的多处理器并行

OFDM(正交频分复用)充分利用频带,具有较高的数据传输速率;ADSP-TS101是具有高速并行处理能力的高性能数学信号处理芯片。针对OFDM水声通信的实时处理要求,本文设计并研制了具有实时处理能力的水声通信系统。该系统包括由多片ADSP-TS101构成的松/紧联合耦合的并行处理结构作为处理核心、无相位偏差的多通道同步采样模块和FPGA(现场可编程门阵列)逻辑控制模块。湖上实验结果表明该系统具有良好的稳定性和实时处理能力,满足工程设计要求。     

Contrast increasing by third-order nonlinear image processing: a numerical study for microscopic rectangular objects

This study deals with the simulation of the image formation involved by using a nonlinear image processor composed of a nonlinear medium placed at the common focus of a 4-f system. A simple linear model is developed taking into account the third-order optical nonlinearity of the medium (in the general case of nonlinear absorption and dephasing) and the optical transfer function of the 4-f system. The simulated images through the system are given for phase and/or amplitude rectangular objects. Our model and its corresponding simulation lead to optimizing nonlinear parameters of the medium for this class of objects, whatever their transmittance is, in order to enhance the visibility in image processors.