在TMS320C6x系列DSP上实现G．729算法

在研究ITU-T的G．729CS-ACELP语音编码算法基础上，本文介绍了使用TI公司的TMS320C6x系列DSP实现该算法的一些问题以及一些应用中的具体优化技术，这在一定程度上降低了该算法的计算复杂度，而输出语音仍然保持了很高的合成品质。然后，本文讨论了如何在一个电信级 的应用内核上对该语音编码算法进行多声道扩展的问题，该多通道编码系统基本能达到实用要求。最后，在实时MCPS、内存要求、处理延时和实时性能等方面对该DSP实现的编码器做了一个评价，可以作为其他语音编码器开发的参考。     

改进的零树编码算法及其DSP实现

基于双正交叠式变换(LBT)的卫星遥感图像数据压缩,需要一种低内存和低复杂度的编码方法.以SPIHT(基于分层树的集合分割排序的编码算法)为基础,通过研究LBT系数的结构,提出了一种改进的零树编码算法.该算法采用了零树结构的快速检验和线性索引技术.本文对该算法在TMS320C6000系列DSP中的实现进行了研究,通过任务合并、优化选项、软件流水、循环展开和多功能指令等方法提高处理速度.该算法已在某侦查卫星数据压缩样机中实现.实验结果表明,该算法内存需求低,压缩质量接近SPIHT,并且计算复杂度大大降低.     

基于DSP技术的H.263+视频编码协议的实现

采用TI公司的TMS320C6201评估板来实现H．263＋编码。首先采用快速算法降低编码的复杂度。然后根据TMS320C6201的结构和C编译器的特点进行C语言程序的优化，来达到降低芯片内存和加快处理速度的目的。     

Oracle性能优化技术在装备信息编码管理系统中的应用

装备信息编码管理系统采用Oracle数据库对装备数据进行分类、编码等标准化工作。由于该系统数据吞吐量比较大,对数据库性能要求比较高。在参与装备信息编码管理系统的开发过程中,从数据库的创建、内存使用和SQL语句优化等方面对数据库进行了优化,并取得了很好的效果。本文从SGA、数据排序、回退段和SQL语句四个方面简要介绍在EICMS中对数据库的优化工作。     

G.729A语音编码算法DSP优化与高速实现

提出了一种将G．729A语音编码算法在TMS320C55xDSP上高效实现的方法，并根据C55x系统结构提供的特性，通过使用双乘加运算、指令并行、循环展开、C55x的专用指令等方法对算法作了高质量的优化，优化实现后的G．729A的运算速度是8．76MCPS，需要15．2kw的程序空间和3．2kw的数据空间，实验结果证明本方法具有运算效率高、代码量少等特点，文中提出的一系列优化方法同样适用于基于C55xDSP等芯片系列其它代码的优化．     

适用于硬件实现的零树编码算法

在Listless Zerotree Coding(LZC)算法的基础上提出了一种内存空间更小的静止图像零树编码算法，该算法与SPIHT等经典的零树编码算法的区别在于编解码过程中不使用列表结构，与LZC算法相比则节约了两个位图(Bit Map)，而这些位图为编码算法提供了重要系数和重要后代集合的位置信息．实验结果证明，本文的算法在不降低图像度量的前提下大大降低了零树编码算法的内存要求，有利于硬件实现．     

位平面算术编码的编码分离与MQ编码器改进

为了降低图像压缩中位平面算术编码的计算复杂度,提出了将算术编码从位平面扫描中分离,方便编码的并行计算。以JPEG2000的位平面编码为例,实现了位平面扫描和MQ编码的分离,然后根据编码分离后的算法特点,设计了一种改进的MQ编码器。对该算法在TMS320C6000系列DSP中的实现进行了研究,通过软件流水、循环展开、条件操作和优化选项等方法提高处理速度。编码独立和编码器改进相结合,使重建图像质量稍有提高,编码速度提高了8倍以上。     

G.729语音编码器在DSP上的实时实现

为满足音视频的同时需求,提出了基于TMS320C64X系列DSP的G.729语音编码器软、硬件设计方案,并着重阐述了在实时实现过程中进行优化的关键技术。经测试表明,优化后的单路语音编码占用了5%左右的CPU资源,比优化前降低了80%,保留了更多的资源用于视频编码,为在同一块DSP芯片上实现音频、视频编码提供可能。     

面向内存的混合容错编码动态调节设计

针对内存系统中高强度的容错编码容易造成过大开销的问题,为同时实现容错强度和容错开销的权衡,提出一种低开销的支持混合容错编码的动态调节设计.通过分析发现常见纠错检错编码数据位长与校验位长存在固定的比例关系,提出一种地址映射逻辑电路;当系统存取内存数据及容错强度发生调节时,该方法可保证容错编码中校验信息的存取,实现对内存容错强度调节以及数据与校验信息在内存中分开存储的支持.实验结果表明,文中设计简单,硬件和性能代价小、功耗开销低.     

遗传算法的编码理论与应用

编码是遗传算法求解问题的前提,文章分析了二进制编码、格雷码编码、实数编码、符号编码、排列编码、二倍体编码、DNA编码、混合编码、二维染色体编码或矩阵编码等编码的实质内容,在树编码和可变长编码基础上阐述了自适应编码的基本理论,提出了基于相似度的可变长编码和基于结构的agent编码方式,给出了函数优化、TSP、KP、JSP、机器人路径规划、图的划分和倒立摆等典型优化问题的编码方案。     

Characterizing History Independent Data Structures

We consider history independent data structures as proposed for study by Naor and Teague. In a history independent data structure, nothing can be learned from the memory representation of the data structure except for what is available from the abstract data structure. We show that for the most part, strong history independent data structures have canonical representations. We provide a natural alternative definition of strong history independence that is less restrictive than Naor and Teague and characterize how it restricts allowable representations. We also give a general formula for creating dynamically resizing history independent data structures and give a related impossibility result.     

无线传感器网络中对等数据存储策略研究

在无线传感器网络中,除了基于基站的数据传输和管理模式以外,对等数据传输以及网内数据管理同样具有非常重要的现实意义。本文论述了传感器网络中对等数据传输的应用特点和研究挑战,以及数据存储策略与对等数据传输的关系。如何设计有效的数据存储策略是对等数据传输和网内数据管理、查询的关键技术和重要研究课题。本文分析了无线传感器网络中对等数据存储策略的分类、负载平衡和查询成功率等相关问题。本文采取基于位置的分类方法,根据策略是否依赖节点的位置信息,将节点间的对等数据存储分为基于位置的策略和无位置策略。本文综述了目前的相关研究,比较了各种策略适用的网络环境和数据查询成功率、网络负载平衡等性能指标。     

Enabling collaborative engineering and science at JPL

We examine the issue of distributed collaboration at the Jet Propulsion Laboratory (JPL). With the goals of “faster, better, cheaper” missions, an efficient, seamless collaboration capability is critical to future JPL space exploration missions. We assert that the current capabilities for distributed collaboration internal to JPL, as well as external, are unsatisfactory. This paper provides a vision of greatly enhanced distributed collaboration capabilities in the near future and into the far future. Our vision focuses primarily on distributed collaborative engineering and science.We believe enhanced capabilities are necessary in two collaborative paradigms: the virtual conference and the shared virtual workspace. We describe these collaborative paradigms, and discuss the tools that exist and the additional capability necessary to achieve the collaborative vision with respect to the types of data we typically work with at JPL. We identify the following types of data frequently exchanged in collaborative activities: project planning data, design data, notes/documentation, communication data, analysis/performance data, verification data and scientific data. Our analysis shows that at present there is good GroupWare support for project planning data and notes/documentation data. Support is improving for design data. Other data types have no or sporadic support at best.     

On Bandwidth-Efficient Data Broadcast

In this paper, we leverage network coding to reduce the bandwidth consumption for data broadcast. We use the concept of mixing in a different way from the traditional network coding. Traditional network coding mixes all data items together. However, we mix each data item from only some data items according to the queried and stored data of receivers. Therefore, each receiver in our approach is required to receive fewer coded data to decode the required information, and the sender thereby is able to broadcast fewer data items. We formulate an optimization problem with integer-linear programming to minimize the bandwidth consumption in data broadcast. We prove that the problem is NP-hard and design an approximation algorithm that can be implemented in the data server. In addition, we show that different mixings of data items lead to different decoding costs for receivers. We design an algorithm to optimally code the data items with the minimum decoding cost.     

A new approximate query engine based on intelligent capture and fast transformations of granulated data summaries

We outline the processes of intelligent creation and utilization of granulated data summaries in the engine aimed at fast approximate execution of analytical SQL statements. We discuss how to use the introduced engine for the purposes of ad-hoc data exploration over large and quickly increasing data collected in a heterogeneous or distributed fashion. We focus on mechanisms that transform input data summaries into result sets representing query outcomes. We also illustrate how our computational principles can be put together with other paradigms of scaling and harnessing data analytics.     

Hilbert填充曲线与空间分布模式探测结合的点数据集空间划分方法

空间数据划分是空间索引、并行GIS数据分解以及分布式数据管理与调度等问题的核心环节之一。本文针对点数据集多目标空间划分问题,在Hilbert空间填充曲线的基础上建立随机、规则、聚集三类空间分布模式的判定模型,设计针对均匀和随机分布模式的等差划分以及针对聚集分布模式的迭代寻优划分。实验表明该方法能够在缺少覆盖范围信息的条件下准确判定空间分布类型,该方法能够兼顾空间聚集性、数据量均衡与空间重叠度三种约束条件。     

数据挖掘过程的多维视图

数据挖掘（DM）是非常具有挑战性的工作,数据挖掘过程是多个因素耦合的决策问题。讨论了当前流行的DM过程CRISPDM和SEMMA的不同之处及优缺点。从机器学习、统计和数据质量角度对挖掘有效性作了讨论,认为一个真正高效的过程应该面向算法,强调探索,以挖掘出高可靠性的具有商业价值的知识目标,并紧跟技术的发展。给出数据挖掘过程的多维视图,将算法分解为组件维、模型维和过程维等维度,以此提出新的DM过程的框架。     

一个数据共享系统的实现

针对一个真实的医疗信息数据共享系统的需求,结合用户需求和数据特点确定数据仓库的主题,采用了数据采集、数据抽取、联机分析处理等技术,并且通过MQ技术实现了数据交换机制,具体的实施则是通过结合一些开源工具的有效使用,最终建立了一套架构先进、功能完善的业务数据共享信息平台.     

Clustering Time Series with Clipped Data

Clustering time series is a problem that has applications in a wide variety of fields, and has recently attracted a large amount of research. Time series data are often large and may contain outliers. We show that the simple procedure of clipping the time series (discretising to above or below the median) reduces memory requirements and significantly speeds up clustering without decreasing clustering accuracy. We also demonstrate that clipping increases clustering accuracy when there are outliers in the data, thus serving as a means of outlier detection and a method of identifying model misspecification. We consider simulated data from polynomial, autoregressive moving average and hidden Markov models and show that the estimated parameters of the clipped data used in clustering tend, asymptotically, to those of the unclipped data. We also demonstrate experimentally that, if the series are long enough, the accuracy on clipped data is not significantly less than the accuracy on unclipped data, and if the series contain outliers then clipping results in significantly better clusterings. We then illustrate how using clipped series can be of practical benefit in detecting model misspecification and outliers on two real world data sets: an electricity generation bid data set and an ECG data set.