基于FAHP的hadoop平台移动终端云存储优化研究

Hadoop云存储架构的设计初衷是实现大文件的高效存储处理,但在处理移动终端下诸如图片等小文件时会引起名称节点索引提取速度过慢和数据节点存储空间利用率不高等问题.针对这一问题,提出一种小文件归档的方案FHAR.方案综合考虑移动终端用户访问的实时性、名称节点服务器内存使用率、数据节点存储空间利用率等方面,利用双层索引的归档技术结合FAHP（模糊多属性决策理论）的系统负载预测算法实现系统的负载均衡,提高服务效率.同时利用数据预取机制对访问操作进行优化.仿真结果表明,该方案有效提高了节点的存储效率与用户访问的实时体验性.     

基于XML信息检索技术的研究

基于XML信息检索技术首先要解决的是建立一个比较高效的索引机制,索引是实现快速查询的重要保证。在介绍了两种基于关系数据库索引技术的基础上,提出一种新的倒排索引技术,并在查询效率和存储开销两个方面进行对比分析,通过实验数据说明新索引技术可取得查询效率和存储开销的最佳权衡。     

基于压缩近邻的查重元数据去冗算法设计

随着重复数据删除次数的增加,系统中用于存储指纹索引的清单文件等元数据信息会不断累积,导致不可忽视的存储资源开销。因此,如何在不影响重复数据删除率的基础上,对重复数据删除过程中产生的元数据信息进行压缩,从而减小查重索引,是进一步提高重复数据删除效率和存储资源利用率的重要因素。针对查重元数据中存在大量冗余数据,提出了一种基于压缩近邻的查重元数据去冗算法Dedup2。该算法先利用聚类算法将查重元数据分为若干类,然后利用压缩近邻算法消除查重元数据中相似度较高的数据以获得查重子集,并在该查重子集上利用文件相似性对数据对象进行重复数据删除操作。实验结果表明,Dedup2可以在保持近似的重复数据删除比的基础上,将查重索引大小压缩50%以上。     

Gzip压缩的硬件加速电路设计

硬件无损压缩技术可以发挥专用电路的速度和功耗优势,被广泛应用于大数据计算以及通信领域.本文以GNUzip（Gzip）数据无损压缩技术为原型设计了一种硬件压缩电路.通过采用双Hash函数、并行匹配处理、面向硬件存储的LZ77压缩存储格式、高效数据拼接器等加速方法,发挥并行计算和流水线结构优势,提升压缩速率.该硬件压缩电路基于Verilog HDL设计,使用现场可编程门阵列（FPGA）进行测试和验证.测试数据表明：与软件压缩方式相比,该硬件压缩电路在获得适中压缩率（65.9%）的同时,其压缩速率得到显著提升,平均压缩速率达171Mb/s,满足网络通信、数据存储等实时压缩应用需求.     

数字视频存储的优化策略

存储是播出/媒体资产管理设施建立的基础.归档和提取服务是建立在存储基础上的框架.随着技术的进步,磁盘和磁带容量不断增加,速度不断提高,而成本则不断降低.在未来很多年内,视频存储体系结构将维持现状:视频服务器的磁盘数字磁带库归档存储;磁盘和数据磁带存储将是归档存储解决方案的基础.     

大规模数据集的分布式索引机制研究

对大规模结构化半结构化数据集分布式存储环境下索引结构的建立进行了研究,提出了一种基于聚簇索引与分布式B 树的混合索引结构.该结构对数据集的输入,存储及查询等操作进行了定义,并分析了索引服务器与客户端的通信模型建立.由于对记录的主关键字与其他属性进行了不同处理,在各种记录检索情况下具有良好的性能.     

大数据环境下网络信息归档与管理研究

计算机技术的普及使得人类开始进入到大数据时代,越来越多的资料通过网络进行存储,并采用计算机进行归档和管理,提升了办公整体效率,提升了办公数据的处理质量。但是大数据环境下网络信息的归档还是存在一些问题,在数据的保管和归档方法方面存在一些不足,因此需要对网络信息归档和管理进行更加全面的研究。文章主要针对大数据环境下网络信息归档与管理现状与对策进行研究。     

基于航天大数据集的相关挖掘与空间态势可视化

针对海量航天数据多源、异构的特点,为了解决数据和信息存储与可视化等问题,首先通过满FP树来对动态数据库中的增量数据条目进行更新,然后进行特性挖掘、重新关联,提高航天信息数据中的关联存储性能,增强不同数据源的数据交互、特征共享能力。实验结果表明,通过基于拉姆达架构的空间态势数据分类处理和归档等技术,对航天空间态势目标高效可视化能力有明显的提高。     

发布/订阅系统中的历史数据分布式存储算法

在分析当前发布/订阅系统在历史数据存储方面相关工作的基础上,提出了一种利用系统中的存储代理对历史数据进行分布式存储的算法.算法将存储代理组成覆盖网,利用一致性哈希定位主题元数据位置,主题数据分段存储在存储设备上,并采用文件索引、并行读取等方法来提高数据存取性能.在对比实验中,验证了算法在负载均衡和扩展性方面的优势.     

嵌入无损编码的海量视频数据存储系统设计

随着当代视频显示分辨率的不断提升,视频显示系统需要实时缓存的数据量越来越大.DDR等实时存储设备受到自刷新率等存储模式的限制,难以满足视频显示系统实时稳定存储的需要.海量视频数据的实时存储成为显示系统亟待解决的关键问题.根据高清视频单帧图像内像素间的空间相关性高的特点,提出了一种内嵌算术编码及哈夫曼编码的嵌入式混合压缩数据存储方法.在传统的视频高速DDR存储控制系统中,首先对单帧视频进行压缩,通过直接减少系统的实时存储量,解决海量视频存储瓶颈的问题.经过试验测试证明,该混合压缩数据存储方法不仅减少了所需存储的数据量,且其数据还原与经典的熵编码比较压缩还原后PSNR值平均高出1～2 dB.     

业务支撑系统灾备方案

随着数据的积累和增长,目前企业面临的最大挑战之一是如何处理对大量数据的透明存取,同时网络的规模越大、计算机应用的越深入、数据量越多,数据的安全对于企业的运转也越来越显得重要。这就提出了数据存储管理方面的要求。数据存储管理系统主要实现全自动的数据备份和恢复,并通过定期对历史数据进行归档处理（将某类具有特定意义的数据永久保留到存储介质上,以备今后查询及决策）,确保系统内有足够的硬盘空间用于后续数据的存储。此外通过对用于数据存储的媒体进行自动的、高效的管理,使得“Light-out”的数据管理模式得以真正实现。     

基于激光三维视觉的船舶靠泊动态监测技术

随着船舶运输业的不断发展,船舶趋向大型化,船舶靠泊风险增加,为了保证船舶靠泊更加安全、高效,有必要在码头配备靠泊监测系统。本文分析了船舶靠泊过程,研究了三维激光扫描仪工作原理和三维点云数据的处理方法,提出了基于激光三维视觉的船舶靠泊监测技术,提取船舶靠泊时船首、船尾相对码头的距离与速度等船舶动态靠泊参数,解决了以往点式激光靠泊监测系统寻找目标能力差的问题,增强了应对不同泊位、船型的适用性,实例验证表明此靠泊技术具有可行性和实用性,可为船舶靠泊提供安全保障。     

SS/TDMA time-slot assignment with knowledge-base and prolog implementation

A software production system for the generation of time slot assignment plans for SS/TDMA systems is proposed and implemented. The approach used in this system conveniently combines algorithms derived for ideal configurations with heuristic procedures that are needed to account for conditions present in real systems. Algorithms that produce plans with a minimum number of switchings and minimum traffic transmission duration can now be easily implemented. The assignment plans obtained are as efficient as those previously reported when the assumptions are the same. The proposed production system, however is capable of handling a much larger variety of constraints and heuristic procedures for which no efficient solutions have been reported.     

An Adaptive Power Control Scheme for Multicast Service in Green Cellular Railway Communication Network

Along with emerging increasing multimedia service need from the big amount of users in the train, wireless cellular railway network has big energy saving potential with its own specific characteristics, which is widely deployed over the world. Enhanced Multimedia Broadcast Multicast Service (E-MBMS) is a prevalent solution for high speed data handling. In this paper, to achieve an energy efficient solution for wireless cellular railway network, we proposed an optimal power control solution to adjust eNodeB’s transmission power adaptively based on actual Signal to Interference plus Noise Ratio (SINR) for Multimedia Broadcast multicast service Single Frequency Network (MBSFN) at User Equipment (UE) side when provide E-MBMS. We established the system model and proposed the optimization algorithms to calculate the E-MBMS transmit power. We also designed the geometry group scheme to significantly reduce the calculation effort. And proposed the Mobility offset parameter to cope with high speed mobility of the train and variation of signal level. Finally, Simulation results showed that this solution can be a good way for high network energy efficiency and can dynamically adjust the output power for the multicast service in the wireless cellular railway communication network.     

Efficient algorithms for multilevel power estimation of VLSI circuits

This paper presents a methodology for calculating highly accurate mean power estimates for integrated digital CMOS circuits. A complementary calibration scheme for ASIC library cells to extract the power relevant parameters is proposed. The circuit models presented allows the prediction of mean power dissipation of gate-level designs in CMOS technologies with an accuracy that is comparable to a SPICE simulation but up to 10 000 times faster. The outlined approach is capable of handling complex circuits consisting of more than 20 000 cells and thousands of memory elements. Very large sets of input data with several millions of patterns can, thus, be simulated in an efficient way. This allows the prediction of mean power dissipation of VLSI circuits in a realistic functional context which provides new assessment possibilities for digital CMOS low-power design methods. Experimental results for some benchmark circuits are detailed in order to demonstrate the significant improvements in terms of performance, accuracy, and flexibility of this approach compared to state-of-the-art power estimation methods     

一种基于RFM模型的新型协同过滤个性化推荐算法

摘要：为了提高个性化推荐效果及预测准确度,特别是针对传统算法中评分矩阵过于稀疏等问题提出一种新颖的协同过滤算法。该算法首先利用RFM模型合理地筛选用户信息,其次通过黏性客户的消费记录稠密化用户—项目评分矩阵,并改进了传统相似度计算公式。通过仿真实验证实了算法的准确性,最后将其应用于一套具有个性化商品推荐功能的系统原型中,证明了该推荐算法的有效性及实用性。     

Proposal for an Inundation Hazard Index of Road Links for Safer Routing Services in Car Navigation Systems

Inundation of roads by heavy rainfall has attracted more attention than traffic accidents, traffic congestion, and construction because it simultaneously causes travel delays and threatens driver safety. For these reasons, in this paper, we propose an inundation hazard index (IHI) of road links, which shows the possibility of inundation of road links caused by rainfall. To generate the index, we have used two key data sources, namely the digital elevation model (DEM) and past rainfall records of when inundation has occurred. IHI is derived by statistically analyzing the relationships between the normalized relative height of the road links calculated from DEM within the watershed and past rainfall records. After analyzing the practical applicability of the proposed index with a commercial car navigation system through a set of tests, we confirmed that the proposed IHI could be implemented to choose safer routes, with reduced chances of encountering roads having inundation risks.s     

A contention/reservation access protocol for speech and data integration in TDMA‐based advanced mobile systems

The performance of third generation mobile systems is greatly influenced by the multiple access protocols used in the radio access system. The paper introduces a multiple access protocol, SIR (Service Integration for Radio access), which has the potential for accommodating the requirements of speech and bursty data traffic in an efficient way. SIR is evolved from an access protocol (PRMA++) studied within the framework of the TDMA-based version of the European evolving standard for third generation mobile systems. In particular, SIR uses the same frame structure and in-band signalling but introduces a contention-free handling of data bandwidth requests while meeting speech service requirements via basic PRMA++ mechanisms. This revised version was published online in June 2006 with corrections to the Cover Date.     

Virtual view synthesis using layered depth image generation and depth-based inpainting for filling disocclusions and translucent disocclusions

View synthesis is an efficient solution to produce content for 3DTV and FTV. However, proper handling of the disocclusions is a major challenge in the view synthesis. Inpainting methods offer solutions for handling disocclusions, though limitations in foreground-background classification causes the holes to be filled with inconsistent textures. Moreover, the state-of-the art methods fail to identify and fill disocclusions in intermediate distances between foreground and background through which background may be visible in the virtual view (translucent disocclusions). Aiming at improved rendering quality, we introduce a layered depth image (LDI) in the original camera view, in which we identify and fill occluded background so that when the LDI data is rendered to a virtual view, no disocclusions appear but views with consistent data are produced also handling translucent disocclusions. Moreover, the proposed foreground-background classification and inpainting fills the disocclusions with neighboring background texture consistently. Based on the objective and subjective evaluations, the proposed method outperforms the state-of-the art methods at the disocclusions.