高性能计算与天文大数据研究综述

数据是天文学发展的重要驱动。分布式存储和高性能计算(High Performance Computing,HPC)为应对海量天文数据的复杂性、不规则的存储和计算起到推动作用。天文学研究中多信息和多学科交叉融合成为必然,天文大数据已进入大规模计算时代。高性能计算为天文大数据处理和分析提供了新的手段,针对一些传统手段无法解决的问题给出了新的方案。文中根据天文数据分类和特征,以高性能计算为支撑,对天文大数据的数据融合、高效存取、分析及后续处理、可视化等问题进行了研究,总结了现阶段的技术特点,提出了处理天文大数据的研究策略和技术方法,并对天文大数据处理面对的问题和发展趋势进行了探讨。

基于胶囊网络模型的抑郁症预测研究

抑郁症已经成为世界第四大疾病,并且具有高发病率,由于精神问题的诊断十分困难,抑郁症的治疗率很低。随着互联网的发展,社交媒体被越来越多的人使用,人们倾向于在社交媒体上表达自己的观点,这就为研究者创造了大量可以使用的数据,但是抑郁症预测不等同于文本分类,存在难以找到判定抑郁的统一标准,以及难以构建神经网络模型两个问题。对此,提出了一种融合局部和整体特征的胶囊网络模型,该模型将情绪词典和胶囊网络进行融合,可以有效地发现微博用户中潜在的抑郁症患者。模型分为局部特征选择和整体特征提取两部分,对于局部特征通过情绪词典进行选取,对于整体特征则使用胶囊网络模型进行学习,将两部分特征进行融合,得到该微博用户的抑郁症预测概率。实验结果表明,提出的模型相较于用于文本分类的模型KNN、DNN、TextCNN和BiRNN等,在针对微博文本的抑郁症预测方面具有较高的准确率,能够有效地识别微博用户中潜在的抑郁症患者。

基于新颖性检测的跌倒风险预测

跌倒是65岁及以上人群因伤害致死的第一位原因.结合受试者个体信息的个性化特征,提出一种基于Kinect三维骨架数据的步态特征提取方法,对老年人的跌倒风险进行评估和预测.将跌倒风险分为高跌倒风险和低跌倒风险两类,考虑数据采集的成本问题,采用新颖性检测模型在不平衡数据集下对特征数据进行训练和评估.实验结果表明,OC-SVM(one-class SVM)检测准确率达86.96%,F1-score为88.55%,能够有效地区分低跌倒风险受试者和高跌倒风险受试者.同时,证明了基于Kinect三维骨架数据预测老年人跌倒风险的潜力.

新工科背景下大学生计算机素质教育研究

本文结合新工科的建设内涵,对新工科背景下计算机素质进行研究,探索了新工科背景下计算机素质教育的培养模式。

Multi-channel Assignment in Multi-hop Wireless Network

Multiple channels are available for use in IEEE 802.11.Multiple channels can increase the available network's capacity,and how to efficiently assign these available channels to optimize the network performance is a challenge.We survey current techniques to solve the problem,and category the techniques single-radio,multi-radio,cross-layer multi-channel assignment.This paper also discusses some interesting issues at last.

A Survey of Secure Routing in Wireless Sensor Networks

We analyze and summarize the literatures on secure routing protocols published since 2003.Firstly,threats and attacks on sensor network routing are sorted and summarized.Then the current secure routing protocols are also classified.According to the taxonomy,some typical secure routing protocols are clarified particularly.Besides,the advantage and disadvantage of these secure routing protocols are compared and analyzed.Finally,we conclude this paper and prospect the future work.

Using Kalman Filter to Guarantee QoS Robustness of Web Server

The exponential growth of the Internet coupled with the increasing popularity of dynamically generated content on the World Wide Web, has created the need for more and faster Web servers capable of serving the over 100 million Internet users. To converge the control method has emerged as a promising technique to solve the Web QoS problem. In this paper, a model of adaptive session is presented and a session flow self regulating algorism based on Kalman Filter are proposed towards Weh Server. And a Web QoS self-regularing scheme is advanced. To attain the goal of on-line system identification, the optimized estimation of QoS parameters is fulfilled by utilizing Kalman Filter in full domain, The simulation results shows thal the proposed scheme can guarantee the QoS with both robusmess and stability .

均匀化处理温度对6014铝合金组织和性能的影响

采用硬度和电导率测试,扫描电子显微镜观察手段,研究均匀化处理温度对6014铝合金组织和性能的影响。研究结果表明,随着均匀化处理温度的升高,6014合金铸态组织中非平衡相逐渐溶解,针状β-AlFeSi相逐渐向颗粒状α-AlFeSi相发生转变,晶内和晶界黑色颗粒状MgSi相逐渐回溶。随着均匀化处理温度的升高,电导率呈下降趋势。均匀化处理温度越高,合金硬度呈先升高后降低的趋势,均匀化处理温度为560℃时硬度最大,达到71.1HB。

Node correlation clustering algorithm for wireless multimedia sensor networks based on overlapped FoVs

In this paper, a clustering algorithm is proposed based on the high correlation among the overlapped field of views for the wireless multimedia sensor networks. Firstly, by calculating the area of the overlapped field of views （FoVs） based on the gird method, node correlations have been obtained. Then, the algorithm utilizes the node correlations to partition the network region in which there are high correlation multimedia sensor nodes. Meanwhile, in order to minimize the energy consumption for transmitting images, the strategy of the cluster heads election is proposed based on the cost estimation, which consists of signal strength and residual energy as well as the node correlation. Simulation results show that the proposed algorithm can balance the energy consumption and extend the network lifetime effectively.

Novel congestion control approach in wireless multimedia sensor networks

Data generated in wireless multimedia sensor networks （WMSNs） may have different importance and it has been claimed that the network exert more efforts in servicing applications carrying more important information. Nevertheless, importance of packets cannot generally be accurately represented by a static priority value. This article presents a dynamic priority based congestion control （DPCC） approach that makes two major innovations in WMSNs. First, DPCC employs dynamic priority to represent packet importance. Second, it prioritizes the local traffic of motes near the base station when WMSN is highly congested. Simulation results confirm the superior performance of the proposed approach with respect to energy efficiency, loss probability and latency as well.

MiniTE:data acquisition routing for wireless sensor networks

Routing is one of the most important supporting parts in wireless sensor networks （WSNs） application that directly affects the application efficiency. Routing time and energy consumption are two major factors used to evaluate WSNs routing. This article proposes a minimum routing time and energy consumption （MiniTE） routing, which can ensure feasibility of the routing protocol both in routing time and energy consumption. Based on the MiniTE, WSNs can be partitioned into different regions according to the received signal strength indication （RSSI）. Messages are sent by nodes in the region to their parent node and again up to their parent node until finally to the sink node. Theoretic evaluation and simulation results are given to verify the features of the protocol.

A Solution of Grid Computing Flow Using MDA Methodology

This paper presents an approach to build a flow of grid services named GFlow. This approach converts all of the services＇ interfaces in globus toolkit 3 into platform independent model format. The conversion uses eclipse modeling framework plug-ins and graphical editing framework plug-ins to parse the models into platform specific models, which are responsible for implementation of grid services＇ clients and servers and their automatic deployment. This paper emphasizes on design method, which uses model drive architecture, and design process, which uses model-view-control.

Improved multipath routing with WNN for the open networks

Multipath routing mechanism is vital for reliable packet delivery, load balance, and flexibility in the open network because its topology is dynamic and the nodes have limited capability. This article proposes a new multipath switch approach based on traffic prediction according to some characteristics of open networks. We use wavelet neural network （WNN） to predict the node traffic because the method has not only good approximation property of wavelet, but also self-learning adaptive quality of neural network. When the traffic prediction indicates that the primary path is a failure, the alternate path will be occupied promptly according to the switch strategy, which can save time for the switch in advance The simulation results show that the presented traffic prediction model has better prediction accuracy; and the approach based on the above model can balance network load, prolong network lifetime, and decrease the overall energy consumption of the network.

QoS-aware replica placement for data intensive applications

In recent years, with the rapid development of data intensive applications, data replication has become an enabling technology for the data grid to improve data availability, and reduce file transfer time and bandwidth consumption. The placement of replicas has been proven to be the most difficult problem that must be solved to realize the process of data replication. This paper addresses the quality of service （QoS） aware replica placement problem in data grid, and proposes a dynamic programming based replica placement algorithm that not only has a QoS requirement guarantee, but also can minimize the overall replication cost, including storage cost and communication cost. By simulation, experiments show that the replica placement algorithm outperforms an existing popular replica placement technique in data grid.

An agent-based incentive mechanism for P2P systems

With the increasing systems, the problems of fair uses of peer-to-peer （P2P） information and resources sharing become serious. P2P systems are self-organizing, distributed systems with no centralized authority, such as Free-riding and Tragedy of Commons. Because of the voluntary participation and lack of motivation, the information and resources available in P2P networks are extremely variable and unpredictable. This article studied the problems existing in P2P networks and propose a different method to stimulate the peers in P2P networks to share information and resources, using agents so as to improve the performance of P2P networks And through simulation experiments and analyzing the results, it can be concluded that this mechanism can effectively solve the problems of fair sharing in P2P systems.

Research on tasks schedule and data transmission of video sensor networks based on intelligent agents and intelligent algorithms

As a novel application technology,wireless video sensor networks become the current research focus,especially on target tracking and surveillance scenario.Based on multiple agents＇ technique,this article introduces a series of intelligent algorithms such as simulated annealing algorithm（SA）,genetic algorithm（GA）,and ant colony optimization algorithm（ACO） or their mixed algorithms,to resolve the optimization of tasks schedule and data transmission.This article analyzes the performance of abovementioned algorithms and verifies their feasibility associated with agents.The simulations demonstrates that the mixed algorithms based on SA and GA obtain the optimal solution to tasks schedule,and those combined with SA-ACO show advantages on multimedia sensor networks routing optimization.

Novel analysis and improvement of Yahalom protocol

The modified version of Yahalom protocol improved by Burrows, Abradi, and Needham （BAN） still has security drawbacks. This study analyzed such flaws in a detailed way from the point of strand spaces, which is a novel method of analyzing protoeol＇s security. First, a mathematical model of BAN-Yahalom protocol is constructed. Second, penetrators＇ abilities are restricted with a rigorous and formalized definition. Moreover, to increase the security of this protocol against potential attackers in practice, a further improvement is made to the protocol. Future application of this re-improved protocol is also discussed.

Design and Realization of Software for Guard Against DDoS Based on Self-Similar and Optimization Filter

This paper proposes a distributed denial-of-service attack detection method based on self similar and wavelet analysis. This method adopts an optimized transmission control protocol cookie technology for filter optimization in order to accurately detect and efficiently filter the traffic of distributed denial-of-service attack. This paper presents the design of our software, and describes all important algorithms of detection and filtering. Experimental results showed that our method has only a low delay to detect abnormal traffic of distributed denial-of-service attacks, and with a high percentage o f filtering.

Energy-efficient node deployment strategy for wireless sensor networks

To balancing energy consumption in wireless sensor networks, we proposed a fixed time interval node broadcasting scheme under variational acceleration straight-line movement model. Simulation results show that the approach proposed in this paper has a superior performance on energy consumption balance compared to uniform broadcasting methods.

Credibility protection for resource sharing and collaborating in non-center network computing environments

Non-center network computing environments have some unique characteristics, such as instability, heterogeneity, autonomy, distribution and openness, which bring serious issues of security and reliability. This article proposes a brand-new credibility protection mechanism for resource sharing and collaboration in non-center network computing environments. First, the three-dimensional hierarchical classified topology （3DHCT） is proposed, which provides a basic framework for realizations of the identity credibility, the behavior credibility and the capability credibility. Next, the agent technology is utilized to construct the credibility protection model. This article also proposes a new comprehensive credibility evaluation algorithm with simple, efficient, quantitative and able to meet the requirements of evaluating behavior credibility and the capability credibility evaluation as well. The Dempster-Shafer theory of evidence and the combination rule are used to achieve the evaluation of the capability credibility. The behavior credibility is evaluated with the current and historical performance of nodes for providers and consumers to realize more accurate prediction. Based on the non-center network computing simulation test platform, simulation is been conducted to test the performance and validity of the proposed algorithms. Experiment and analysis show that the proposed algorithms are suitable for large-scale, dynamic network computing environments, and able to maintain the credibility for networks without relying on central node, make a non-center network gradually evolve into an orderly, stable and reliable computing environment efficiently.