An SVM-based machine learning method for accurate internet traffic classification

Accurate and timely traffic classification is critical in network security monitoring and traffic engineering. Traditional methods based on port numbers and protocols have proven to be ineffective in terms of dynamic port allocation and packet encapsulation. The signature matching methods, on the other hand, require a known signature set and processing of packet payload, can only handle the signatures of a limited number of IP packets in real-time. A machine learning method based on SVM (supporting vector machine) is proposed in this paper for accurate Internet traffic classification. The method classifies the Internet traffic into broad application categories according to the network flow parameters obtained from the packet headers. An optimized feature set is obtained via multiple classifier selection methods. Experimental results using traffic from campus backbone show that an accuracy of 99.42% is achieved with the regular biased training and testing samples. An accuracy of 97.17% is achieved when un-biased training and testing samples are used with the same feature set. Furthermore, as all the feature parameters are computable from the packet headers, the proposed method is also applicable to encrypted network traffic.     

An Improved Protocol for Deadlock and Livelock Avoidance Resource Co-allocation in Network Computing

A multitude of applications require simultaneous access to multiple kinds of resources scatted in distributed sites. This problem is known as resource co-allocation which has evolved as a hot topic in network computing. How to design a kind of high-performance protocol for deadlock and livelock avoidance resource co-allocation becomes a challenging problem. In this paper, we propose a new protocol OODP³ (Optimal ODP³) which is based on the currently popular protocol ODP³ (Order-based Deadlock Prevention Protocol with Parallel requests).OODP³ not only inherits the advantage of ODP³ but also guarantees the fulfillment of resource co-allocation within polynomial time. Theoretical proof is conducted to verify the correctness of OODP³. Experimental results also show that OODP³ achieves the better performance improvements than the existing deadlock and livelock avoidance protocol.     

A power- and coverage-aware clustering scheme for wireless sensor networks

A common and critical operation for wireless sensor networks is data gathering. The efficient clustering of a sensor network that can save energy and improve coverage efficiency is an important requirement for many upper layer network functions. This study concentrates on how to form clusters with high uniformity while prolonging the network lifetime. A novel clustering scheme named power- and coverage- aware clustering (PCC) is proposed, which can adaptively select cluster heads according to a hybrid of the nodes' residual energy and loyalty degree. Additionally, the PCC scheme is independent of node distribution or density, and it is free of node hardware limitations, such as self-locating capability and time synchronization. Experiment results show that the scheme performs well in terms of cluster size (and its standard deviation), number of nodes alive over time, total energy consumption, etc.     

Guaranteed cost active fault-tolerant control of networked control system with packet dropout and transmission delay

The problem of guaranteed cost active fault-tolerant controller (AFTC) design for networked control systems (NCSs) with both packet dropout and transmission delay is studied in this paper. Considering the packet dropout and transmission delay, a piecewise constant controller is adopted. With a guaranteed cost function, optimal controllers whose number is equal to the number of actuators are designed, and the design process is formulated as a convex optimal problem that can be solved by existing software. The control strategy is proposed as follows: when actuator failures appear, the fault detection and isolation unit sends out the information to the controller choosing strategy, and then the optimal stabilizing controller with the smallest guaranteed cost value is chosen. Two illustrative examples are given to demonstrate the effectiveness of the proposed approach. By comparing with the existing methods, it can be seen that our method has a better performance.     

Grid-based collaborative simulation system for vehicle crashworthiness

Globalization and increasing market competition are forcing the automotive industry to develop technology that will enable more efficient collaboration between carmakers and their suppliers. One of the options that can meet this demand is to adopt grid technology in the collaboration process. For the system to be successfully applied, the trust relationship among Virtual Organization (VO) members becomes a very important factor. Systems lacking in secure mechanisms require all the collaborating members to fully trust each other, and thus make their application quite limited. Based on grid technology, this paper proposes a collaborative simulation system for vehicle crash safety design, and discusses the trust relationship among VO members. The grid middleware VEGA and web services working for the collaboration also contribute to the simulation system. Also, a test bed is set up on the basis of the Local Area Network (LAN). At the end of this paper, a real case involving a carmaker and a door supplier is presented to demonstrate how the grid-based collaborative simulation system works.     

基于FPGA的同步时钟报文检测电路的设计

时钟同步技术是解决基于网络的分布式测控系统完成同步测控任务的关键技术,IEEE-1588协议是一种应用于分布式测量和控制系统中的精准时钟同步协议.提出了一种基于IEEE-1588协议在以太网物理层和MAC层之间的介质无关接口(MII/RMII)处检测同步报文的策略和实现精确时间标记方案[1],在此硬件支持方案和方法的基础之上,充分利用FPGA宏模块资源采用较为简便实用的方法设计实现了同步报文检测电路,该部分电路的设计是采用硬件时间标记方案实现IEEE1588高精度时钟同步的基础.在QuartsII 7.2平台下对设计电路进行优化综合和时序仿真,通过在线实时检测验证了电路设计的正确性.初步验证结果表明设计达到课题要求,应用性能良好.     

基于双链量子遗传算法的过程神经元网络训练

基于函数正交基展开的过程神经元网络训练,由于参数较多BP算法不易收敛。针对这一问题,本文提出了一种基于双链量子遗传算法的解决方案。首先按权值参数的个数确定染色体上的基因数,完成种群编码,然后通过染色体评估获得当前最优染色体,以该染色体为目标,用量子旋转门完成种群中个体的更新,用量子非门实现个体变异增加种群多样性。在该方法中,每条染色体携带两条基因链,因此可扩展对解空间的遍历性,加速优化进程。以两组二维三角函数的模式分类问题为例,仿真结果表明该方法不仅收敛速度快,而且寻优能力强。     

基于CAN总线的电动汽车能源管理系统

为保证电动汽车电子控制系统间通信的实时性、可靠性和互操作性,提出基于CAN总线的电动汽车能源管理系统;以AT89C52芯片为核心构建了能源管理系统的基本结构框架和总体功能方案;设计了CAN总线多功能集中器及其相应程序,使能源管理系统可以通过CAN/USB转接口和上位机(PC)通信,也可通过CAN总线与运动控制器相互传输数据,实现数据信息共享,并可动态显示能源节点状态;电池电压监控实验验证了能源管理系统的有效性和可行性,FUDS工况和整车实验验证了其稳定性和可靠性.     

飞机气动力加载系统控制软件设计与实现

针对飞机气动力载荷加载试验的特点,在分析气动力加载系统工作原理的基础上,设计了上下位机结构的计算机控制系统,重点论述了基于LabWindows/CVI平台的上位机监督控制与数据采集软件的具体实现;在软件开发过程中利用多线程机制及高精度定时技术,解决了多通道分布式控制系统中的多任务及实时性等技术难题;该测控软件在某型飞机地面模拟试验中运行良好,很好地满足了多通道气动力加载控制的实时性和可靠性要求。     

高速网络Fast TCP拥塞控制算法的改进

随着网络技术的飞速发展和接入性能的不断提高,如今全世界的互联主干网络呈现出一种高速网络的特性。在这种网络特性下,传统网络中的TCP拥塞控制协议已经开始显现出不适应性,发现了FAST TCP协议中＂持续拥塞＂的形成机制,建立了数学模型来准确的描述＂持续拥塞＂所导致数据流之间带宽分配不公平的问题,进一步利用优先服务队列的方法解决了FAST TCP在网络中的持续拥塞现象。