FUPS- DV:用于桌面虚拟化的全时抢占CPU调度算法

桌面虚拟化通常运行混合负载,且更注重交互式性能,现有的虚拟机调度算法无法很好适应这两个特点.本文提出了一种全时抢占CPU调度算法,通过灰盒技术探测虚拟机内部信息用于辅助虚拟机调度,并结合远程桌面的负载特性进行优化.评测表明,5台WindowsXP虚拟机同时运行混合负载,优化后播放幻灯片的显示延迟降低了至少60％.     

Real-time enhancement of reference signals for feedforward controlof random noise due to multiple uncorrelated sources

The problem envisaged in this paper is the optimization of the reference set for broadband feedforward control of sound and vibration in the presence of multiple, uncorrelated primary sources. A real-time algorithm is presented that generates a set of orthogonal virtual references out of a given set of reference signals. The algorithm is based on recursive singular value decomposition (SVD) techniques. The technique simplifies controller design in the sense that primarily, comparable control performance can be achieved using a reduced set of virtual references instead of the extensive original reference set. Hence, controller dimensions (and, thus, the memory requirements for implementation) and the computational burden reduce accordingly. Second, it is shown that convergence speed of the adaptive filtered-X LMS feedforward control algorithm is enhanced. The technique is applied to active control of structure-borne road noise inside a car cabin. Time domain simulation of a control configuration, utilizing either an extensive set of original references as well as a reduced virtual reference set, shows that in the frequency ranges of interest, a set of four virtual references yields the same noise reductions as the original set of 12 references     

Efficient algorithm for virtual topology design in multihoplightwave networks

One of the major challenges in the virtual topology design of a WDM star based system, is to incorporate in the optimization process both realistic objective functions and real system behavior. The authors present an efficient discrete optimization algorithm that meets these goals while incorporating the prevailing traffic conditions. They simulate the real system and then approximate the objective function by a short term simulation. The optimization process is based on an ordinal optimization approach, i.e., is insensitive to the approximation of the objective function obtained by short term simulation. Another crucial issue in virtual topology design is how to evaluate the quality of the solution obtained by the algorithm. They propose a new metric to evaluate the solution quality, the so called suboptimal quality, obtained by deriving the solution's relative position in the state space according to the performance order. The experiments presented in the paper attest to the quality (efficiency and robustness) of the optimization algorithm and its suitability to solve the wavelength assignment problem     

IEEE 802.16网络中保障QoS的多用户业务调度算法

基于IEEE 802.16协议,研究了宽带无线接入网中保障Qos的上行多用户多业务资源分配和调度问题.首先,对该问题建立跨层分析模型,即基于TDM-OFDMA物理层的多址接入队列调度模型.随后提出了基于凸优化方法的最小剩余聚合工作负载算法.可以证明,只要到达业务的QOS参数值在该算法的稳定域内,那么它就能渐进地保证业务的QoS要求,同时还能最小化调度系统的剩余聚合工作负载.仿真实验的数值结果进一步证明了本算法的确能够在可接受的复杂度范围内保证4类服务的Qos要求.     

Performance optimization of distributed-system models withunreliable servers

Models of distributed systems with servers subject to breakdown and repair are investigated for optimization of performance measures. The optimization problems are the cost minimization, response time minimization, and throughput maximization. The system is modeled by a preemptive-resume priority queuing network. The mean-value analysis algorithm is applied to derive a relationship between the multiprogramming level and performance measure formulas. Based on this relationship the Lagrange multiplier technique is applied to carry out the optimization of performance measures. Optimal service rates are obtained that reach a target throughput while minimizing the total cost. Servers are also treated individually in order to minimize the mean response time of a particular server in the system in order to find the optimal service rates which minimize the response time of a particular server while reaching a target throughput. Formulas are derived for determining the maximum throughput of the system; unfortunately, it has no closed-form solution. However, it can be solved using the binary search and insert value method. Numerical examples illustrate the solutions     

一种生存性军事虚拟网络映射新方法

针对生存性的军事虚拟网络映射问题,提出了生存性的军事虚拟网络映射需要遵循的原则。构建了虚拟网络映射模型,并采用蝙蝠算法进行求解。针对故障情况,提出了区分服务的故障恢复策略,对于高优先级虚网请求提前构建保护路径,对于低优先级虚网请求则提出基于链路可靠性的故障迁移算法,为了减少带宽消耗适当考虑了节点迁移策略。最后通过仿真验证了算法在虚拟网络运行成功率、故障修复率和链路利用率方面相比其他算法具有更好的性能。     

一种适用于稀疏多径信道的自适应均衡算法

针对传统的自适应均衡算法在稀疏多径信道下性能表现不佳的问题,提出了一种基于基追踪降噪的自适应均衡算法。该算法利用稀疏多径信道下均衡器权值的稀疏性,将自适应均衡器的训练过程看作压缩感知理论中稀疏信号对字典的加权求和,并利用重构算法直接对稀疏权值进行求解,解决了迭代参数设置和收敛慢的问题。采用基追踪降噪作为重构算法并选用变量分离近似稀疏重构对该最优化问题进行求解,既提高了权值的重构精度又降低了计算的复杂度。仿真结果表明,所提算法能够以较低的计算量和较少的训练序列达到更优性能,这对提升系统的通信性能具有参考价值。     

非正交多址接入系统中基于受限马尔科夫决策过程的网络切片虚拟资源分配算法

针对无线接入网络切片虚拟资源分配优化问题,该文提出基于受限马尔可夫决策过程(CMDP)的网络切片自适应虚拟资源分配算法。首先,该算法在非正交多址接入(NOMA)系统中以用户中断概率和切片队列积压为约束,切片的总速率作为回报,运用受限马尔可夫决策过程理论构建资源自适应问题的动态优化模型;其次定义后决策状态,规避最优值函数中的期望运算;进一步地,针对马尔科夫决策过程(MDP)的“维度灾难”问题,基于近似动态规划理论,定义关于分配行为的基函数,替代决策后状态空间,减少计算维度;最后设计了一种自适应虚拟资源分配算法,通过与外部环境的不断交互学习,动态调整资源分配策略,优化切片性能。仿真结果表明,该算法可以较好地提高系统的性能,满足切片的服务需求。     

Kurtosis-Based Blind Multiuser Detection in Multipath CDMA Channels

1　ＩｎｔｒｏｄｕｃｔｉｏｎＭｕｌｔｉｐｌｅ ＡｃｃｅｓｓＩｎｔｅｒｆｅｒｅｎｃｅ (ＭＡＩ)ｏｗｉｎｇｔｏｍａｎｙｓｉｍｕｌｔａｎｅｏｕｓｕｓｅｒｓｃｏｎｓｔｉｔｕｔｅｓｔｈｅｍａｉｎｌｉｍｉｔａ ｔｉｏｎｏｆＤＳ ＣＤＭＡｓｙｓｔｅｍｓ .ＭｕｌｔｉｕｓｅｒｄｅｔｅｃｔｉｏｎｔｅｃｈｎｉｑｕｅｓｃａｎｅｆｆｉｃｉｅｎｔｌｙｓｕｐｐｒｅｓｓＭＡＩａｎｄｓｕｂ ｓｔａｎｔｉａｌｌｙｉｎｃｒｅａｓｅｔｈｅｃａｐａｃｉｔｙｏｆＣＤＭＡｓｙｓｔｅｍｓ.Ｖａｒｉｏｕｓｍｕｌｔｉｕｓｅｒｄｅｔｅｃｔｉｏｎｓｃｈｅｍ…     

基于自适应遗传算法的稀布阵天线优化

稀布阵能够以较少的天线单元,通过稀布的方式和先进的处理算法获得与传统等间距阵相当的孔径效能,降低了系统的硬件成本,具有灵活布置的优点、重要的研究意义和应用价值。稀布阵与同口径满布阵相比副瓣电平较高。文中提出了一种基于自适应遗传算法的稀布阵列综合优化方法。该算法采用实值编码方式,引入自适应遗传算子,有效提高了优化效率,避免陷入个体早熟及局部收敛,得到了更低的副瓣电平。给出了具体实现步骤以及仿真实例,结果表明,采用自适应遗传算法,得到了更低的稀布阵天线副瓣电平。     

基于遗传算法的RLS自适应算法研究

在通信系统中采用信道均衡技术是改善信道特性行之有效的方法,但研究算法的同时,往往需要通过大量的仿真实验取平均值来选取最优的参数值,本文首先利用MATLAB仿真软件对线性调制下RLS自适应算法进行仿真分析,然后引入遗传算法的寻优特性及其优点,对RLS最佳遗忘因子λ的选取进行了寻优,得出最佳遗忘因子λ的取值,提供了参数选择的一条捷径,最后通过对比最优λ与参照λ,计算RLS算法均衡已知信号的均方误差值,证明了该方法的可行性。     

A Survey of Adaptive Optimization in Virtual Machines

Virtual machines face significant performance challenges beyond those confronted by traditional static optimizers. First, portable program representations and dynamic language features, such as dynamic class loading, force the deferral of most optimizations until runtime, inducing runtime optimization overhead. Second, modular program representations preclude many forms of whole-program interprocedural optimization. Third, virtual machines incur additional costs for runtime services such as security guarantees and automatic memory management. To address these challenges, vendors have invested considerable resources into adaptive optimization systems in production virtual machines. Today, mainstream virtual machine implementations include substantial infrastructure for online monitoring and profiling, runtime compilation, and feedback-directed optimization. As a result, adaptive optimization has begun to mature as a widespread production-level technology. This paper surveys the evolution and current state of adaptive optimization technology in virtual machines.     

QPID-AVQ：一种基于队列的PID控制的AVQ算法

本文针对AVQ算法存在的不足,在PD-AVQ算法基础上,为进一步提高系统的稳定性,引入积分选项,构成PID-AVQ算法,运用控制理论推导出了满足系统稳定条件的取值范围.利用PID-AVQ算法更新虚拟带宽,综合考虑平均队列长度和包到达速率,提出了一种新的基于队列的PID控制自适应虚拟队列管理算法QPID-AVQ,可根据网络实际情况调节参数,保持队列长度稳定.仿真结果表明,当控制参数在稳定范围内取值时,QPID-AVQ算法能较好地适应网络状态变化,使队列长度始终维持在期望值附近,而不受用户数量的影响,具有较好的稳定性、抗干扰能力和较高的带宽利用率,综合性能优于PD-AVQ和RED算法.     

基于CMAC网络强化学习的电梯群控调度

电梯群控调度是一类开放、动态、复杂系统的多目标优化问题.目前应用于群控电梯调度的算法主要有分区算法、基于搜索的算法、基于规则的算法和其他一些自适应的学习算法.但已有方法在顾客平均等待时间等目标上并不能够达到较好的优化性能.本文采用强化学习技术应用到电梯群控调度系统中,使用CMAC神经网络函数估计模块逼近强化学习的值函数,通过Q-学习算法来优化值函数,从而获得优化的电梯群控调度策略.通过仿真实验表明在下行高峰模式下,本文所提出的基于CMAC网络强化学习的群控电梯调度算法,能够有效地减少平均等待时间,提高电梯运行效率.     

Adaptive control of blood pressure

Stochastic adaptive controllers have been developed for automatic control of blood pressure during infusions of cardiostimulatory or vasoactive drugs. An adaptive algorithm based upon a minimum variance control law is presented. A more advanced algorithm obtained by augmenting the performance measure to include the rate of charge of the control signal is also presented. An autoregressive-moving-average (ARMA) model, representing the dynamics of the system, and a recursive least-squares parameter estimation technique are used for both algorithms. A series of experiments was performed in dogs, utilizing an electronically activated drug infuser. Stable control was achieved, even when the circulatory state of the animal underwent major changes, using either algorithm. On the basis of theoretical considerations and experimental results, we expect that these adaptive controllers will significantly improve the performance of drug infusion systems in clinical applications.     

面向数据密集型工作流的能耗感知调度策略

随着数据中心规模的扩大,高能耗问题已经成为高性能计算领域的一个重要问题。针对数据密集型工作流的高能耗问题,提出通过引入“虚拟数据访问节点”的方法来量化评估工作流任务的数据访问能耗开销,并在此基础上设计了一种“最小能耗路径”的启发式策略。在经典的HEFT算法和CPOP算法基础上,通过引入该启发式策略设计并实现了2种具有能耗感知能力的调度算法（HEFT-MECP和CPOP-MECP）。实验结果显示,基于最小能耗路径的启发式调度算法能有效降低数据访问操作的能耗开销,在面对大型的数据密集工作流任务时,该启发式调度策略体现了较好的适应性。     

Acoustic feedback cancellation based on weighted adaptive projection subgradient method in hearing aids

Acoustic feedback is an important factor that degrades the overall performance of hearing aids, and acoustic feedback cancellation has always been the research focus in the field of signal processing in hearing aids. The newly suggested adaptive projection subgradient method (APSM) for adaptive signal processing solves the problem of difficulty in finding the exact projection operator in the realization of affine projection by taking the subgradient projection hyperplane as the searching region for relaxed projection. This work applies APSM in the acoustic feedback cancellation system of hearing aids for the first time, and proposes a weighted adaptive projection subgradient method (WAPSM), which takes into consideration the exponential decay weight factor to incorporate the prior information of estimation system. The new method is compared with the traditional NLMS algorithm and APSM algorithm in simulation experiments. Incorporating the prior information of estimation system by setting the proper weighting matrix, WAPSM achieved notable improvements on the speed, stability and accuracy of the misalignment convergence. Numerical experiments demonstrate that the proposed algorithm is more robust for low SNR and real speech segment input than the traditional algorithms.     

OFDM系统中子信道的自适应比特、功率分配

研究了一种改进的OFDM系统中子信道的自适应比特、功率分配算法。该算法可适用于多径频率选择性衰落信道,它根据信道的瞬时特性,动态地为OFDM系统的各子信道分配传输比特数和发送功率。依据不同的分配准则,采用了MA和RA两种优化算法。仿真结果表明,采用自适应技术可以大大提高OFDM系统的性能。     

自适应动态Meta粒子群优化算法综合多方向图共形阵列

基于全波仿真得到的广义阵元有源方向图,该文提出一种用于综合多方向图共形阵列的新方法:自适应动态Meta粒子群优化(ADMPSO)算法。在传统Meta粒子群优化(MPSO)算法基础上,定义了优势子群和非优子群的概念,并通过植入非优子群裁减、优势子群规模膨胀以及惯性权重自适应更新等机制,实现了优化过程中多子群的自适应动态调整,全面提高了算法性能。ADMPSO成功用于12元微带锥面共形阵列非赤道面的多方向图综合,综合过程考虑了由共形载体导致的阵元极化指向各异特征,在公共激励存在约束情况下,使阵列同时实现了笔形、平顶,以及余割平方波束总功率方向图,其与该阵列全波数值仿真完全吻合,优化结果和收敛速度相比于其他算法均有显著改善。     

基于三维仿真平台的二维跟踪算法性能检测

对于现阶段昂贵武器不能重复外场实验进而影响光电跟踪经纬仪跟踪算法性能检测的问题,采用了一个模拟真实导弹飞行的三维仿真虚拟平台,通过改变参数模拟真实导弹飞行姿态等改变。主要利用坐标系转换和旋转映射技术解决了在三维仿真平台上检测二维跟踪算法性能的问题。两组实验结果表明:旋转映射技术在检测重心算法跟相关算法时稳定跟踪时平均帧频达到85 帧/s,并且比直接映射技术更节省计算时间,因此,已成功将二维跟踪算法应用于三维仿真平台,并可将该平台作为一个基准检测不同二维跟踪算法的性能。