基于改进卷积神经网络的上市企业财务风险预测研究

针对现有财务危机模型在预测方面存在效果弱、精度低的问题,提出采用布谷鸟算法（Cuckoo Search, CS）优化卷积神经网络的财务危机预警模型。选择纺织类上市企业为研究对象,筛选出能够反映财务危机的指标;使用布谷鸟算法优化卷积神经网络的卷积层和池化层的权重,以提高神经网络的性能。仿真实验结果表明,相比于其他经元启发式算法优化的卷积神经网络,所提算法在财务预警方面具有较好的效果。     

多用户MIMO下行链路中线性预编码器和译码器的联合优化

文章研究多用户MIMO下行链路中线性预编码器和译码器的联合优化问题。采用的优化准则是在总发射功率约束下最小化系统的和均方误差（Sum MSE,SMSE）,提出了一种启发式的迭代算法。通过在接收端引入相同的自动增益控制因子,该算法对预编码器和译码器进行交替优化,直到SMSE达到最小值。分析和仿真结果表明：该算法具有较低的复杂度和良好的收敛特性,符号差错率也优于现有算法。     

支持DiffServ的流量工程体系中最小化抢占的路径选择策略

在支持具有不同优先级和不同QoS需求的多业务网络中,抢占是带宽分配和管理的有效策略,但抢占也对网络造成了振荡。在路由策略中引入抢占信息可最小化重路由,提高网络的稳定性。本文在分析MPLS网络现有路径选择策略和抢占策略的基础上,提出一个最小化抢占的带宽约束路径选择算法。其基本思想是在路径选择中考虑低优先级LSP的资源预留信息,在链路开销函数中引入抢占信息即将被抢占的带宽和将被影响的最高优先级。我们的算法能在最小化抢占和重路由的同时实现负载均衡,有效地改善全局网络性能。     

可生存的虚拟网络多层映射方法研究

本文研究了可生存虚拟网络多层映射问题,首先对其建立了整数线性规划模型（ILP）,然后针对较大规模问题提出一种高效的启发式算法VNP-SVNME对其进行求解.实验表明,VNP-SVNME算法的资源映射开销相对ILP仅平均高15%,且优于现有的启发式可生存算法.此外,VNP-SVNME算法的映射时间相对ILP大大降低,可以满足在线虚拟网络映射的需求.     

IP over WDM网络中一种新型虚拓扑构造算法

对IP over WDM的虚拓扑设计算法进行了研究，选择网络整体流量的均衡性作为优化目标，提出了一种新型的虚拓扑构造算法。在算法中，选择负载分布方差作为评价流量均衡性的指标，并给出其定义和计算方法。主要针对拥塞概率性能，在光层和IP层将提出的新型虚拓扑构造算法与其他算法进行比较。在光层的算法比较部分，选择负载均衡启发式路由算法（RLBH，routing with load balancing heuristics）与提出的算法进行比较；在IP层的算法比较部分，选择固定路径最小拥塞路由算法（FPLC，fixed-path least-congestion）与提出的算法进行比较，并对比较结果进行了讨论。比较结果显示在拥塞概率性能方面，提出的新型虚拓扑构造算法优于RLBH算法和FPLC算法。     

基于MACR和CAL启发式的求差知识编译算法

DKCHER算法是基于超扩展规则的求差知识编译算法.本文首先研究了DKCHER算法的执行流程,并定义了互补量的概念,然后设计了启发式策略MACR（maximum complementary amount of clauses with middle result）,用于动态选择与中间结果互补量最大的子句.针对互补展开过程,设计了动态启发式策略CAL（optimal sequence sorted by complementary amount of literals）,将互补展开中的文字按照与输入公式互补量的大小进行排序并展开.将上述两种启发式策略与DKCHER算法相结合,分别设计了MACR_DKCHER算法、CAL_DKCHER算法和MACR_CAL_DKCHER算法.实验结果表明,MACR启发式策略能够提升DKCHER算法的编译效率和编译质量,编译效率最高可提升9倍,编译质量最高可提升1.9倍;CAL启发式策略在子句数和变量数比值较大的实例上,能够提高DKCHER算法的编译效率,但会降低DKCHER算法的编译质量;MACR_CAL启发式最高可将DKCHER算法的编译效率提高12倍,但会导致DKCHER算法的编译质量有所降低.     

一种基于Vxworks的多任务调度算法

实时操作系统的任务调度算法是研究嵌入式系统开发的热点。针对机载二次配电系统（the Second Power Discribution Association,缩写为sPDA）的任务调度．提出了一种在信号量通信的基础上选择时间片轮转的静态权重优先级抢占混合的调度算法,并对算法进行了仿算。     

支持DiffServ的流量工程体系中最小化抢占的路径选择策略

在支持具有不同优先级和不同Qos需求的多业务网络中，抢占是带宽分配和管理的有效策略，但抢占也对网络造成了振荡。在路由策略中引入抢占信息可最小化重路由，提高网络的稳定性。本文在分析MPLS网络现有路径选择策略和抢占策略的基础上，提出一个最小化抢占的带宽约束路径选择算法。其基本思想是在路径选择中考虑低优先级LSP的资源预留信息，在链路开销函数中引入抢占信息即将被抢占的带宽和将被影响的最高优先级。我们的算法能在最小化抢占和重路由的同时实现负载均衡，有效地改善全局网络性能。     

一种基于Vxworks的多任务调度算法

实时操作系统的任务调度算法是研究嵌入式系统开发的热点。针对机载二次配电系统（the Second Power Distribution Association,缩写为SPDA）的任务调度,提出了一种在信号量通信的基础上选择时间片轮转与静态权重优先级抢占混合的调度算法,并对算法进行了仿真。     

基于改进蚁群算法的电力工程数据分析管控方法研究

为了平衡电力工程数据管控的收敛速度与解的多样性,提出了一种新型的启发式通信异构双种群蚁群算法（HHACO）。该算法通过将启发式通信、3-OPT局部搜索规则及异质蚁群系统相结合,并进行设计。其中启发式交流是一种间接交流策略,有助于改善解的偏差,而3-OPT算法能够提高解的多样性,异质蚁群则侧重于平衡收敛速度。通过分析电力建设项目目标控制,建立了费用-工期-质量动态多目标优化模型,进而实现了对电力工程进度的综合管控。将HHACO与经典蚁群优化算法、粒子群算法和模拟退火算法进行比较的结果表明,HHACO在大规模问题上的性能更优,且在保证质量的基础上缩短了1/4的工期,具有良好的实际应用效果。     

New preemption policies for DiffServ-aware traffic engineering to minimize rerouting in MPLS networks

The preemption policy currently in use in MPLS-enabled commercial routers selects LSPs for preemption based only on their priority and holding time. This can lead to waste of resources and excessive number of rerouting decisions. In this paper, a new preemption policy is proposed and complemented with an adaptive scheme that aims to minimize rerouting. The new policy combines the three main preemption optimization criteria: number of LSPs to be preempted, priority of the LSPs, and preempted bandwidth. Weights can be configured to stress the desired criteria. The new policy is complemented by an adaptive scheme that selects lower priority LSPs that can afford to have their rate reduced. The selected LSPs will fairly reduce their rate in order to accommodate the new high-priority LSP setup request. Performance comparisons of a nonpreemptive approach, a policy currently in use by commercial routers, and our policies are also investigated.     

A Preemption-Aware On-line Routing Algorithm for MPLS Networks

In the context of Multiprotocol Label Switching (MPLS), we look at the problem of organizing the mapping of Label Switched Paths (LSPs) in an optimal way throughout the network on the basis of a given objective function. This problem is highly combinatorial and makes dynamic and real-time features a difficult issue for any LSP routing scheme. For this reason, we propose a computationally efficient, though approximate, on-line scheme adapted to an incremental optimization of the network state. We also propose an integration of a computationally efficient preemption mechanism into this approximation. The efficiency of this algorithm is obtained through a very rigid but seemingly appropriate policy concerning which LSPs are appropriate for preemption. A feedback from preemption on the routing process is proposed, in order to regulate the rate of the reroutings and to tune the balance between stability and continuous reorganization in a network.     

EDF调度算法抢占行为的研究及其改进

通过对采用抢占式EDF算法的嵌入式系统中各实时任务抢占行为的分析,建立了一个周期性任务集的抢占模型,从数学上描述了抢占关系、可调度性、调度开销与实时任务的周期、执行时间、最终期限、启动时间等属性之间的关系.依据该抢占模型,提出了一个改进的抢占式EDF调度算法,通过将基于遗传算法的优化方法离线计算得到的实时任务启动时间作为目标系统的一个调度参数,减少抢占次数,改变抢占关系,从而提高系统的可调度能力和实时性能.最后用实验验证了改进的抢占式EDF调度算法的有效性.     

一种改进的进程调度算法在机顶盒上的设计与实现

调度算法对操作系统中的实时性起着关键的作用.就此问题,深入分析了基于最小优裕度算法的内核进程调度算法,为更好地满足Linux系统对嵌入式系统实时可靠性的要求,对现有的LSF调度策略进行分析,指出LSF算法调度算法的不足.为了改进LSF算法的颠簸现象,基于抢占阈值的思想,提出适用于LSF算法的抢占阈值分配方法,动态地给每...     

支持DiffServ的流量工程体系中的抢占策略

在支持具有不同优先级和不同QoS需求的多业务网络中，抢占是带宽分配和管理的有效策略。重点介绍了支持Diffserv的流量工程体系中的抢占问题，包括抢占策略的分类、信令实施以及抢占算法描述，并讨论了抢占与路由的结合问题。     

IP over WDM网中的策略路由算法

业务量工程允许管理者通过赋予业务主干不同的业务量工程属性来体现一定的管理策略,在为业务主干建立标记交换路径(LSP)时也应该考虑这些策略的影响,该文讨论了业务主干具有不同优先权属性时的 LSP建立问题,针对中断 LSP个数最少和中断业务量最小两种指标,分别提出不同的解决策略:最小连接数中断法(MCNIM)和最小连接带宽中断法(MCBIM),并在不同负载的动态业务下对所提算法进行了仿真研究,给出了仿真结果。     

An algorithm for optimal assignment of a wavelength in a tree topology and its application in WDM networks

In this paper, we present a polynomial time algorithm that gives an optimal solution to the routing and wavelength assignment (RWA) problem in a tree topology. One of the major design issues in wavelength-division multiplexed networks is the assignment of the limited number of wavelengths among network stations so that greater capacity can be achieved. The problem of RWA is known to be NP-hard problem. Many researchers have tackled the problem of RWA with a number of efficient heuristic algorithms. This paper presents an algorithm that optimally assigns a single wavelength to maximize one-hop traffic in a tree topology. The algorithm uses dynamic programming and is shown to be optimal with a time complexity of O(N/sup 4/). We also propose a heuristic scheme to use our optimal algorithm for wavelength assignment in a general graph. The heuristic works on the tree subgraphs of a given graph and the remaining spare wavelengths can be assigned with an existing RWA policy.     

Optimal Resource Allocation and Fairness Control in All-Optical WDM Networks

This paper investigates the problem of optimal wavelength allocation and fairness control in all-optical wavelength-division-multiplexing networks. A fundamental network topology, consisting of a two-hop path network, is studied for three classes of traffic. Each class corresponds to a source-destination pair. For each class, call interarrival and holding times are exponentially distributed. The objective is to determine a wavelength allocation policy in order to maximize the weighted sum of users of all classes (i.e., class-based utilization). This method is able to provide differentiated services and fairness management in the network. The problem can be formulated as a Markov decision process (MDP) to compute the optimal allocation policy. The policy iteration algorithm is employed to numerically compute the optimal allocation policy. It has been analytically and numerically shown that the optimal policy has the form of a monotonic nondecreasing switching curve for each class. Since the implementation of an MDP-based allocation scheme is practically infeasible for realistic networks, we develop approximations and derive a heuristic algorithm for ring networks. Simulation results compare the performance of the optimal policy and the heuristic algorithm, with those of complete sharing and complete partitioning policies.     

流量工程中静态路由算法的研究

该文提出了一种应用于流量工程环境中的静态路由算法。考虑当前的网络资源情况,分优先级别在网络中计算并配置标记交换路径(Label Switched Path,LSP),当某一优先级有多条 LSP需要并行配置时,利用遗传算法搜索最优或较优的配置方案,使得网络的链路带宽使用率低于管理员定义的某个限定值,达到合理分布资源的目的。此外,提出了一种改进的 Dijkstra 算法计算 LSP的最短路径。     

Protection approaches for dynamic traffic in IP/MPLS-over-WDM networks

Due to the explosive growth of data-related traffic driven by the Internet, network reliability becomes an important issue. We investigate various protection approaches to handle failures for dynamic traffic demands in IP/MPLS-over-WDM networks. An LSP can be protected at either the IP/MPLS layer or the optical layer. In IP/MPLS layer protection, an LSP is protected by providing a link-disjoint backup LSP between its end nodes. In optical layer protection, an LSP is protected by the backup lightpath of each lightpath traversed by the LSP. We present two integrated routing algorithms: hop-based integrated routing algorithm and bandwidth-based integrated routing algorithm (BIRA) to set up the restorable bandwidth-guaranteed paths efficiently. Then we present a multilayer protection scheme for multiclass traffic in such networks. This scheme takes into account the different QoS and recovery requirements of the traffic to provide protection capability either at the MPLS layer or at the optical layer in a cost-effective manner. We use the connection blocking probability and number of optical-electrical-optical conversions as performance metrics to compare various protection approaches.