多种群纵横双向学习和信息互换的鲸鱼优化算法

鲸鱼优化算法(WOA)相较于传统的群体智能优化算法,具有较好的寻优能力和鲁棒性,但仍存在全局寻优能力有限、局部极值难以跳出等问题.针对上述不平衡问题,该文提出一种多种群纵横双向学习的种群划分思路,子群相互独立,子群内个体受到来自横向和纵向两个方向的最优值影响,从而规避局部最优,在探索和开发之间取得均衡.对纵向种群的所有个体,该文提出一种线性下降概率的个体置换策略,促进不同子群的信息流动,加快算法收敛.基于不同个体的历史进化信息,来进行策略算子选择,从而区别于现有基于随机数的策略算子选择方法.利用基准函数进行跨文献对比,数值结果表明该文算法具有很好的优越性和稳定性,在大多数问题上都获得了全局极值,具有较好的问题适用性.     

多种群纵横双向学习和信息互换的鲸鱼优化算法

鲸鱼优化算法(WOA)相较于传统的群体智能优化算法,具有较好的寻优能力和鲁棒性,但仍存在全局寻优能力有限、局部极值难以跳出等问题。针对上述不平衡问题,该文提出一种多种群纵横双向学习的种群划分思路,子群相互独立,子群内个体受到来自横向和纵向两个方向的最优值影响,从而规避局部最优,在探索和开发之间取得均衡。对纵向种群的所有个体,该文提出一种线性下降概率的个体置换策略,促进不同子群的信息流动,加快算法收敛。基于不同个体的历史进化信息,来进行策略算子选择,从而区别于现有基于随机数的策略算子选择方法。利用基准函数进行跨文献对比,数值结果表明该文算法具有很好的优越性和稳定性,在大多数问题上都获得了全局极值,具有较好的问题适用性。     

基于自适应蜣螂算法的无人机三维路径规划方法

山区地势具有陡峭、沟深壑大的环境特点,导致基于启发式算法的山区无人机路径规划速度慢、质量差,针对该问题提出了基于自适应动作策略蜣螂算法的路径规划方法。以路径长度、飞行安全性以及路径平滑度构建路径规划目标函数;在蜣螂算法中引入种群相似性动作变异策略和反向学习策略,平衡局部优化和全局优化能力;通过对比麻雀算法、蜣螂算法和灰狼算法在12个基准函数上的算法性能,结果表明所提方法具有更快的收敛速度、不易陷入局部最优。山区路径规划仿真实验表明,所提方法比蜣螂算法的路径规划质量提高了37.66%。     

基于自适应信息素调整的连续空间优化蚁群算法

对于基本蚁群算法(ACA)不适用求解连续空间问题,并且极易陷入局部最优的缺点,提出了一种基于自适应的蚁群算法。路径搜索策略采用基于目标函数值搜索筛选局部最优解的策略,确保能够迅速找到可行解。信息素更新策略采用自适应的启发式信息素分配策略,使算法能够快速收敛到全局最优解。对2个求函数极值问题进行优化并与其他算法进行比较,结果表明该算法能很好的应用于对连续对象的优化,同时具有较高的寻优精度高,搜索速率快,良好的全局优化性能。     

一种基于排序操作的进化算子自适应遗传算法

提出了一咱基于排序操作的进化算子自适应的遗传算法,该算法中,每个体按适应值大小进行排序,个体的选择、交叉、交异算子的概率根据个体排序值来自适应地确定,其中选择概率还随进化过程而调节,利用Ｍａｒｋｏｖ链的分析法证明了该算法的全局收敛性,最后,实验结果表明该算法同传统的遗传算法相比不仅能收敛到全局最优解,而且具有交快的收敛速度。     

基于混沌的多策略优化麻雀算法及应用

针对原始麻雀搜索算法在寻优过程中出现多样性降低,难以跳出局部最优,以及收敛精度不够等问题,提出一种基于混沌的多策略优化麻雀算法.首先,通过Circle混沌映射进行种群初始化,生成分布更加均匀的麻雀种群,增加种群的多样性;其次,引入自适应比例,对发现者的种群规模占种群总规模的比例进行动态变化,平衡算法的全局搜索与局部挖掘能力;然后引入Levy飞行改进发现者位置更新方式,提高算法的搜索范围与局部搜索能力,并且加快收敛于最优值的速度;最后,选择逐维变异与反向学习相融合的方式来扰动当前全局最优位置,通过贪婪算法来筛选出变异前后的最优值作为当前全局最优值,从而提高算法跳离局部最优的能力.本次选择12个基准函数和Wilcoxon秩和检验进行验证,并于六种其他算法进行对比,证明了以上多种策略对于算法的性能提升明显.同时,将该改进算法应用于工程实践中,本文选择压缩弹簧设计优化问题,验证所提改进算法在工程设计中的可行性与优越性.     

基于小生境遗传算法的移动机器人路径优化

针对标准遗传算法存在收敛性慢和局部最优解的缺陷,结合移动机器人行走特点,提出一种基于预选择机制小生境技术的改进遗传算法中移动机器人路径规划方法.该方法兼顾对局部最优解和全局最优解的搜索,维持群体的多样性,避免了早期收敛现象的发生;同时也增强了自然群体进化的并行性,加快了搜索进程.计算机仿真结果表明,该算法在收敛速度和输出全局最优解概率方面相对于标准遗传算法有了显著提高.     

基于改进深度强化学习的虚拟网络功能部署优化算法

针对网络功能虚拟化/软件定义网络(NFV/SDN)架构下,网络服务请求动态到达引起的服务功能链(SFC)部署优化问题,该文提出一种基于改进深度强化学习的虚拟网络功能(VNF)部署优化算法.首先,建立了马尔科夫决策过程(MDP)的随机优化模型,完成SFC的在线部署以及资源的动态分配,该模型联合优化SFC部署成本和时延成本,同时受限于SFC的时延以及物理资源约束.其次,在VNF部署和资源分配的过程中,存在状态和动作空间过大,以及状态转移概率未知等问题,该文提出了一种基于深度强化学习的VNF智能部署算法,从而得到近似最优的VNF部署策略和资源分配策略.最后,针对深度强化学习代理通过ε贪婪策略进行动作探索和利用,造成算法收敛速度慢等问题,提出了一种基于值函数差异的动作探索和利用方法,并进一步采用双重经验回放池,解决经验样本利用率低的问题.仿真结果表示,该算法能够加快神经网络收敛速度,并且可以同时优化SFC部署成本和SFC端到端时延.     

基于自适应节点选择蚁群算法的QoS选播路由算法

针对基本蚁群算法在求解QoS选播路由问题中存在的容易陷入局部最优和收敛速度慢的缺陷,提出一种基于自适应节点选择的蚁群算法对该问题进行求解.该算法根据解的情况自适应调整节点选择策略;依据各路径上信息素的＂集中＂程度判断解的早熟、停滞情况,并对可能陷入局部最优的解进行信息素混沌扰动更新,以便跳出局部极值区间.仿真实验表明,算法全局搜索能力较强,能够跳出局部极值区间,快速地收敛到全局最优解,算法是可行、有效的.     

基于遗传算法的ISAR运动补偿新算法

提出了基于遗传算法进行逆合成孔径雷达运动补偿的新算法.针对遗传算法解的收敛性问题,在遗传算法中采用了最优个体保存策略,使得解以概率1收敛于全局最优解;针对遗传算法运算量大的问题,提出了采用运动补偿后最小熵和最小距离作为适应度函数,由此形成2种称为GAMCE和GAMCD的ISAR运动补偿新算法.仿真数据和实测数据验证了所提新算法的有效性.     

Wavelength retuning without service interruption in an all‐optical survivable network

This paper proposes a new wavelength retuning (WRT) scheme in an all‐optical WDM network. Compared with the existing WRT schemes developed for all‐optical networks, which can alleviate the wavelength‐continuity constraint but cannot avoid service interruption or data loss, the proposed scheme is able to alleviate the wavelength‐continuity constraint and reduce the connection blocking probability with no service interruption to the on‐going traffic. This is achieved by allocating two routes, one for active path and one for backup path, to each incoming connection request and conducting WRT only on the backup path. The backup path provides an alternate path in case of a failure, while the active path carries traffic under normal conditions. Thus, WRT on the backup path will not cause any impact on data transmission. An optimal backup path WRT scheme and a heuristic algorithm are developed and the performance evaluation on the proposed schemes is presented. The simulation results show that the proposed optimal scheme reduces the connection blocking probability by 46.8% on average, while the proposed heuristic scheme reduces the blocking probability by 28.3% on average, all compared with the scheme without WRT. Copyright © 2009 John Wiley & Sons, Ltd.     

FM-DCSK UWB接收机优化

基于搜索者优化算法（SOA,Seeker Optimization Algorithm）,首次将启发式搜索优化的理念引入到调频差分混沌键控超宽带接收机的积分时间最优化问题中。分析了积分时间对系统性能的影响,在IEEE 802.15.4a室内信道下得到给定信噪比的最优积分时间及其分布,同时也给出了最优积分时间并随信噪比变化的趋势。仿真结果表明,采用SOA较之传统方法更能高效地找到最优值。结论给该系统实现中的参数选择提供了有力依据,新理念的引入也将对难以用闭合数学模型表示的系统优化提供有益指导。     

基于多元优化算法的路径规划

本文提出了一种基于多元优化算法和贝塞尔曲线的启发式智能路径规划方法.该方法通过用贝塞尔曲线描述路径的方法把路径规划问题转化成最优化问题.然后,使用多元优化算法来寻找最优的贝塞尔曲线控制点以获得最优路径.多元优化算法智能搜素个体协同合作交替的对解空间进行全局、局部迭代搜索以找到最优解.多元优化算法的搜索个体（元）按照分工不同可以分为全局元和局部元.在一次迭代中,全局元首先探索整个解空间以找出更优的潜在解区域.然后,局部元在各个潜在解区域进行局部开采以改善解质量.可见,搜索元具有分工不同的多元化特点,多元优化算法也就因此而得名.分工不同的搜索元之间高效的沟通和合作保证了多元优化算法的良好性能.为了评估多元优化算法的性能,我们基于标准测试地图比较了多元优化算法与其它三种经典启发式智能路径规划算法.结果表明,我们提出的方法在最优性,稳定性和有效性上方面优于其它方法.     

基于残差图像的迭代终止条件及其在NCSR算法中的应用

现有图像去模糊(deblurring)算法多依赖于 正则化技术以迭代逼近求解最优目标函数 方式实现关于原图像的最佳估计,由于缺乏迭代终止判定条件(ITC),这些算法通常采用固 定迭代 次数实现以至执行效率不高,同时所获得图像的质量在很多时候也未必是最佳的。鉴于在迭 代过程中每一步所获得的中间估计图像经与模糊核卷积后和模糊图像之间残差图像的亮度 值具有显著的高斯分布特点,提出使用广义高斯分布(GGD)模型为迭代过程中的残差图像建 模并以GGD模型参数值作为衡量去模糊效果的度量 (DM,deblurring measure)。基于DM,在保障去模糊图像质量的前提下,设计了ITC自适应 地终止迭代过程以提高去模糊算法的计算 效率。在经典的非局部集中稀疏表示(NCSR,nonlocally centralized sparse representat ion)去模糊算法上完成的大量 的实验表明,对于运动、高斯和失焦3种典型的模糊失真,新提出的ITC能 够有效判定在每一步迭代过程中所获得的估计图像是否已达到最佳的图像质量,从而实现 在保障去模糊图像质量的前提下大幅度提高NCSR算法计算效率的目的。所提出的ITC判定方 法具有普适性,调整相关参数后也可以应用于其它迭代型的去模糊算法。     

基于改进混合蛙跳算法的CVRP求解

该文提出基于实数编码模式的混合蛙跳算法(Shuffled Frog Leaping Algorithm,SFLA)求解容量约束车辆路径问题(Capacitated Vehicle Routing Problem,CVRP);把具有极强局部搜索能力的幂律极值动力学优化(Power Law Extremal Optimization,-EO)融合于SFLA,针对CVRP对-EO过程进行设计和改进。改进的-EO采用新颖的组元适应度计算方法;采用幂律概率分布来挑选需要变异的组元;根据最邻近城市表,采用幂律概率分布挑选变异组元的最佳邻近城市,执行线路间或线路内的变异。求解测试库中的实例,证明该改进算法有效。     

自适应免疫遗传算法在认知决策引擎中的应用

提出一种自适应免疫遗传算法,设计自适应免疫遗传算子。该算法利用交叉率和变异率自适应调整策略,既防止交叉变异中的个体退化,又保证种群的多样性,并能快速收敛到全局最优解。仿真分析表明,与遗传算法等其他算法相比,该算法具有收敛速度快、平均适应度高、稳定性好等优点,能满足认知引擎参数优化的需要。     

Hybrid protection algorithms based on game theory in multi-domain optical networks

With the network size increasing, the optical backbone is divided into multiple domains and each domain has its own network operator and management policy. At the same time, the failures in optical network may lead to a huge data loss since each wavelength carries a lot of traffic. Therefore, the survivability in multi-domain optical network is very important. However, existing survivable algorithms can achieve only the unilateral optimization for profit of either users or network operators. Then, they cannot well find the double-win optimal solution with considering economic factors for both users and network operators. Thus, in this paper we develop the multi-domain network model with involving multiple Quality of Service (QoS) parameters. After presenting the link evaluation approach based on fuzzy mathematics, we propose the game model to find the optimal solution to maximize the user’s utility, the network operator’s utility, and the joint utility of user and network operator. Since the problem of finding double-win optimal solution is NP-complete, we propose two new hybrid protection algorithms, Intra-domain Sub-path Protection (ISP) algorithm and Inter-domain End-to-end Protection (IEP) algorithm. In ISP and IEP, the hybrid protection means that the intelligent algorithm based on Bacterial Colony Optimization (BCO) and the heuristic algorithm are used to solve the survivability in intra-domain routing and inter-domain routing, respectively. Simulation results show that ISP and IEP have the similar comprehensive utility. In addition, ISP has better resource utilization efficiency, lower blocking probability, and higher network operator’s utility, while IEP has better user’s utility.     

Wavelength converter placement under different RWA algorithms in wavelength-routed all-optical networks

Sparse wavelength conversion and appropriate routing and wavelength assignment (RWA) algorithms are the two key factors in improving the blocking performance in wavelength-routed all-optical networks. It has been shown that the optimal placement of a limited number of wavelength converters in an arbitrary mesh network is an NP-complete problem. There have been various heuristic algorithms proposed in the literature, in which most of them assume that a static routing and random-wavelength assignment RWA algorithm is employed. However, the existing work shows that fixed-alternate routing and dynamic routing RWA algorithms can achieve much better blocking performance. Our study further demonstrates that the wavelength converter placement and RWA algorithms are closely related in the sense that a well-designed wavelength converter placement mechanism for a particular RWA algorithm might not work well with a different RWA algorithm. Therefore, the wavelength converter placement and the RWA have to be considered jointly. The objective of this paper is to investigate the wavelength converter placement problem under the fixed-alternate routing (FAR) algorithm and least-loaded routing (LLR) algorithm. Under the FAR algorithm, we propose a heuristic algorithm called minimum blocking probability first for wavelength converter placement. Under the LLR algorithm, we propose another heuristic algorithm called weighted maximum segment length. The objective of the converter placement algorithms is to minimize the overall blocking probability. Extensive simulation studies have been carried out over three typical mesh networks, including the 14-node NSFNET, 19-node EON, and 38-node CTNET. We observe that the proposed algorithms not only outperform existing wavelength converter placement algorithms by a large margin, but they also can achieve almost the same performance compared with full wavelength conversion under the same RWA algorithm.     

Evaluation of algorithms for efficient partitioning and scheduling of data bursts in satellite communications

We evaluate four scheduling algorithms for satellite communications that use the Time Division Multiple Access methodology. All the algorithms considered are based on the open‐shop model. The open‐shop model is suitably represented or modified to exploit some existing algorithms to solve the satellite communication problem. In the first two algorithms, namely pre‐emptive scheduling with no intersatellite links and greedy heuristics with two intersatellite links, a (traffic) matrix representation of the open‐shop model is used to get a near optimal schedule. In the next two algorithms, generalized heuristic algorithm and the branch and bound algorithm, the open‐shop model is modified to accommodate the inter‐satellite link and this modified open‐shop model is used to solve for a near optimal schedule. The basic methodology of all the algorithms are briefly described and their performance was evaluated through extensive simulations. The performance criteria to evaluate the algorithms are—run time of the algorithms, schedule lengths, and optimality of the algorithm against theoretical bounds. Three of the above‐mentioned algorithms are evaluated by comparing the performance criteria under similar conditions. Optimal branch and bound algorithm is not evaluated due to its high complexity. The general heuristic algorithm is found to give a good trade off between computation time and optimality. The computation time is comparable with the pre‐emptive scheduling algorithm and greedy heuristic algorithm and the schedule length achieved is near to the lower bound value. Copyright © 2005 John Wiley & Sons, Ltd.     

The revisit of QoS routing based on non‐linear Lagrange relaxation

The development of efficient quality of service (QoS) routing algorithms in a high‐speed networking or the next generation IP networking environment is a very important and at the same time very difficult task due to the need to provide divergent services with multiple QoS requirements. Recently, a heuristic algorithm H_MCOP, which is based on a non‐linear Lagrange relaxation (NLR) technique, has been proposed to resolve the contradiction between the time complexity and the quality of solution. Even though H_MCOP has demonstrated outstanding capability of finding feasible solutions to the multi‐path constrained (MCP) problem, it has not exploited the full capability that an NLR‐based technique could offer. In this paper, we propose a new NLR‐based heuristic called NLR_MCP, in which the search process is interpreted from a probability's perspective. Simulation results indicate that NLR_MCP can achieve a higher probability of finding feasible solutions than H_MCOP. We also verify that the performance improvement of a MCP heuristic has a tremendous impact on the performance of a higher level heuristic that uses a MCP heuristic as the basic step. Copyright © 2006 John Wiley & Sons, Ltd.