实数遗传算法的改进及性能研究

提出一种粒子群优化方法(PSO)与实数编码遗传算法(GA)相结合的混合改进遗传算法(HIGAPSO).该方法采用混沌序列产生初始种群、非线性排序选择、多个交叉后代竞争择优和变异尺度自适应变化等改进遗传操作;并通过精英个体保留、粒子群优化及改进遗传算法(IGA)三种策略共同作用产生种群新个体,来克服常规算法中收敛速度慢、早熟及局部收敛等缺陷.通过四个高维典型函数测试结果表明该方法不但显著提高了算法的全局搜索能力,加快了收敛速度;而且也改善了求解的质量及其优化结果的可靠性,是求解优化问题的一种有潜力的算法.     

An artificial intelligence-based spectrum sensing methodology for LoRa and cognitive radio networks

The artificial intelligence-based spectrum sensing approach is extremely important in terms of effective bandwidth utilization for low power wide area networks (LPWANs) based on cognitive radio networks (CRNs). Most studies perform spectrum detection with CRNs using optimization or deep neural network methods. However, optimization-based spectrum detection approaches based on current LPWANs are scarce. For this purpose, in this study, a hybrid optimization methodology integrated with CRNs is proposed for LoRa, which is one of the most compatible LPWAN technologies in the Internet of Things (IoTs) recently. In the particle swarm optimization (PSO) part of this hybrid methodology, agent users are created so that secondary users (SUs) could use the licensed band of primary users (PUs) in cognitive radio. On the genetic algorithm side, LoRa error rates are minimized in order to further improve the performance of the proposed method. In this way, effective spectrum sensing is performed in the LoRa network. Various LoRa-CRN experiments have been carried out in the simulation environment, and the probability of detection and false alarm performances have been compared with both theoretical and proposed approaches in terms of quality estimation parameters. It is clear from the results that the proposed methods give successful results for the LoRa-CRNs.     

Multi-hour, multi-traffic class network design for virtualpath-based dynamically reconfigurable wide-area ATM networks

The virtual path (VP) concept has been gaining attention in terms of effective deployment of asynchronous transfer mode (ATM) networks in recent years. In a recent paper, we outlined a framework and models for network design and management of dynamically reconfigurable ATM networks based on the virtual path concept from a network planning and management perspective. Our approach has been based on statistical multiplexing of traffic within a traffic class by using a virtual path for the class and deterministic multiplexing of different virtual paths, and on providing dynamic bandwidth and reconfigurability through virtual path concept depending on traffic load during the course of the day. In this paper, we discuss in detail, a multi-hour, multi-traffic class network (capacity) design model for providing specified quality-of-service in such dynamically reconfigurable networks. This is done based on the observation that statistical multiplexing of virtual circuits for a traffic class in a virtual path, and the deterministic multiplexing of different virtual paths leads to decoupling of the network dimensioning problem into the bandwidth estimation problem and the combined virtual path routing and capacity design problem. We discuss how bandwidth estimation can be done, then how the design problem can be solved by a decomposition algorithm by looking at the dual problem and using subgradient optimization. We provide computational results for realistic network traffic data to show the effectiveness of our approach. We show for the test problems considered, our approach does between 6% to 20% better than a local shortest-path heuristic. We also show that considering network dynamism through variation of traffic during the course of a day by doing dynamic bandwidth and virtual path reconfiguration can save between 10% and 14% in network design costs compared to a static network based on maximum busy hour traffic     

Multi-Objective Evolutionary Algorithm Based on Decomposition for Energy Efficient Coverage in Wireless Sensor Networks

Wireless sensor networks (WSNs) have become a hot area of research in recent years due to the realization of their ability in myriad applications including military surveillance, facility monitoring, target detection, and health care applications. However, many WSN design problems involve tradeoffs between multiple conflicting optimization objectives such as coverage preservation and energy conservation. Many of the existing sensor network design approaches, however, generally focus on a single optimization objective. For example, while both energy conservation in a cluster-based WSNs and coverage-maintenance protocols have been extensively studied in the past, these have not been integrated in a multi-objective optimization manner. This paper employs a recently developed multi-objective optimization algorithm, the so-called multi-objective evolutionary algorithm based on decomposition (MOEA/D) to solve simultaneously the coverage preservation and energy conservation design problems in cluster-based WSNs. The performance of the proposed approach, in terms of coverage and network lifetime is compared with a state-of-the-art evolutionary approach called NSGA II. Under the same environments, simulation results on different network topologies reveal that MOEA/D provides a feasible approach for extending the network lifetime while preserving more coverage area.     

A genetic algorithm‐based flow update scheduler for software‐defined networks

Software‐defined networking (SDN) facilitates network programmability through a central controller. It dynamically modifies the network configuration to adapt to the changes in the network. In SDN, the controller updates the network configuration through flow updates, ie, installing the flow rules in network devices. However, during the network update, improper scheduling of flow updates can lead to a number of problems including overflowing of the switch flow table memory and the link bandwidth. Another challenge is minimizing the network update completion time during large‐network updates triggered by events such as traffic engineering path updates. The existing centralized approaches do not search the solution space for flow update schedules with optimal completion time. We proposed a hybrid genetic algorithm‐based flow update scheduling method (the GA‐Flow Scheduler). By searching the solution space, the GA‐Flow Scheduler attempts to minimize the completion time of the network update without overflowing the flow table memory of the switches and the link bandwidth. It can be used in combination with other existing flow scheduling methods to improve the network performance and reduce the flow update completion time. In this paper, the GA‐Flow Scheduler is combined with a stand‐alone method called the three‐step method. Through large‐scale experiments, we show that the proposed hybrid approach could reduce the network update time and packet loss. It is concluded that the proposed GA‐Flow Scheduler provides improved performance over the stand‐alone three‐step method. Also, it handles the above‐mentioned network update problems in SDN.     

Simple index modulation profile with fast-converging design optimization for multichannel fiber Bragg grating filters

In this paper, a novel hybrid algorithm featuring a simple index modulation profile with fast-converging optimization is proposed towards the design of dense wavelength-division-multiplexing systems (DWDM) multichannel fiber Bragg grating (FBG) filters. The approach is based on utilizing one of other FBG design approaches that may suffer from spectral distortion as the first step, then performing Lagrange multiplier optimization (LMO) for optimized correction of the spectral distortion. In our design examples, the superposition method is employed as the first design step for its merits of easy fabrication, and the discrete layer-peeling (DLP) algorithm is used to rapidly obtain the initial index modulation profiles for the superposition method. On account of the initially near-optimum index modulation profiles from the first step, the LMO optimization algorithm shows fast convergence to the target reflection spectra in the second step and the design outcome still retains the advantage of easy fabrication.     

一种新型的自适应混沌遗传算法

针对标准二进制编码遗传算法的缺陷,提出一种基于实数编码技术的新型自适应混沌遗传算法用于求解优化问题.该算法利用信息熵理论产生较好的初始群体分布,并依据概率分布函数构造杂交算子,同时结合混沌动力学特性和人工神经网络理论,设计了一种自适应混沌变异算子,使算法能有效维持群体多样性,防止和克服进化过程中的"早熟"现象,算法操作简单、易于实现.最后通过对几个经典测试函数的数值实验,验证了该算法在提高解的精度和加快收敛速度方面都有显著改善,从而为解决函数优化问题提供了一种行之有效的新方法.     

A line search approach for high dimensional function optimization

This paper proposes a modified line search method which makes use of partial derivatives and re-starts the search process after a given number of iterations by modifying the boundaries based on the best solution obtained at the previous iteration (or set of iterations). Using several high dimensional benchmark functions, we illustrate that the proposed Line Search Re-Start (LSRS) approach is very suitable for high dimensional global optimization problems. Performance of the proposed algorithm is compared with two popular global optimization approaches, namely, genetic algorithm and particle swarm optimization method. Empirical results for up to 10,000 dimensions clearly illustrate that the proposed approach performs very well for the tested high dimensional functions.     

基于改进遗传算法的多目标JSP调度研究

针对多目标车间作业调度问题(JSP),提出了一种混合遗传算法,将多目标遗传算法得出的初步优化结果作为粒子群算法的初始粒子,利用粒子群算法强化局部搜索,加快收敛速度,改善了简单遗传算法局部搜索能力差、迭代效率低的问题.仿真结果表明了该算法对JSP调度的良好效果.     

A novel heuristic simulation‐optimization method for critical infrastructure in smart transportation systems

A simulation‐based optimization is a decision‐making tool that helps in identifying an optimal solution or a design for a system. An optimal solution and design are more meaningful if they enhance a smart system with sensing, computing, and monitoring capabilities with improved efficiency. In situations where testing the physical prototype is difficult, a computer‐based simulation and its optimization processes are helpful in providing low‐cost, speedy and lesser time‐ and resource‐consuming solutions. In this work, a comparative analysis of the proposed heuristic simulation‐optimization method for improving quality‐of‐service (QoS) is performed with generalized integrated optimization (a simulation approach based on genetic algorithms with evolutionary simulated annealing strategies having simplex search). In the proposed approach, feature‐based local (group) and global (network) formation processes are integrated with Internet of Things (IoT) based solutions for finding the optimum performance. Further, the simulated annealing method is applied for finding local and global optimum values supporting minimum traffic conditions. A small‐scale network of 50 to 100 nodes shows that genetic simulation optimization with multicriteria and multidimensional features performs better as compared to other simulation‐optimization approaches. Further, a minimum of 3.4% and a maximum of 16.2% improvement is observed in faster route identification for small‐scale IoT networks with simulation‐optimization constraints integrated model as compared to the traditional method. The proposed approach improves the critical infrastructure monitoring performance as compared to the generalized simulation‐optimization process in complex transportation scenarios with heavy traffic conditions. The communicational and computational‐cost complexities are least for the proposed approach.     

Lagrangean Based Methods for Solving Large-Scale Cellular Network Design Problems

The cellular network design (CND) problem is formulated as a comprehensive linear mixed integer programming model integrating the base station location (BSL) problem, the frequency channel assignment (FCA) problem and the topological network design (TND) problem. A solution algorithm based on Lagrangean relaxation is proposed for solving this complex cellular network design problem. Pursuing the optimum solution through exact algorithms to this problem appears to be unrealistic considering the large scale nature and NP-hardness of the problem. Therefore, the solution algorithm strategy consists in computing effective lower and upper bounds for the problem. Lower bounds are evaluated through a Lagrangean relaxation technique and subgradient method. A Lagrangean heuristic is developed to compute upper bounds based on the Lagrangean solution. The bounds are improved through a customized branch and bound algorithm which takes in account specific knowledge of the problem to improve its efficiency. Thirty two random test instances are solved using the proposed algorithm and the CPLEX optimization package. The results show that the duality gap is excessive, so it cannot guarantee the quality of the solution. However, the proposed algorithm provides optimal or near optimal solutions for the problem instances for which CPLEX also provides the optimal solution. It further suggests that the proposed algorithm provides optimal or near optimal solutions for the other instances too. Finally, the results demonstrate that the proposed algorithm is superior to CPLEX as a solution approach for the CND problem.     

基于GA的BP神经网络在多目标优化中的应用

针对传统多目标优化的求解方法通常存在目标权值主观性大,优化目标仅为各目标加权和以及在求解过程中各目标优化的不可操作性等问题,文中提出了一种新颖的多目标优化算法,其将改进后的遗传算法与BP神经网络融合,提出了基于遗传算法的BP神经网络融合算法。该算法将遗传算法与BP神经网络算法相结合,充分发挥遗传算法的全局搜索能力优势和BP算法的局部搜索能力特点,使得多目标优化问题得以求解,加快收敛速度,从而提高了收敛精度。     

Optimization of via formation in photosensitive dielectric layersusing neural networks and genetic algorithms

Via formation using photosensitive polymer technology can reduce process cost by reducing process complexity and is hence of great interest in electronics packaging substrate fabrication. However, to overcome technical difficulties and to facilitate low-cost manufacturing, process modeling, optimization and control are required. In this paper, a process optimization approach for via formation in dielectric layers composed of photosensitive benzocyclobutene (BCB) for high density interconnect (HDI) in MCM-L/D substrates is presented. A series of designed experiments are used to characterize the via formation workcell (which consists of the spin coat, soft bake, expose, develop, cure, and plasma de-scum unit process steps). Neural network process models are then constructed to characterize via yield and geometry, as well as film thickness, retention, and uniformity. These models are used for process optimization using genetic algorithms (GA's) and hybrid combinations of GA's with the Powell algorithm and with the simplex algorithm. The optimized process recipes are verified experimentally. Comparison of the three approaches reveals that the hybrid GA/simplex method yields superior recipes     

Optimizing backscattering from arrays of perfectly conducting strips

Eight different numerical optimization algorithms tackled the problem of finding the best spacings for an array of perfectly conducting strips in order to get desirable backscattering characteristics. Local optimizers worked well when the problem was relatively simple and had few parameters. As the complexity of the problem increased, the genetic algorithm proved a better approach. In general, a hybrid genetic algorithm (GA) worked best, because it combined the power of the local search with a global search. This paper presents optimized results that were averaged over twenty independent runs, and discusses the pros and cons of the various approaches.     

Optimal design of digital IIR filters by using hybrid taguchi genetic algorithm

A hybrid Taguchi genetic algorithm (HTGA) is applied in this paper to solve the problem of designing optimal digital infinite-impulse response (IIR) filters. The HTGA approach is a method of combining the traditional GA (TGA), which has a powerful global exploration capability, with the Taguchi method, which can exploit the optimum offspring. The Taguchi method is inserted between crossover and mutation operations of a TGA. Based on minimizing the L/sub p/-norm approximation error and minimizing the ripple magnitudes of both passband and stopband, a multicriterion combination is employed as the design criterion to obtain the optimal IIR filter that can fit different performance requirements. The proposed HTGA approach is effectively applied to solve the multiparameter and multicriterion optimization problems of designing the digital low-pass (LP), high-pass (HP), bandpass (BP), and bandstop (BS) filters. In these studied problems, there are many parameters and numerous local optima so that these studied problems are challenging enough for evaluating the performances of any proposed GA-based approaches. The computational experiments show that the proposed HTGA approach can obtain better digital IIR filters than the existing GA-based method reported recently in the literature.     

Beamforming in the Presence of Mutual Coupling Based on Constrained Particle Swarm Optimization

Two approaches towards beamforming based on constrained particle swarm optimization (PSO) are presented. One approach relies on the penalty function method, for which a new mathematical expression to select the penalty factors is derived. The other method uses a new PSO strategy for solving constrained optimization problems. To maximize performance of the beamforming procedure, the optimal system parameters for the PSO algorithm are derived. Mutual coupling and platform effects are fully accounted for by using the measured active element radiation patterns. The new PSO strategy is compared with a standard genetic algorithm. Using the measured radiation patterns of a seven-element antenna array, a real-life example is presented that demonstrates the possibilities of the new approach.      

Normalized Incremental Subgradient Algorithm and Its Application

The problem of minimizing the sum of a number of component functions is of great importance in the real world. In this paper, a new incremental optimization algorithm, named normalized incremental subgradient (NIS) algorithm, is proposed for a class of such problems where the component functions have common local minima. The NIS algorithm is performed incrementally just as the general incremental subgradient (IS) algorithm and thus can be implemented in a distributed way. In the NIS algorithm, the update of each subiteration is based on a search direction obtained by individually normalizing each component of subgradients of component functions, resulting in much better convergence performance as compared to the IS algorithm and other traditional optimization methods (e.g., Gauss-Newton method). The convergence of the NIS algorithm with both diminishing stepsizes and constant stepsizes is proved and analyzed theoretically. Two important applications are presented. One is to solve a class of convex feasibility problems in a distributed way and the other is distributed maximum likelihood estimation. Numerical examples, arising from two important topics in the area of wireless sensor networks-source localization and node localization-demonstrate the effectiveness and efficiency of the NIS algorithm.     

基于遗传算法和BP算法的多层感知机杂交训练算法

基于梯度下降的神经网络训练算法易于陷入局部最小,从而使网络不能对输入模式进行准确分类。本文提出综合遗传算法和BP算法的杂交算法GA-QP,它结合遗传算法的全局搜索特性和BP的局部收敛特性,实现对神经网络的有效训练。实验表明该算法优于BP算法,实验结果令人满意。     

分布式多网关无线mesh网公平协作路由算法

现有的协作路由协议不能公平地分配无线网络资源,无法满足网络最小流的吞吐量需求。将多并发流的协作路由问题形式化成一个最大化网络整体效用的凸优化问题,并基于对偶分解和子梯度,提出一种分布式的多网关无线mesh网公平协作路由算法FCRMG。实验结果表明,与基于期望传输时间的非协作路由和基于竞争感知的协作路由相比,FCRMG算法在保证网络吞吐量的前提下,能显著提高最小业务流的吞吐量。     

基于混沌免疫遗传算法的神经网络及应用

借助混沌随机序列构造初始种群,将免疫机制引入传统遗传进化过程,有效克服传统遗传算法种群“退化”和“早熟”的不足,保持种群多样性,构造得到混沌免疫遗传优化算法。进而将混沌免疫遗传优化算法与BP神经网络相结合,分别用混沌免疫遗传优化算法和自适应BP算法对网络权值进行全局优化和局部二次优化,建立基于混沌免疫遗传算法的神经网络模型。利用所建立的混合神经网络模型对渤海某海域年极值冰厚进行训练预测,并将模型预测结果与实际数据以及动态拓扑预测的结果进行对比,表日周基于混沌免疫遗传算法的神经网络模型具有很高的预测精度和工程适用性。