Hyperchaos synchronization using PSO-optimized RBF-based controllers to improve security of communication systems

This paper studies the Lorenz hyperchaos synchronization and its application to improve the security of communication systems. Two methods are proposed to synchronize the general forms of hyperchaotic systems, and their performance in secure communication application is verified. These methods use the radial basis function (RBF)-based neural controllers for this purpose. The first method uses a standard RBF neural controller. Particle swarm optimization (PSO) algorithm is used to derive and optimize the parameters of the RBF controller. In the second method, with the aim of increasing the robustness of the RBF controller, an error integral term is added to the equations of RBF neural network. For this method, the coefficients of the error integral component and the parameters of RBF neural network are also derived and optimized via PSO algorithm. For better comparison, the proposed methods and an optimal PID controller optimized by PSO are applied to the Lorenz hyperchaotic system for secure communication. Simulation results show the effectiveness and superiority of the proposed methods in both performance and robustness in comparison with the PID controller.     

基于改进遗传算法寻优的神经网络PID控制及应用

根据神经网络PID控制器初值的选取影响系统控制性能的特点,提出了一种基于改进遗传算法寻优的神经网络PID控制方法.即先利用遗传算法对PID控制器参数离线寻优,将求出的参数值作为控制器的比例、积分、微分系数的初值,再进行神经网络PID控制.对一类液位过程的实时控制结果表明采用本方法的控制系统具有较好的稳定性和鲁棒性.     

基于PSO-RBF神经网络的模拟电路诊断*

为了提高径向基神经网络(radial basis funtion neural network,RBFNN)进行模拟电路故障诊断的速度与准确性,提出了一种基于粒子群算法(particle swarm optimization,PSO)优化RBFNN的故障诊断方法。该方法利用PSO优化RBFNN的结构参数,克服了神经网络中模型结构和参数难以设置的缺点,避免了参数选择的盲目性;同时对模拟电路的响应信号采用小波包分解,提取有效故障特征。仿真结果表明,方法具有更高的诊断精度和更快的收敛速度,能有效地实施模拟电路的故障定位。     

Nonfragile asynchronous control for uncertain chaotic Lurie network systems with Bernoulli stochastic process

This paper proposes a novel nonfragile robust asynchronous control scheme for master‐slave uncertain chaotic Lurie network systems with randomly occurring time‐varying parameter uncertainties and controller gain fluctuation. The asynchronous phenomenon occurs between the system modes and the controller modes. In order to consider a more realistic situation in designing a reliable proportional‐derivative controller, Bernoulli stochastic process and memory feedback are introduced to the concept of nonlinear control system. First, by taking full advantage of the additional derivative state term and variable multiple integral terms, a newly augmented Lyapunov‐Krasovskii functional is constructed via an adjustable parameter. Second, based on new integral inequalities including almost all of the existing integral inequalities, which can produce more accurate bounds with more orthogonal polynomials considered, less conservative synchronization criteria are obtained. Third, a desired nonfragile estimator controller is achieved under the aforementioned methods. Finally, 4 numerical simulation examples of Chua's circuit and 3‐cell cellular neural network with multiscroll chaotic attractors are presented to illustrate the effectiveness and advantages of the proposed theoretical results.     

An Evolutionary Radial Basis Function Neural Network with Robust Genetic-Based Immunecomputing for Online Tracking Control of Autonomous Robots

This paper presents an evolutionary radial basis function neural network with genetic algorithm and artificial immune system (GAAIS-RBFNN) for tracking control of autonomous robots. Both the GAAIS-RBFNN computational intelligence and online tracking controller are implemented in one field-programmable gate array (FPGA) chip to cope with the optimal control problem of real-world mobile robotics. The hybrid GAAIS paradigm incorporated with Taguchi quality method is employed to determine the optimal structure of RBFNN. The control parameters of tracking controller are online tuned by minimizing the performance index using the proposed GAAIS-RBFNN to achieve trajectory tracking. Experimental results and comparative works are conducted to show the effectiveness and merit of the proposed FPGA-based GAAIS-RBFNN tracking controller using system-on-a-programmable-chip technology. This FPGA-based online hybrid GAAIS-RBFNN intelligent controller outperforms the existing bio-inspired RBFNN controllers using individual GA and AIS algorithms.     

神经网络的混沌特性及其在加密通信中的应用

首先概括了能实现混沌动力学特点的主要神经网络模型及其产生混沌同步，混沌序列和混沌吸引子等复杂性的基本原理，介绍了如何利用混沌同步，混沌轨迹序列和混沌吸引子等复杂性特点实现通信加密算法，最后总结有关神经网络的混沌特性及其加密通信应用中需要进一步研究的一些课题。     

遗传算法优化BP 神经网络的短时交通流混沌预测

为了提高BP神经网络预测模型对混沌时间序列的预测准确性,提出了一种基于遗传算法优化BP神经网络的改进混沌时间序列预测方法.利用遗传算法优化BP神经网络的权值和阈值,然后训练BP神经网络预测模型以求得最优解,并将该预测方法应用到几个典型混沌时间序列和实测短时交通流时间序列进行有效性验证.仿真结果表明,该方法对典型混沌时间序列和短时交通流具有较好的非线性拟合能力和更高的预测准确性.     

一种基于改进遗传算法的模糊神经网络控制器及其在烧结终点控制中的应用

针对烧结过程这一复杂、多参数耦合的高度非线性系统,融合遗传算法、神经网络和模糊控制的优点,提出一种基于改进遗传算法的模糊神经网络控制方法,并应用于烧结过程终点控制．首先采用遗传算法对给定的模糊神经网络控制器结构参数进行离线优化,然后利用BP算法较强的局部搜索能力和对对象的适应能力,进一步进行参数的在线调整．同时,为解决传统遗传算法早熟和收敛速度慢的问题,从交叉和变异算子、适应度函数选取等方面对遗传算法进行改进．采用精英保留策略,提高了全局搜索性能和收敛速度．仿真结果表明,所提出的控制器优于常规的模糊神经网络控制器(Fuzzy Neural Network Controller, FNNC)．算法的实际应用效果良好,为解决烧结终点控制问题提供了一条新的途径．     

Chaotic secure communication based on a gravitational search algorithm filter

A new chaotic secure communication scheme based on a gravitational search algorithm (GSA) filter is proposed. In this scheme, useful signals are delivered via an encoder, a chaotic transmitter, a GSA-based filter, a chaotic receiver, and a decoder. The security of such a communication system is promoted due to the unpredictable features of the chaotic map and the unknown encoding-modulation scheme. By using a GSA filter technique the resistance of the system to noise is enhanced. To verify the effectiveness of the proposed scheme, it is compared with the current state-of-the-art schemes in simulations. At the same time, comparisons with a genetic algorithm (GA) filter and a particle swarm optimization (PSO) filter are made. Numerical simulations confirm that the proposed method is better in estimating the states and information symbols, and has a lower bit error rate than other schemes.     

基于神经网络的电力系统暂态稳定分布式自适应控制

针对电力系统中普遍存在的系统非线性和参数不确定性等问题,提出一种基于径向基函数神经网络(RBFNN)的分布式自适应控制器,以提高多机电力系统的暂态稳定性.利用基于RBFNN的方法对系统中的未知非线性项和外部扰动进行补偿,设计相应的自适应参数估计方法,逼近未知非线性项的理想权值矩阵.该策略基于多智能体框架,分布式控制器通过通信网络接收测量装置测量的实时数据,并控制储能装置动作,使受到扰动后各发电机能够迅速实现频率同步.利用李雅普诺夫稳定性理论,证明所提出的分布式控制方法的收敛性.最后,通过仿真研究验证所提出的分布式控制方法的有效性.     

时滞不确定飞行系统神经网络非脆弱H∞控制

针对具有参数摄动和状态时延的时滞不确定飞行系统,提出了一种神经网络非脆弱H_∞控制方案。该方案将鲁棒H_∞控制和神经网络控制结合起来,利用径向基神经网络的非线性逼近能力,对飞行系统的非线性不确定项进行逼近。由线性矩阵不等式（LMI）设计系统标称部分的鲁棒控制器,然后利用神经网络的输出来消除系统控制输入中的不确定部分。Lyapunov稳定性分析中,综合考虑了系统参数摄动、时延和神经网络逼近误差的影响,并证明了在所设计的飞行控制器作用下,闭环系统的稳定性。仿真实例验证了提出的飞行控制方案的可行性和有效性。     

改进径向基函数神经网及其在手写体字符识别中的应用

提出一种基于半模型矢量量化（ＳＦＶＱ）技术的改进径向基函数神经网（ＩＲＢＦＮＮ）分类器,并且用于无约束手写体数字的识别。作者在模糊聚类和矢量量化的基础上利用半模糊的思想提出了半模糊矢量量化算法,并在其中加入了有监督的控制,从而使系统在聚类过程中可以确定比较合适的类别数并使聚类结果能更好地反映训练集的概率分布。以半模糊矢量量化作为预处理的改进ＲＢＦ网,应用了多尺度补偿等办法,能够充分利用训练样本集的     

异结构混沌系统同步及其在保密通信中的应用

针对统一混沌系统和Qi混沌系统的模型，根据主动控制思想和Lyapunov稳定性定理，构造主动控制器使得这两个异结构的混沌系统实现同步，并且实现了驱动系统参数已知而响应系统参数未知的异结构混沌系统的白适应同步，并且能较怏的辨析出系统的未知参数．并利用MATLAB进行数字仿真，将同步的异结构混沌系统应用到混沌掩盖和混沌扩频保密通信中．仿真结果表明，异结构混沌系统能够实现稳定的同步，且在混沌掩盖和混沌扩频保密通信中，有用信号均能有效地在接收端恢复出来．     

Design of a Proportional-Integral-Derivative Controller for an Automatic Generation Control of Multi-area Power Thermal Systems Using Firefly Algorithm

Essentially, it is significant to supply the consumer with reliable and sufficient power. Since, power quality is mea sured by the consistency in frequency and power flow between control areas. Thus, in a power system operation and control, automatic generation control (AGC) plays a crucial role. In this paper, multi-area (Five areas: area 1, area 2, area 3, area 4 and area 5) reheat thermal power systems are considered with proportional-integral-derivative (PID) controller as a supplemen tary controller. Each area in the investigated power system is equipped with appropriate governor unit, turbine with reheater unit, generator and speed regulator unit. The PID controller parameters are optimized by considering nature bio-inspired firefly algorithm (FFA). The experimental results demonstrated the comparison of the proposed system performance (FFA-PID) with optimized PID controller based genetic algorithm (GA PID) and particle swarm optimization (PSO) technique (PSO PID) for the same investigated power system. The results proved the efficiency of employing the integral time absolute error (ITAE) cost function with one percent step load perturbation (1% SLP) in area 1. The proposed system based FFA achieved the least settling time compared to using the GA or the PSO algorithms, while, it attained good results with respect to the peak overshoot/undershoot. In addition, the FFA performance is improved with the increased number of iterations which outperformed the other optimization algorithms based controller.      

爆破扫雷器遗传优化RBF模糊控制器研究

研究爆破扫雷器电液伺服系统的控制问题。针对液压系统的非线性、发射架位置和带弹量不同引起的系统负载及转动惯量变化,常规控制方法难以得到满意的动静态性能指标。在分析电液伺服系统传递函数的基础上,设计了基于RBF神经网络的模糊控制器,并采用改进遗传算法对控制器的节点中心矢量和权值等参数进行优化,在优化目标函数中使用控制量平方项,增大其影响力。最后对系统在不同模拟负载及转动惯量的情况下进行了仿真,系统动态响应快、超调小,为优化设计提供了依据。     

递阶遗传粒子群算法在神经网络设计中的应用

将递阶遗传粒子群算法（HGAPSO）应用于神经网络设计,可以在对网络拓扑结构优化的同时对连接权重进行求解。该算法结合了遗传算法在解决离散问题和粒子群算法在解决连续问题上的优势,并利用BP算法沿误差最速下降的能力对连接权重进一步学习,达到全局最优和快速搜索的有机结合。通过对混沌时序信号的预测,表明递阶遗传粒子群算法在较大程度上提高了神经网络的学习性能和泛化能力。     

参数扰动下的混沌同步控制及其保密通信方案

研究了变参数混沌系统的同步控制问题及其保密通信方案设计.主要创新点可归结如下:第一,提出了利用混沌系统状态变量来构造变参数系统参数扰动项的思想,研究了混沌系统参数在另一组混沌序列持续扰动下的同步控制问题.混沌系统对参数具有敏感依赖性,当混沌系统的参数被另一组混沌序列持续扰动时,系统将产生更加难以预测的混沌行为.第二,对加密后的信息信号再进行一次非线性叠来生成加密端与解密端之间的通信信号,而不是直接以加密信号作为通信信号,使得实际传输的通信信号具有更复杂的形式从而难以被破译.此外,该方案只需向接收系统发送一路信号,且同步控制器简单,实用性较强.数值模拟和理论分析表明该方案具有良好的通信、保密特性和较广的适用范围.     

Synchronization of two different chaotic systems using Legendre polynomials with applications in secure communications

In this study, a new controller for chaos synchronization is proposed. It consists of a state feedback controller and a robust control term using Legendre polynomials to compensate for uncertainties. The truncation error is also considered. Due to the orthogonal functions theorem, Legendre polynomials can approximate nonlinear functions with arbitrarily small approximation errors. As a result, they can replace fuzzy systems and neural networks to estimate and compensate for uncertainties in control systems. Legendre polynomials have fewer tuning parameters than fuzzy systems and neural networks. Thus, their tuning process is simpler. Similar to the parameters of fuzzy systems, Legendre coefficients are estimated online using the adaptation rule obtained from the stability analysis. It is assumed that the master and slave systems are the Lorenz and Chen chaotic systems, respectively. In secure communication systems, observer-based synchronization is required since only one state variable of the master system is sent through the channel. The use of observer-based synchronization to obtain other state variables is discussed. Simulation results reveal the effectiveness of the proposed approach. A comparison with a fuzzy sliding mode controller shows that the proposed controller provides a superior transient response. The problem of secure communications is explained and the controller performance in secure communications is examined.     

Design of an intelligent prediction-based neural network controller for multi-scroll chaotic systems

An indispensable part of the precise control of multi-scroll chaotic systems, model identification has received increasing attention in recent years. Because of plant uncertainty and unmodeled dynamics, conventional control methods cannot guarantee a sufficiently high-performance for stabilizing multi-scroll chaotic systems. In an effort to tackle the matter better, we propose an intelligent controller called the adaptive neural network prediction-based controller (NN-PbC ). The specified neural network is trained with the system model, which is extracted from a time series. In actual practice, the data are divided into two sets. One set is used for training and the other set for testing. In fact, a generalized NN will perform well for both training and testing data. The prediction-based control method is then applied to the obtained neural network model to stabilize the multiple equilibrium points. The stability of the closed-loop system is proven. In addition, simulation examples on two typical multi-scroll chaotic systems are presented to demonstrate the effectiveness of the proposed controller.     

Secure speech communication based on the combination of chaotic oscillator and logistic map

This paper uses a combination of a logistic map and a five-term chaotic oscillator to develop a secure speech communication technique. Firstly, an active controller is designed to synchronize the five terms chaotic oscillators at the transmitter and receiver sides. Then, the first encryption level of row-column shift and a bit XOR operation are performed using the chaotic oscillator. It is followed by a bit-level permutation using the combined key, generated by the chaotic oscillator and the logistic map to enhance the security. Various security analysis methods, such as spectrogram analysis, histogram analysis, correlation test, periodogram analysis, SNR analysis, etc. are used to assess the claim of higher security made in this paper. The simulation results confirm that the proposed technique is more secured than some similar available techniques. The proposed technique’s ability to resist various attacks is tested and found to have adequate resistance to various attacks.