【作者】 梅蓉；

【导师】 吴庆宪；

【作者基本信息】 南京航空航天大学 ， 模式识别与智能系统， 2011， 博士

【摘要】 由于混沌系统具有随机性、宽频谱和初值敏感性等特性,因此常用于保密通信系统中以提高通信系统的保密性能。混沌同步是实现混沌保密通信的关键技术之一,研究动态特性更为复杂的混沌系统,并设计鲁棒性能高的混沌同步控制器是研究混沌保密通信的热点问题。同时状态时滞与不确定性干扰是实际混沌系统不可避免的两个因素,研究混沌系统同步时考虑这两个因素更符合实际情况。本论文主要研究时滞不确定混沌系统同步控制并应用于保密通信,其主要研究工作和创新点如下:首先,给出了时滞混沌系统的构建和判定原则,在此基础上构建了时滞Lorenz混沌系统、时滞三角型混沌系统、时滞混沌神经网络及新的时滞混沌系统,并分别对这四类混沌系统进行了特征分析和混沌判定。其次,针对具有时滞的不确定Lorenz混沌系统,设计了异结构主动同步/反同步控制器,并进行仿真验证。同时针对具有干扰的时滞不确定Lorenz混沌系统,分别设计了基于径向基神经网络的自适应滑模单维同步控制器和基于径向基神经网络与非线性干扰观测器的鲁棒受限同步控制器,从而实现混沌同步。然后,针对具有参数不确定的时滞三角型混沌系统,设计了单维自适应滑模控制器实现混沌同步。在此基础上研究了带有结构不确定项的三角型混沌系统同步控制问题,通过设计基于神经网络干扰观测器和回馈递推方法的自适应控制器来实现混沌系统同步,仿真结果证明了所提出的同步控制方法的有效性。接着,针对时变延迟不确定递归混沌神经网络提出了同结构鲁棒自适应神经网络同步控制方案。同时针对具有外界噪声的不确定时变延迟分布式神经网络,设计了H∞同步控制器和基于干扰观测器的鲁棒同步控制器,并通过仿真来验证所提出的同步控制方法的有效性和正确性。最后,对参数具有摄动情况下的同结构时滞混沌系统提出了基于干扰观测器的鲁棒同步控制并应用于保密通信。研究了干扰情况下,基于混沌调制技术的Lorenz混沌同步保密通信技术。针对时变的明文信号,解决了明文信号必须低频率才能被正确恢复的问题。将混沌掩盖和混沌调制技术相结合,对同结构时变延迟混沌神经网络设计了H∞同步保密通信,在实现同步后能有效恢复出隐藏的多路明文信号。同时给出了时滞Lorenz混沌系统的电路和同步电路,并推导出各元件的有效值,在此电路的基础上利用电路仿真软件EWB进行了仿真实验。实验结果表明,利用混沌掩盖技术实现了图像保密传输,并能正确有效地恢复出原图像。更多还原

【Abstract】 To improve the secret performance, chaotic systems are usually used in the secure communication to fully utilize their inherent characteristics such as random, the wide frequency chart, initial sensitivity, and so on. Chaotic synchronization is one of key technologies for achieveing chaotic secure communication. To obtain the chaotic systems with more complex dynamic character and to design the chaotic synchronization controller with high robust performance is the hot topic in the development of chaotic secure communication. At the same time, the state time delay and the uncertainty are two inevitable factors of the actual chaotic systems. Thus, the state time delay and the uncertainty are considered in the chacotic synchronization design which is more suitable for the practical application. This thesis focuses on the design of chaotic synchronization control of the various time delayed uncertain chaotic systems and its application in the secure communication. The main works and innovative points of this paper are as follows:First of all, the constitution means and judgement principles of the delayed chaotic systems have been studied. The delayed Lorenz chaotic systems, the delayed triangular chaotic systems, the delayed chaotic neural networks and a new class of time delay chaotic systems have been constructed and the characters of these constructed chaotic systems have been analysed and judged.Then, the active synchronization and anti-synchronization controllers of the different delayed uncertain Lorenz chaotic systems have been designed and the simulation results have been presented. The synchronization controllers of the delayed uncertain Lorenz chaotic systmes with disturbance have been developed. One-dimension synchronous controller has been designed via using the adaptive sliding mode control based on radial basis fuction (RBF) neural ntwork. Furthermore, the robust constrained synchronous controller has been proposed by using the nonlinear disturbance observer based on RBF neural network. The synchronous performance of the developed synchronization control schemes can be guaranteed.Subsequently, one-dimension adaptive sliding-mode controller has been developed to realize the synchronization of the uncertain parameter triangle chaotic systems. On the based of the study, the synchronous control of the uncertain parameter triangle chaotic systems with the uncertain configuration has been further studied. The synchronization of the chaotic systems has been achieved by designing the disturbance observer based on the neural network and the backstepping adaptive control technique. The simulation results have demonstrated that the proposed synchronization control is effective.Following, the robust adaptive neural network synchronous control scheme has been proposed for the identical time-varying delay uncertain recurrent chaotic neural networks. Two various synchronous controllers have been developed for the uncertain time-varying delay neural networks with noise. One controller is the H∞synchronous controller and the other controller is the robust controller based on the disturbance observer. The simulation results have demonstrated that the proposed synchronization control schemes are effective and correct.Finally, the synchronous control of a class of delay uncertain chaotic systems with parameter perturbations and its application in secure communication have been studied based on disturbance observer. On the other hand, the secure communication of the Lorenz chaotic synchronization under the disturbance has been achieved by using the chaotic modulation technique. At the same time, the original time-varying signals have been considered in the secure communication. In general, the time-varying original signals can be only recovered under the low frequency. Howerver, the developed Lorenz chaotic synchronization can overcome the problem. Furthermore, the secure communication of the H∞synchronization for the identical time-varying delay chaotic neural networks has been proposed by using the techniques of chaotic masking and chaotic modulation. The developed secure communication technology can recover the hidden multiplex original signals. At the same time, the synchronization circuit of the delayed Lorenz chaotic systems is designed and the effective values of circuit components are deduced. The simulation results based on the designed circuit are presented using the Electronic Workbench simulation software. The simuation results also demonstrate that the image can be transmitted securely by using the chaotic masking technique and can be recovered effectively via the developed synchronization circuit of the delayed Lorenz chaotic system.更多还原