Chaotic Secure Communication System Based on Predictive Kalman Fault Estimator with Input Constraint

• Main Authors: Fang-ChengGuo, 郭芳誠
• Other Authors: Jason Sheng-Hong Tsai
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/77077224184241987750

碩士 === 國立成功大學 === 電機工程學系碩博士班 === 101 === Chaotic systems are often applied to encryption on secure communications, but they may not provide a high-degree security. In order to improve the communication of security, chaotic system needs to be add other secure signal, but additional dimensions or signals may cause signals to divergence. In this thesis, we redesign a communication scheme that could create secure signals with additional secure signals, and the scheme could keep system convergence. First, we introduce the universal state-space adaptive observer-based fault diagnosis/estimator and the high-performance tracker for the sampled-data linear time varying system with unanticipated decay factors in actuators/ system states. Second, a residual generation scheme and a mechanism for auto-tuning switched gain is also presented, so that the introduced methodology is applicable for the fault detection and diagnosis (FDD) for actuator and state faults to yield the high tracking performance recovery. The evolutionary programming-based adaptive observer is then applied to the problem of secure communication. However, since the given reference input has serious variation at some time instants, the tracker can easily induces the large input which might not conform to the input constraint of some physical systems. To overcome this disadvantage, the proposed modified linear quadratic analog tracker (LQAT) can effectively restrict the control input within the specified constraint interval, under the acceptable tracking performance. The effectiveness and efficiency of proposed design methodology are illustrated through tracking control simulation examples.