I. Introduction
It is well known that since the pioneering work of Carroll and Pecora [1], synchronization of two chaotic dynamical systems is one of the most important applications of chaos and has been paid much attention. In some engineering applications such as secure communications, chaotic synchronization has been investigated extensively and found valuable since chaotic signals are typically broadband, noiselike, difficult to predict, and can hide information efficiently and securely [2]. To recover the message the receiver should synchronize with the transmitter in secure communications. Therefore, the problem of chaotic synchronization should be solved. Many approaches for synchronization of chaotic systems based on exact chaotic model have been proposed. In [3] and [4], the feedback control techniques to achieve chaotic synchronization were reported. However, some parameters of chaotic system cannot be exactly known in priori. If chaotic system is partly known, that is, its differential equation is known but some or all of the parameters are uncertain, the synchronization scheme based on exact model of chaotic systems may be infeasible. To solve the problem of chaotic synchronization with uncertain parameters, adaptive strategy for a variety of chaotic systems has been presented [5]–[7].