| Fractional Fourier transform (FRFT) is a generalization of the classical Fourier transform. FRFT is a new time-frequency method for signal analysis, which can be interpreted as an expression in the time-frequency plane, rotating the signal to the fractional Fourier domain by a certain angle. FRFT combines the information in both time and frequency domain, completing the Fourier Analysis Theory. In recent years, FRFT becomes a widely used application in the fields including signal analyis and processing, optics, and communication technology.Long-Range Dependence (LRD) means that two stochastic data is relevent even if there is a long range of time interval between them. LRD is the essential feature of some stochastic process like netwok traffic, financial data. Triditional mathematical models cannot describe the process generating these data, so that it is important to build Self-Similar model. Hurst parameter measures the LRD level, which makes estimating it crucial in stochastic process analysis. We introduce the design and implementation of a system for Hurst parameter estimation.In chapter 1, we introduce the system development background and the research status of the field. In Chapter 2, the system design related thoery is introduced. In Chapter 3, we describe the system demand, giving the outline design and goal of development. In Chapter 4, we define the functional architecture of every system module. In Chapter 5, we design the system in detail. In Chaper 6, system is implemented on MATLAB GUI platform, and we also do some simulations for testing. In Chapter 7, we summerize the whole thesis and system, and look in to the future for further iprovement.The main contributions of this thesis are:(1) Introducing a new method for Hurst parameter estimation based on FRFT.(2) Implementing the Hurst parameter estimation system based on the thoery above on the platform of MATLAB GUI.(3) Simulating Hurst parameter estimation using actual data. |
PDF Full Text Request