Algorithm Of Modulation Identification For Spread-spectrum And Frequency-hopping Signals Based On Cognitive Radio System

With rapid growth of the wireless communications business, the available spectrum resources is becoming increasingly scarce. As an intelligent sharing spectrum technologies and adaptive adjustment technologies, the conception of Cognitive Radio (CR) produced. Through dynamic testing and use of idle spectrum, CR can greatly increase the utilization of the spectrum and effectively solve the problem of the scarcity of wireless channel resources. In the CR system, the parameters such as absence of communication signals and symbol rate are unknown, so it is important for the realization of cognitive radio to accomplish the work of signal detection, parameter estimation, and signal modulation classification.Based on the cognitive radio framework, this paper focus on the blind signal modulation classification in the spread spectrum and frequency hopping communication system. The so-called blind modulation classification, which refers that the receiver cann't obtain the necessary information to realize the correct reception of signal in the non-cooperative communication, must research on signal detection and estimation pretreatment etc.techniques before mddulation classification. The main work is as follows:First, We discussion signal preprocessing techniques, and explore the cyclostationary of spread spectrum communication signals. According to the synchronizatio problem of timing error, frequency offset, and unknown early-phase exited in actual receiver, we proposed an above parameters estimation algorithm relying on the second-order cyclic statistics, which lay the foundation for subsequent modulation classification.Second, we study on modulation classification problem for M-ary Frequency Shift Keying (MFSK) signal which is used to major modulation in frequency hopping system. A joint Cumulants and box dimension knowledge being used to classify MFSK signals is proposed and the perfomance is outperform the only cumulants or box dimension knowledge respsctively. Finally, the second-order cylic statistics are introduced into the analysis of spread spectrum and continuous phase modulation(CPM) signals. By exploring the Second-order cyclostationarity of Binary Phase Shift Key Direct Sequency(BPSK-DS) signal, Quarter Phase Shift Key Direct Sequency(QPSK-DS) signal and Minimum Shift Keying(MSK) signal, we present a method to classify these signals.