OFDM Signal Detection And Modulation Identification

Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier modulation technology, which has higher spectrum efficiency and better anti-multipath-fading ability. It has been widely used in the communication systems,and is one of key technologies of 4th generation mobile communication systems. With rapid development of communication technology, noncooperative communication technology is facing some new challenges. The thesis, with the nonoperative receiving of OFDM signal as application background, focuses on the OFDM signal detection and modulation identification.The OFDM signal detection, parameters estimation and modulation identification are analyzed and studied in this thesis, which is based on the features of OFDM signal, such as gaussianity, autocorrelation and cyclostationarity. The main work is as follows.1. OFDM signal detection. The existing OFDM signal detection methods are surveied. Through analysis of cumulants of OFDM signal, a new detection method for OFDM signal over frequency selective fading channels is proposed, which is mainly based on new characteristic quantities. Theoretical analyses and computer simulations show that, compared with other existing methods, the proposed method not only can effectively eliminate the impact of frequency selective fading channels, but also has better detection performance under low SNR condition.2. OFDM signal parameters estimation. Estimation approaches of some important parameters of OFDM signal are described, which include symbol rate estimation method based on autocorrelation, subcarrier-number estimation method based on cyclostationarity and timing and frequency offset joint estimation method based on Least Squares.3. OFDM subcarriers modulation identification. The existing methods for identification of OFDM subcarriers modulation are summarized and analyzed. Through combining the ideas of single carrier modulation classification and the existing methods, a hierarchical method for OFDM subcarricer modulation identification is presented. The proposed method, which utilized high-order cumulants as characteristic quantity, can effectively identify typical subcarrier modulations (including null subcarrier).