Modulation Classification Of Digital Signal Based On OFDM Channel Systems

With the transmission environment of communication signal become more and more intensive, the demand of updating technologies in the communication field is increasing. So the technologies such as modulation, transmission, classification are developing toward the orientation of better performance and more simple system. Modulation classification is one of the key technologies in communication system. It’s an important issue, which have many scholars have been studying for decades. There have already been lots of different modulation classification algorithms, which fall into two categories:decision theoretic and statistical pattern recognition. Decision theoretic is based on the theory of probability judgment, stochastic processes and so on. This algorithm requires a lot of tedious knowledge which is hard to get access to. Consequently, the statistical pattern recognition has become a mainstream in recent years.In order to reduce the complexity of the system and improve the performance, this dissertation bases on the analysis of previous works in the theory of statistical pattern recognition, researches the modulation classification of digital signal on the basis of OFDM system, and focuses on the modulated signal of ASK, FSK, PSK, and specially focus on MPSK. In this dissertation, we abstract some features to classify these signals and prove the performance of OFDM system. And the simulation results are also given in some chapters and sections.Our main research results can be summarized mainly as follows:1. The dissertation gives a detailed analysis of OFDM system and takes QPSK as an example, sets the transmit number of sub-carriers as128, and simulate the performance of transmission system to verify the system’s high spectrum efficiency and good anti-noise performance.2. Based on spectral analysis, this dissertation extracts two features as the parameter to identify ASK, FSK, PSK from each other, the two features are the number of spectral peaks as m, and the normalized spectrum average energy as H. We draw a flowcharts of modulation classification process and simulate the performance of two features and also the performance of the recognition algorithm as well.3. With the modulation of MPSK based on OFDM channel system, the deduction and analysis of the second-order cumulants and the fourth-order cumulants of MPSK signal, An innovative feature to distinguish the modulation order of the MPSK is proposed. And this feature, due to its simple extracted approach and easy realization, can efficiently achieve the goal of modulation classification. This dissertation makes a comparison of the simulation of the MPSK recognition rate. The modulation algorithm is based on high-order cumulants. And the result shows that the advantage of high-order cumulants analysis is verified.