Research On Modulation Identification Of High-order QAM Signals Over Multipath Fading Channel

The modulation identification technology has been a research hotspot in the field ofcommunication, especially in fields like the software radio, the monitoring of spectrum, theelectronic countermeasures and so on. And there has emerged in large numbers of newtheories and research methods on modulation identification. Most of studies in this field areassumed with the ideal channel environment. However, the signal transmission usually goesthrough multipath fading or gets a variety of interference of other signals in the practicalwireless communication scenario, so the research of modulation identification in theconsideration of the multipath fading channel is more practical.The orthogonal frequency-division multiplexing (OFDM) and the M-ary QuadratureAmplitude Modulation(MQAM), which are both with high spectral efficiency, havebecome the important methods of the modern digital communication systems.OFDM-MQAM signal has been extensively used not only in WLAN transmissiontechnique, but also in the cable access network and the satellite communicationtransmission. With the improvement of demand spectrum efficiency, the sub-carriermodulation order of OFDM-MQAM signal is higher (such as the ADSL signal has morethan2048order). And with the increase of the order of QAM modulation signal, theEuclidean distances between the signal samples are smaller. In addition, the transmission inthe multipath fading channel may lead to the divergence and distortion of the signalconstellation diagram with the effect of the noisy and multipath channel transmission.These aspects result in more difficulties in the signal modulation identification.In the view of the research on the OFDM-MQAM signal modulation identification inthe multipath fading channel environment, this article mainly includes the followingcontents:(1)Blind equalization is used to eliminate the signal distortion caused by the effect ofmultipath fading channel. On the basis of research and analysis on the classic blindequalization algorithms, such as the constant modulus algorithm (CMA), the correctionconstant modulus algorithm (MCAM), and the dual-mode blind equalization algorithms,and based on the feature of OFDM signal frame structural, a new secondary blindequalization algorithm in time domain is proposed, which is suitable for OFDM-MQAMsignals.(2) The research on modulation identification algorithms of OFDM-MQAM signals, including the identification of the OFDM signal and single-carrier signal based on wavelettransform, the identification of OFDM hollow carrier and modulating sub-carriers based onthe high-order cumulants, the identification of the OFDM-MQAM signal sub-carriermodulation based on the clustering algorithm.