Research On OFDM Singal Detection Based On Higher Statistic

Orthogonal Frequency Division Multiplexing (OFDM) technique is a efficienct data transmission techique with good immunity for the effects of multi-path channel propagation, intersymbol interference, narrowband interference.and higth frequency using ratio. Wide used in Asymmetrical Digital Subscriber Line( ADSL), Digital Audio Broadcasting (DAB) , Hign Definition Television(HDTV), Wireless Local Area Network (WLAN) and etc.systems.The future Radio Mobile Communication System requires high speed data transmission and multimedia in wireless networks,so Orthogonal Frequency Division Multiplexing will be more and more attented,and will become one of key technologis in the next mobile communication generation.Communication signal auto identify and modulate parameter estimation are key technologis in non-cooperative communication and software radio.In this paper,the mobile fading channel characteristic and the fundament of OFDM are discussed firstly,and then briefly introduces the the OFDM key techniques such as cyclic prefix, Equalizer equalizer and Channel Estimation.Secondly,research the algorithm to detect the OFDM based on four order cumulants.above all accomplished signal sampling, Hilbert transform, signal normalization work.Then using the fact that the OFDM is asymptotically Gaussian, introduce four order cumulants to detect the OFDM singal, after formula derivation and Simulation indicate that this method has good effect.Finally, because the signal-carrier modulated signals after the Rayleigh fading channel also have the asymptotically Gaussian,using a more practical algorithm based on higher order moment. At the same time research the OFDM signal time-domain parameter the blind mode detection using variable length auto-correlation algorithm, The results indicate that the algorithm has beautiful identification defection, have quite theory and practicality value.