| Many researchers begin to concern the realization of signal automatic identification ofmulti-carrier modulation after the appearance of multi-carrier modulation. Wavelet packetmodulation is a new type of multi-carrier modulation technologies which based on wavelet packettransform, it has the advantages of high spectrum utilization, good anti-multipath interference andanti-intersymbol interference ability, also easy to realizing multi-rate transmission etc. In this thesis,the wavelet packet modulation(WPM) signals and orthogonal frequency divisionmultiplexing(OFDM) signals are served as research targets. By extracting the suitable parameter ofa signal, this thesis is to realize the identification between wavelet packet modulation signal andOFDM signal in the additive Gaussian white channel, Rayleigh fading channel and frequencyselective fading channel.1. Based on the difference of wavelet packet modulation signal and OFDM signal in thefractional Fourie domain distribution, the picture of received signal’s fractional Fourie domaindistribution is processed by the image forming technology. In additive Gaussian white noise channel,Rayleigh fading channel and frequency selective fading channel, the characteristic parameters of theformed picture of fractional Fourie domain distribution are discussed, in addition, signal recognitionmethod of wavelet packet modulation, which is based on support vector machine, is proposed. Themodulation identification between OFDM and WPM is finally studied. The effectiveness of thismethod is done by the computer simulations.2. The basic principles and realization of wavelet packet modulation and OFDM are introduced.According to the non-linear and quadratic phase-coupling, the theory of bispectrum analysis isapplied to extract the classification feature and the wavelet packet modulation signal is identified inthe additive Gaussian channel, Rayleigh fading channel and Frequency selective channel.Simulation results show that good performance is obtained. |
PDF Full Text Request