Automatic Modulation Recognition On Basic Modulation Styles In Softwareradio

The articles on automatic modulation recognition I have seen so far are all based on one aspect, for example, on constant-envelope modulation (CW, FSK, PSK), on N- QAM modulation and on MFSK or MPSK modulation. The article regarding to all basic modulation styles(DSB, FM, PM, 2ASK> 2FSK and 2PSK) is absent, so this paper makes a rearch in the aspect.This article mainly has an analysis in three aspects: one is based upon signal' s own characteristic parameters, the other relies on the parameters gotten by wavelet analysis and the final is the application of artificial neural network in automatic modulation recognition.By computer simulation in MATLAB software environment, we know that: On the extraction of signal' s characteristic parameters, the first method can effectively filter out noise . It is influenced less by noise and can reach a better recognition result. In the second method, wavelet analysis mainly relies on instant information and is greatly influenced by noise, but it can also come to a better recognition result in relatively lower-power noise environment. On the intelligence of judgement threshold, artificial neural network can find a position in automatic modulation recognition. It can make judgement threshold automatize so as to simply and accurately judge.