The Research Of Digital Communication Signals Recognition And Parameter Estimation

As a very important issue in the field of the non-cooperative communication signalprocessing, automatic modulation recognition of communication signals is widely used incivilian areas and military areas of signal interception, signal monitoring, software radioand satellite communications. The purpose of modulation recognition is that,by processingthe received signal, such as estimating carrier frequency and symbol rate of the signal, onecan correctly judge out the modulation type of the signal.Modulation recognition is usually composed of three parts,which are signalpreprocessing, feature extraction and classification. The main work and innovation of thisthesis contains:First, it introduces the modulation recognition algorithm based on the instantaneouscharacteristics which are raised by E.E.Azzouz, and A.K.Nandi. These characteristics aresuch classical that the most of the following research are based on them; Second, thepaper discusses how to extract the signal characteristics of the three instantaneouscharacteristics; then, this paper proposes five new characteristic parameters on the basis ofthe existing algorithms. By the simulation using MATLAB for seven digital modulationsignals including2ASK,4ASK2FSK,4FSK,2PSK4PSK and16QAM, the simulationresults show that the proposed new algorithm greatly improve the recognition accuracy.On the other hand, the paper discusses the signal preprocessing of modulationrecognition by analyzing and comparing the advantages and disadvantages of somedifferent methods of the carrier frequency and symbol rate estimation.And according to thedisadvantages of existed algorithms, this thesis puts forward improved algorithms.