Classifier Design Recognition of Modulation of Communication Signals
Abstract
Modulation recognition of communication signals is one of the key technologies in
communication reconnaissance, is an important task in the field of electronic warfare. In
this paper, aiming at the problem of communication signalmodulation recognition,
in-depth study of the signal feature extraction method,on the basis of the design method of
classifier.
Modulation recognition is widely usedin signal distinction, interference identification
and other fields. Its purpose is in a complex environment with noise conditions, in the
absence of a priori knowledge, are processed by the reception of the signals, then judge
the modulation signal.
Neural network is a method of simulation of biological neural work, has some
similarities with the human brain function, self learning and memory function,algorithm is
intelligent, very useful for dealing with nonlinear problems, and fornow in various fields,
such as nonlinear do sales forecasting, computer intelligent control, image processing,
optimization algorithm and so on. Based on the predecessors, the research of signal
recognition deeper problems. The specific work is as follows:
1, an overview ofsome common signal modulation. Some frequently usedmodulation
signal mathematical prototype, investigate some basic properties ofcommon digital
modulation signals, such as time domain waveform and spectrum.
2, the neural network classifier for the whole introduction.
3, the training algorithm classifier, and puts forward the improved algorithm of BP
network to the. Simulation results show that the improved BP network, isfeasible in
theory.
Keywords： Modulation recognition; frequency characteristic; neural network
