Research On Modulation Recognition Technology Of Communication Signals

Communication signals modulation recognition technology is widely used inmilitary and civil fields. In this paper, according to the digital communicationsignal modulation technology, proposed digital communication signalsmodulation recognition algorithm based on decision theory and based on higherorder cumulants of the recognition within and between signals.Digital communication signals modulation recognition algorithm based ondecision theory is extracting signal characteristic parameters, through defaultingjudgment threshold artificially, comparing the characteristic parameters with thethreshold, then identifying the signal modulation method. The instantaneousparameters of signals are extracted, the results can be converted to thecharacteristic parameters, in this paper, we use zero center normalizationalgorithm. Simulating the real workspace, selecting the threshold is according tothe simulation experiment. Identification results can be obtained throughcomparing the decision threshold with the characteristic parameters of signals. Inthe application of MATLAB simulation experiment on the algorithm, thealgorithm can effectively accomplish signal recognition. Algorithm used in thispaper starting from three parameters including amplitude, phase and frequency,getting more comprehensive characteristic parameters to used to identify, to alarge extent, improving the algorithm recognition rate, while retaining the featurewith real-time algorithm of decision theory is strong.The principle of communication signals modulation recognition algorithmbased on higher order cumulants is doing frequency processing on the digitalsignals with noise pollution, using the normalized processing, then completingpretreatment process. Building higher order cumulants theoretical value, thevalue is based on signal energy, using higher order cumulants theoretical value toconstruct characteristic parameters, taking it as a benchmark, if the signal to be identified tends to characteristic parameter then it can be considered as thismodulation mode. To verify its feasibility, this paper uses the gauss white noise toprocess the signal, applying simulation experiment to make comprehensiveanalysis on the data, the data prove that higher order cumulants algorithm hashigh recognition rate. And the more complex signal, the better the performance.The comparison of decision theory and higher order cumulants algorithms basedon the recognition rate shows that the higher order cumulants method ofidentification is better under low SNR, which embody the algorithm has strongernoise resistance.