Modulation Classification And Parameter Estimation For CPM Signals

Continuous phase modulation (CPM) technology which is a memoryless nonlinearmodulation scheme is widely used in satellite and mobile communication systems with its goodbandwidth and power efficiency. In non-cooperative reception conditions, based on the analysisof the characteristics of signals, the problems about the automatic modulation recognition andthe blind estimation of modulation parameter of CPM signals are studied.The relevant theoriesand algorithms are also realized on a general signal processing platform.. The main work andcontribution of this paper is summarized as follows:1. Based on the analysis of the mathematical model, modulation principle and lineardecomposition of the CPM signal, study the characteristics of CPM signal: the instantaneousvalue, the power spectral density, cumulants and cyclostationary.2. For most of existing algorithms only suit for CPM signals with half integer modulationindex, a recognition algorithm for CPM signal is proposed based on the cyclic characteristics andmodulation order can be estimated. By analyzing the cyclostationarity characteristics of thedigital modulation signals and comparing the Difference, derived the first-order cyclic momentexpression of signals which have been nonlinear transformed, a feature is proposed todiscriminate CPM signals with an arbitrary index from PSK, QAM signals effectively. Theproposed approach doesn’t need the prior knowledge of symbol rate, carrier phase, timingrecovery, and so on. Simulation results show that the proposed approach could achieve satisfyingclassification performance even when the SNR was lower.3. For the problems that the cyclostationary identification algorithm is only suit for CPMsignal with single index and can’t distinguish between integer index CPM signals from the FSKsignal, a CPM signal recognition algorithm is proposed based on the characteristics of theinstantaneous amplitude. By analyzing the characteristics of the instantaneous amplitude ofsignal, based on the analyses of the relationship between the symbol period and the instantaneousamplitude spectrum of signals, the characteristics reflecting the signal envelope volatility isimproved to distinguish arbitrary index CPM signals with FSK, PSK, QAM signals. Simulationresults show that the proposed algorithm which the modulation index and shaping pulse haveno effect on needs less prior knowledge and low computational complexity.4. A symbol rate blind estimation algorithm is proposed based on wavelet transforming ofinstantaneous frequency. By analyzing the CPM signal instantaneous frequency function, usingits relationship with the symbol rate, combined with the singularity detection of wavelettransform, the symbol rate and the number of modulation index of CPM signals are estimated. Compared to the existing symbol rate estimation algorithm, the proposed algorithm reduces thesignal-to-noise ratio requirements, improves the performance of CPM signal parameterestimation effectively under non-cooperative reception.5. The structure design of the CPM signal recognition and analysis system, and theconstruction of related functional modules are completed. After testing the system performance,the results show that the proposed algorithm can accurately complete the actual CPM signalidentification and parameter estimation, achieving anticipative results.