Research On Modulation Parameter Estimation For CPM Signals In Fading Channels

Continuous phase modulation (CPM) signals are widely used in modern satellite and mobilecommunications due to their excellent spectral and power efficiency and their constant-modulusproperty. The complexity of wireless channels and the variety of radio communications bringgreat challenge to blind processing of communication signals.Especially, the received signal isdistorted by channel fading, which causes more difficulties for signal analyzing innon-cooperative reception conditions such as electromagnetic spectrum monitoring, signalreconnaissance and electronic countermeasure.Aiming at the effective analysis of CPM signals under non-cooperative reception conditionsin fading channels, this dissertation mainly focuses on the key techniques in characteristicsanalysis and modulation parameters estimation for CPM signals. The relevant theories andalgorithms are also realized on a general signal processing platform. The main work andinnovative achievements can be summarized as follows:1. Based on the analysis of signal particularity and modulation principle, this paperestablishes blind analysis mathematical models for CPM signals in fading channel, and pointsout the characteristics of CPM signals: the spectral particularities, the linear representation andcyclostationarity.2. A blind symbol rate estimation algorithm is proposed based on asecond-order/zero-conjugate cyclic cumulants feature This dissertation deduces the simplifiedexpressions of square spectrum for instantaneous frequency function of CPM signals, analysesthe relationship between the symbol period and the discrete spectral lines of square spectrum,then achieves the symbol rate estimation of CPM signals. Comparing with the existingalgorithms, the proposed algorithm with lower pre-processing requirement dispenses withmodulation index adjusting. The experimental results show that it has good noise resistantperformance and can be used to accomplish effective symbol rate estimation of CPM signals inmultipath fading channels.3. A novel modulation order identification algorithm is proposed based on the first ordercyclic moment of CPM signals. According to the analysis and derivation of the first ordercyclostationarity, modulation order is identified by the relationship between it and the discretespectral lines of cyclic moment.Simulation results sustain the theoretical analysis and indicatethat the proposed algorithm with lower computational complexity is independent of symbolperiod, insensitive to the influence of frequency shaping pulses, compared with the existingalgorithm.4. An algorithm is proposed based on the above two cyclostationary features aiming atestimating jointly the symbol period, the modulation order and the modulation index. Firstly, thesymbol rate is estimated by studying square spectrum for the instantaneous frequency function.Secondly, the modulation order is identified by analyzing the first order cyclostationarity ofCPM signals. Finally, the modulation index is estimated based on the relationship of the interval of cyclic frequency, the symbol period and the modulation index. Experimental results indicatethat the proposed algorithm simplifying the parameters estimation process, improving the rightprobability of the parameters of sigle-h CPM signals and having good performance in multipathfading channel.5. A comprehensive CPM signal analysis system is preliminary implemented and tested byreal CPM signals. Experimental testing results validate that the system is pragmatic and theproposed algorithms lead to an efficient performance in practical signal analysis scenarios.