Mpskqam Study On Analysis Of Recognition Of Communication Signals

In this dissertation, we make a deep research on the sub-set modulation recognition of the complex modulated signal MPSK, MQAM and the constant envelope signal OQPSK, π/4 QPSK, MSK. Based on the two basic method of modulation classification which is the feature extraction methods and the maximum likelihood methods, we use four algorithms to classify these signals mentioned above respectively.The main works can be summarized as follows:1) This dissertation analyzes the basic characters of the following signals: MPSK, QAM, OQPSK, MSK. Including the charactors of modulations, the structure of modulators, the trait of constellation diagramm, the characters of instantaneous amplitude and nonlinear phase.2) An improved qLLR (quasi Log-likelihood rate) modulation classification algorithm is proposed, in which an adaptive threshold is used. The algorithm can classify MPSK (M=2, 4, 8, 16) modulation automatically in additive white Gaussian noise(AWGN) and the baud rate estimation is not needed, but the carrier frequence is required. The computer simulation shows its performance at low SNR.3) The new algorithm for classification MQAM signals using log-likelihood function is proposed in Gaussian noise and ideal communication channel environment. Via simulation, we validate the hypothesis of the distributings of the amplitude of 16QAM and 32QAM. Then we plot the graph of means and variances after the predigestion of the expression. At last, we get the performance of the classifier. The probability of correct classification can reach 90% when SNR is above 18dB.4) This dissertation analyzes an automatic identification algorithm base on the strong feature of spectrum which is the sprctrum feature extraction method. The simulation is both under the theoretic data and collected data. Experiments results indicate that this method can identify the complex signals MPSK, MFSK, MSK etc without the prior information such as carry frequency, SNR etc and also show its...