Research Of Synchronization Parameter Estimation In Satellite Based AIS Signal

In satellite-based AIS,because of the complex channel and fast running speed of the satellite,there are large time delay and Doppler frequency shift in the satellite signal receivers,which seriously affect the detection and signal separation.This paper mainly studied how to accurately estimate time delay and frequency offset under he Gauss white noise channel.According to existing parameter estimation algorithm have some defects,the properties and application of the training sequence is researched in depth,and new parameter estimation algorithms is proposed(1)A related fold time delay estimation algorithm is proposed based on the periodicity of modulation signal of training sequence.According to the analysis of the symmetry of the training sequence modulation signal,this algorithm calculates the correlation of the received signal,and searches the maximum value of the correlation coefficient,so as to obtain accurate estimation of time delay.The simulation results show that the estimation range of this algorithm is large,and the larger effects of frequency offset is,the more accurate estimation will be.(2)A time delay algorithm is proposed based on three order cumulants.The algorithm studies the distribution characteristics of the training sequence modulation signal.And on this basis,by using the differences of the higher order cumulant between the modulation information of training sequence and modulation signal of the signal symbols,the time delay is estimated accurately.The simulation results show that the estimation accuracy of the algorithm has been greatly improved,especially in under the influence of correlated noise,it can also achieve better estimation performance.(3)A linear regression based on phase difference frequency offset estimation algorithm is proposed.After the phase difference among the modulation signal of training sequence in the received signal,the frequency offset is estimated by linear regression equation.The simulation results show that this algorithm has higher estimation accuracy in high frequency,and the estimated range is large.And theoretically,the tolerance of this algorithm is 32 bT.A frequency offset estimation algorithm is proposed based on frequency domain correlation.This algorithm can realize accurately estimation of the frequency offset by the frequency domain correlation of the received signal and re-modulated training sequence.The simulation results show that the performance of this algorithm in estimation accuracy,estimation range,tolerance of phase shift and time delay has been greatly improved.