Reserach On Low Complexity Coherent Demodulation Of CPM

Continuous phase modulation(CPM)belongs to phase continuous constant envelope modulation,with high power and spectrum utilization,which has been wildly concerned in the areas of mobile communications,satellite communications,deep space communications and so on.The optimal demodulation of CPM is coherent demodulation based on the maximum likelihood criterion,and the optimal demodulation structure is the matched filter bank followed by the optimal path search in the state grid.However,the more excellent performance CPM is,the more complex the state grid is,and the higher complexity the demodulation is.Therefore,this thesis mainly studies the low complexity coherent demodulation technology of CPM.This thesis first studies the important concepts of CPM's state transition and decomposition models.On the basis of this,the optimal receiver structure of CPM and the influence of parameters on CPM performance are studied.It is proved that the large M-ary and memory length can make CPM have more excellent performance and bring higher demodulation complexity.Secondly,the low-complexity CPM coherent demodulation method is studied.The results show that these methods are more effective for CPM with small state grid,but have little effect on CPM with large grid widths.Again,the Fano algorithm of sequence decoding is studied.In this paper,three improved algorithms are proposed to lower the erasure probability of Fano algorithm,which are constraint algorithm,return layer limited with condition algorithm and adaptive return layer limited algorithm respectively.Simulation results show that compared with the traditional improved algorithm,the proposed improved algorithm can reduce the deletion probability of Fano algorithm more significantly,and improve the stability of Fano algorithm greatly.Then,the performance of CPM Fano algorithm demodulation is studied and compared with the Viterbi algorithm demodulation.The simulation shows that although the the performance of bit error rate is slightly lost,the complexity of Fano algorithm demodulation is much lower than that of Viterbi demodulation.The offset correction algorithm proposed in this thesis can reduce this performance gap so that the performance of bit error rate of Fano demodulation is almost the same as that of Viterbi demodulation,while it increases the computational complexity at low SNR.Finally,the improved algorithms proposed in this thesis are applied to CPM Fano demodulation.Simulation results show that,compared with the traditional algorithm,the improved algorithms can more effectively reduce the erasure probability of Fano demodulation,and improve the Fano demodulator's stability and practicability greatly.