Research On Low Complexity Algorithm Of Continous Phase Modulation

Continuous phase modulation(CPM)is an attractive modulation which combines a constant signal envelope and an excellent bandwidth efficiency.CPM offers the possibility to use nonlinear cost-effective and power efficient amplifiers.Continuous phase modulation formats are in wide use due to their efficient use of power and bandwidth.Notwithstanding those favor features,however,CPM is still only adopted in few simple applications because of its high complexity.Receiver complexity reduction techniques are neceessary.Besides,the complexity is exploding exponentially with the increase of the memory length of CPM pulse and the M-ary alphabet.In this paper,we introduce the background of this article and review the current research work on the reducing complexity of CPM signal.Then,we foucus on the novel low complexity algorithm based on those work.Simulation results show that the new algorithm can reduce complexity significantly with a negligible performance loss.The detailed descriptions of this article are as follows.Chapter 1 presents the background of this article and the current research work on the reducing complexity of CPM signal.Then we briefly intrudoce the research work and the arrangement of other chapters.Chapter 2 presents the maximum likelihood sequence detection(MLSD)based on the per-survivor processing(PSP)technique.Firstly,the system model of CPM scheme is described and the receiver block diagram based on the MLSD algorithm is provided.Then,the PSP technique is introduced to estimate the phase state.This new algorithm can simplify the trellis paths and reduce the calculated metrics.Finally,simulation results show that the proposed algorithm suffers a tiny performance loss with significant complexity reduction.Chapter 3 presents the Laurent decomposition(LD)of CPM signal.In this chapter,a binary CPM signal can be exactly decomposed as a sum of finite number of pulse-amplitude modulated(PAM)signal components.Then,this property is extended to M-ary CPM waveforms by expressing the CPM signals as the products of several constitutional binary CPM signals.We realize the optimum and suboptimum receiver based upon this linear representation of the signal.The complexity reduction LD(CR-LD)algorithm is deduced with the data-independent feature of the Laurent function(LF).This algorithm optimizes the receiver state trellis and reduce the calculation metrics in half.Numerical examples,based on theoretical analysis and computer simulations,are provided.Chapter 4 presents a new per survivor processing Laurent decomposition based(PSP-LD)low complexity detection scheme for CPM signals.In the proposed detection method,the phase stated is estimated from the detection of survivor path.Based on the phase estimation result,the branch metrics are calculated,which simplifies the trellis structure.The block diagram of the PSP-LD algorithm is shown and the trellis paths change is detailed.Simulation results show that this algorithm suffers an acceptable performance loss in quaternary CPM signal.Finally,the influence of different CPM signal parameters is analyzed.The performance curve and calculation metrics are analyzed.This algorithm can reduce the number of states and simplified the receiver structure.Chapter 5 makes a comprehensive summary of this paper,illustrates the shortcomings of the main work and gives suggestions on the future work.