Demodulation Techniques Of Multi-h Continuous Phase Modulation

Multi-h CPM has constant envelope and continuous phase.It has many different parameters,so we can get better error resilient performance and better spectral performance by adjusting these parameters.Multi-h CPM gets a lot of attention of telemetry communication researchers because of its good performance.This paper studies demodulation techniques of Multi-h CPM and it mainly includes the following four parts:1.The modulation model and demodulation model of Multi-h CPM have been studied.The phase of CPM modulated signal carries data information to be transmitted and this phase is of memorability.Best demodulated method of CPM is maximum likelihood sequence detection(MLSD).This demodulation structure mainly includes the matched filters in the front end and the trellis decoder in the back end.The matched filters are mainly responsible for correlation of the reference signal and the received signal to get decision metric.The Viterbi algorithm is used by the trellis decoder to obtain the maximum likelihood decoding results.2.Many algorithms to reduce complexity of demodulation structure of Multi-h CPM have been studied.In order to reduce the complexity of demodulation structure,this paper mainly studies three algorithms,namely frequence pulse truncation,Walsh space decomposition and PAM decomposition.This part gives not only the symbol error rate curves of every algorithm but also the complexity to demodulate and the performance loss after adopting these simplified algorithms.Through the comparison between algorithms which can reduce complexity and MLSD can we kown that demodulation complexity and performance loss are contradictory.3.Soft output Viterbi algorithm of Multi-h CPM have been studied.In order to communicate further and more reliably,we need to combine CPM modulation with coding technology.Then soft decision and Viterbi algorithm are used to get the best performance.Soft output Viterbi algorithm and Viterbi algorithm are the same broadly,but there is a little difference that soft output Viterbi algorithm considers priori information and it produces soft information in demodulation.Then this softinformation is used as input of channel decoding in the back end to get the information bits.4.Engineering capability analysis of an algorithm which can reduce the complexity(frequence pulse truncation) has been studied.First we study the principle and draw the simulation flow char of this algorithm.Then we analyse complexity from time aspect and space aspect.At last this paper gives levels of complexity by using parameters of real simulation.