Study And Implementation Of Demodulation Technique For Satellite Burst Signals

This paper is focusing on a study and implementation of the detection and demodulation of satellite burst signals laying emphasis on symbol timing estimation, carrier frequency offset estimation and carrier phase estimation etc. The work finished in this paper is backgrounded with an army key engineering project.First, the theoretic basis of demodulation technology for burst signal is briefly covered, including Maximum Likelihood criterion and cyclostationarity. And then the signal detection algorithms both in time and in frequency domains are discussed, mainly including energy detection algorithm, double sliding windows detection algorithm, adaptive threshold control energy detection algorithm and short autocorrelation algorithm etc.; A detection algorithm in frequency domain based on preambles is presented, which could be used for real-time detection. Several feedforward symbol timing estimation algorithms for PSK signals are discussed and derived from ML function, and the corresponding simulation results are also provided. The estimation algorithms of sinusoidal signals are summarized up, and an improved estimation algorithm with high resolution based on the phase of DFT is proposed which is suitable for the case when the frequency to be estimated just lie between two adjacent spectrum lines. A carrier frequency offset estimation algorithm based on cyclostationarity is derived and a new algorithm based on two-step estimation is proposed, which can greatly reduce the computational complexity. The phase estimation algorithms based on M-power for MPSK signals, the HA algorithm for high order QAM signals and blind algorithm of CEO are discussed and derived, based on which a new phase estimation algorithm without ambiguity is presented which is suited for burst-mode transmissions.Monte Carlo simulation results have proved the feasibility and effectiveness of the algorithms proposed in this paper.