Research Of Synchronization Technique For Short Burst Communication At Low Signal-to-noise Ratio

Short burst signal has the advantage of short duration,strong anti-interference and anti-capture ability.It's hard to capture and detection.These excellent characteristics make the short burst signal widely used in the communication system with strong demand for concealment.In these communication systems,there is a relatively high-speed movement between both sides of communication,which will produce a large Doppler shift and thus cause serious deviation between the transmited and received signals.At the receiving end of the communication system,The synchronous receiver needs to estimate the carrier frequency and carrier phase accurately.For short burst signals,the synchronization overhead can be quite limited.At the same time,it is more difficult to synchronize at low SNR.Therefore,it is of great significance to study the efficient synchronization technology for short burst communication at low signal to noise ratio(SNR).In this paper,we firstly study some classical carrier synchronization estimation algorithms in additive Gaussian white noise(AWGN)chanel,including FFT-based Gaussian interpolation algorithm,time domain correlation M&M algorithm and EAC algorithm and the coded-adied(CA)expectation-maximization(EM)algorithm.The performance of these algorithms are simulated and analyzed.For burst communication system,high estimation accuracy and large estimation range cannot be achieved at the same time at low SNR,a hybrid data-aided(DA)and code-aided carrier synchronization scheme is proposed.Firstly,the synchronization parameters coarse estimation is performed based on maximization of periodogram and cross-correlation algorithm utilizing known pilot symbols.Secondly,the fine estimation is obtained based on EM algorithm exploiting the pilot symbols and soft information from the extended Turbo decoder.On the basis,the cascade of periodogram maximization and multiple cross-correlations has been proposed,which improves the estimation performance with low complexity.Simulation results show that the proposed algorithm can satisfy the requirements of frequency estimation range and accuracy,BER performance of Turbo decoding is very close to perfect synchronization and the estimation range can reach half of the symbol rate at a relatively efficient pilot cost.In addition,we study the problem of frame synchronization estimation in short burst system.The paper briefly introduces two basic frame synchronization estimation methods and several insertion methods of synchronization codes.At the same time,the commonly used synchronization code-Barker code and m sequence are described.Then we proposed the MAP cascade EM algorithm.Simulation results show that at low signal to noise ratio the algorithm can achieve very accurate capture of the frame head.