Research On OFDM Time And Frequency Synchronization Algorithm Based On L-DACS1 System

L-band Digital Aeronautical Communication System(L-DACS)is an important ground-to-air data link candidate technique of future airborne communication. Employed with orthogonal frequency division multiplexing, L-DACS is facing high transmission delay and large Doppler in ground-to-air data link, making its synchronization technology a vital issue to be addressed. L-DACS1 is extremely sensitive to symbol timing and frequency deviation, which will degenerate the system performance, and the system should meet stricter synchronous requirements when considering airborne communication’s far transmitting range, complex channel conditions and large Doppler frequency shift. This research set up training sequence based on Zadoff-Chu(ZC), and the proposed algorithm obtain accurate symbol timing synchronization and carrier frequency synchronization, which have settled down the high transmission delay and large Doppler shift in L-DACS1. The research contents are as follows:(1) The influence of symbol timing offset estimation and frequency offset to L-DACS1 is analyzed. This paper did both theoretical research and simulation analysis into the effects of symbol timing offset estimation and frequency offset to L-DACS1, and further illustrated the importance of accurate symbol timing and frequency synchronization to L-DACS1, then explained the impacts of the system to symbol timing carrier frequency synchronization.(2) A timing synchronization algorithm based on ZC is proposed. By comparing with traditional timing synchronization algorithm, Simulation results reveal the proposed algorithm show better performance under aeronautical channel and accurate timing even under poor SNR, which would satisfy the demands of L-DACS1, save the expense of network transmission bands and improve the timing accuracy, thus obtaining more accurate symbol timing synchronization.(3) A frequency offset estimation algorithm based on ZC is proposed. This algorithm obtained integer frequency offset estimation and fine frequency estimation by setting up training order based on ZC. The contrast simulation with the representative frequency offset estimation algorithm show that the proposed algorithm can not only enlarge the range of frequency offset estimation, overcome the drawbacks of ML algorithm and get better synchronization, but also can improve the accuracy of the frequency offset estimation under poor SNR.