Joint Frequency Estimation Method of Preamble and Preamble

Abstract

The invention discloses a method for joint frequency estimation of preamble and preamble. The invention adopts the phase deviation of two known symbols (preamble and postamble) with an interval to invert the frequency deviation, and includes the steps of: calculating the center symbol of the preamble angle; calculate the center symbol angle of the postamble; calculate the frequency offset. The present invention is simple to implement, only needs a small number of known symbols, and can more accurately estimate the frequency deviation in a small range, and its frequency estimation performance is far greater than other algorithms using the same known number of symbols.

Description

technical field [0001] The invention relates to the technical field of satellite communication, in particular to a method for precise frequency estimation of bursts with pre- and post-synchronization codes. Background technique [0002] In the satellite communication system of the MF-TDMA system, the system transmits information according to the TDMA mechanism, with the burst as the smallest unit. Due to the frequency deviation and Doppler frequency shift between the transceivers, the received baseband signal has a certain frequency deviation, which seriously affects the reception performance of the signal. Therefore, whether the frequency offset can be estimated and calibrated quickly and correctly is the premise of receiving data correctly, and it is an essential step in the satellite communication system. [0003] In the MF-TDMA system, after the system completes the acquisition and synchronization, the system corrects the time and frequency of receiving other bursts acc...

AI Technical Summary

Problems solved by technology

[0005] The M&M algorithm has a large deviation correction range and a relatively low threshold of about 0dB. The implementation process is relatively complicated and is suitable for rough estimation of frequency offset. To work under SNR, if you need to work under medium and low SNR, you need to accumulate the signal first and then perform Kay algorithm, which requires a large number of known symbols and high requirements for burst structure; Fitz algorithm and L&R algorithm The range of deviation correction is small, the performance of the algorithm is not much different and there is no obvious limit, it is only suitable for fine frequency estimation, but as the signal-to-noise ratio decreases, the root mean square error performance of the algorithm deviates more and more from the Cramereau boundary, And it is necessary to perform correlation calculations, which is more complicated to realize

[0006] To sum up, the existing algorithms need a large number of continuous known symbols when doing frequency fine estimation. The more known symbols, the better the frequency estimation performance and the more complicated the calculation process; for continuous known For a burst with a small number of symbols, the performance of the existing algorithm for fine frequency estimation is poor, and the calculation is more complicated

Images