Research On Synchronization Technology Of Shortwave Burst Signals And Its DSP Realization

The application of short burst communication technologes in HF system improves the anti-interference capability of shortwave communication. The uncooperative reception (blind reception processing) of short burst signals has become the hotspot and difficulty in the new research field. Based on this background, this paper studies one key technology in blind reception processing of short burst signals, which is synchronization technology. The main research contents are:1. The detection technology for short burst signals is studied. Taking several common detection-adjudgment statistics and adjudgment thresholds into consideration, the method of choosing one among them are analyzed. Moreover, their performances and application situations are discussed.2. The basic theory of the sample rate conversion and timing error estimation technology are studied. The paper has discussed the integral sample rate conversion technology and the non-integral sample rate conversion technology. Then, their structures are introduced respectively. Thirdly, the timing error estimation technology is studied. This paper presents a feedforward timing error estimation modified algorithm after some classical timing error estimation algorithms are analyzed and simulated. Through augmenting sampling parameter, the performance of the algorithm is improved. Compared with the maximum mean power estimate algorithm,the accuracy is improved due to the use of correlation operation. And compared with the nonlinear transform algorithm, the improved algorithm is simpler and easier to be realized. Finally, the performance of the algorithm is tested in simulation.3. The frequency offset estimation technology is studied. Based on autocorrelation function of the signal with modulation removed, one improved frequency offset estimation algorithm is proposed. In this algorithm, accurate frequency offset estimation is obtained by calculating the weight sum of the rough estimation which is obtained from long and short delay autocorrelation functions. The improved algorithm has low complexity. The weight coefficient can be set differently when the signal-to-noise rate is known, which makes the algotithm better flexible. Another improved algorithm is proposed based on this theory, which uses multi-line spectrum to estimate the frequency offset. Both of the improved algorithms expand the range of frequency offset estimation, and meanwhile keep the accuracy of the algorithms from dropping down.4. The carrier phase offset estimation technology is studied. Firstly the ML algorithm and the constellation character algorithm are analyzed and simulated. And then a methord of removing phase hopping in phase offset estimation is proposed. Due to phase offset estimation together with phase error correction, the influence of phase offset can be removed.5. The algorithms chosen in this paper are realized on DSP chip. The synchronization technology is used in the practical signal demodulation processing and the validity and feasibility are tested by adopting the practical signal.