CAJ | 학술논문

在分析S＆C算法的基础上，提出一种基于新训练序列的改进的正交频分复用（OFDM）同步方法。改进后的算法仍使用两个训练符号，以训练符号本身的强自相关性来消除定时同步中的峰值平台问题，并将训练符号的循环前缀（CP）置为0，以尽量消除CP对定时同步的影响。新的训练符号可直接在时域估计出栽波频偏。改进后的算法在定时同步的准确性上较S＆C算法有了较大提高。给出了在MATLAB环境中使用蒙特卡罗方法的仿真结果。仿真结果表明。在高斯白噪声以及多径信道下，利用序列的强相关性可实现较为精确的同步，虚漏誓率小于5‰。
재분석S＆C산법적기출상，제출일충기우신훈련서렬적개진적정교빈분복용（OFDM）동보방법。개진후적산법잉사용량개훈련부호，이훈련부호본신적강자상관성래소제정시동보중적봉치평태문제，병장훈련부호적순배전철（CP）치위0，이진량소제CP대정시동보적영향。신적훈련부호가직접재시역고계출재파빈편。개진후적산법재정시동보적준학성상교S＆C산법유료교대제고。급출료재MATLAB배경중사용몽특잡라방법적방진결과。방진결과표명。재고사백조성이급다경신도하，이용서렬적강상관성가실현교위정학적동보，허루서솔소우5‰。
On the basis of the study of the S＆C algorithm, an advanced OFDM synchronization algorithm based on a new training sequence is proposed. The advanced algorithm still uses two training sequences, and can eliminate the plateau of correlation peak in the timing synchronization by the strong autocorrelation character of training sequence. Since the timing metric function is usually affected by the cyclic prefix, CP is set to 0. The new training sequence is proposed to estimate the timing and the frequency offset error jointly in time domain. It is much better than the S＆C algorithm in timing synchronization correction. Simulation result is also given in the MATLAB environment by using Monte Carlo. The simulation result shows that in the Ganssian white noise and multi-path channel, the algorithm of the strong correlation can achieve more precise synchronization, the false alarm rate is less than 5‰.