基于训练序列的OFDM符号定时与频偏联合估计

在最大似然算法的符号定时与频偏联合估计算法中，存在频偏估计依赖于定时估计和频偏估计范围小的缺点，采用基于循环前缀的联合估计算法，结合S＆C算法原理，设计具有共轭对称特性的训练序列，并根据训练序列的特性计算符号定时位置值和频率偏差值，为了验证算法的有效性，对算法进行了仿真验证，结果表明，该算法不但频偏估计过程独立于符号定时估计，而且频偏估计范围大．

The frequency offset estimation for OFDM based on ML algorithm

The maximum likelihood algorithm has some weaknesses during joint estimating symbol timing and fre-quency offset, which includes the frequency offset estimation depends on the symbol time estimation and the range of frequency offset estimation is small. The joint estimation algorithm based on circulation prefix is adopted. Combined with the principle of S＆C algorithm, the training sequences which have the conjugate symmetry property are first designed. Secondly, the symbol timing and frequency offset were calculated according to the characteristics of the training sequence. Finally, the algorithm was verified by simulation, the results indicate that the frequency offset estimation was not only independent of the symbol timing estimation, but also the range of the frequency offset estimation was expanded.