频偏条件下基于循环平稳的定时估计算法

Cyclostationarity-Based Timing Estimation Considering the Frequency Offset

研究了利用信号的循环平稳统计特性提取定时误差的估计算法,由于性能的优良是建立在大量统计信息的基础上,因而所需的观测时间长度较大,同时算法受频偏的影响也随观测时间长度的增大而增大。针对该特点,研究了频偏对算法性能的影响及存在频偏环境下算法中参数的选择。仿真结果表明,频偏对该类算法的影响不可忽略,尤其是在观测时间长度较大的情况下。实用中应结合具体频偏大小,合理选择参数取值。

Symbol timing delay can be estimated by employing cyclostationarity features of signals. The super performance of estimators is based on large numbers of statistical information. Thus, the observation interval is required to be quite long. However, the performance of estimators become complex when the frequency offset is considered. Due to this problem, the effect of frequency offset and the choice of parameters under this circumstance are studied. Simulation results show that the effect of frequency offset can not be ignored, especially when the observation interval is quite long. In practice, the effect of frequency offset must be considered when choosing the rational parameters.