基于遗传算法的OFDM高精度盲频偏估计器

摘要

针对正交频分复用系统频偏估计问题，提出了一种基于遗传算法的高精度盲频偏估计器。该算法首先将带载波间干扰的正交频分复用系统视为正交码分多址接入系统，将载波间干扰系数视为码分多址接入系统的扩频码，利用估计的载波频偏构造扩频码，进而重建接收信号；然后利用重建信号跟实际接收信号最小重建误差原则构造了盲频偏估计的代价函数，并设计了对应于遗传算法的适应度函数；最后利用遗传算法强大的并行搜索和全局寻优能力，通过最小化代价函数求得最佳频偏估值。仿真结果表明，所提算法具有较高的估计精度，低信噪比下性能显著，有效降低了时间复杂度，且不受频偏估计范围和信号调制类型的限制。%For the frequency offset estimation in Orthogonal Frequency Division Multiplexing( OFDM) sys-tems,a high accuracy blind carrier frequency offset estimator is proposed. The OFDM system with inter-carrier interference( ICI) is considered as a multi-carrier CDMA system and the ICI coefficient matrix is considered as the spreading codes in CDMA system. The spreading code is designed with estimated fre-quency offset to reconstruct the received signal. The cost function is constructed based on minimum recon-struction error principle between the received signal and actual received signal,and the fitness function is also designed corresponding to the genetic algorithm ( GA ) . With the strong ability of parallel search and global optimization property of the genetic algorithm,the appropriate frequency offset value is achieved by minimizing the cost function. The simulation results show the proposed algorithm performs higher estimation accuracy than conventional methods,especially in low signal-to-noise ratio( SNR) condition,and it reduces time complexity effectively,without the limitation of frequency offset estimation range and modulation type.