基于新序列的OFDM定时估计改进算法

An improved timing estimation algorithm for OFDM using new training sequence

定时估计是准确实现OFDM(orthogonal frequency division multiplexing)系统同步的关键技术之一.实际应用中,为了提高系统对多径时延扩展的抵抗能力,还应尽量减少符号的定时同步误差.为了提高符号定时部分的估计精度,提出了改进算法.算法中设计了一种具有良好自相关特性的新训练序列,结合新序列的数据结构,算法由粗估计和细估计2部分完成.Matlab仿真表明,与传统算法相比,在高斯噪声和多径信道的影响下,所提出的算法具有较小的均方误差,抗干扰能力强的特点,并为频率偏移估计打下了良好的基础.

Timing estimation is one of the key techniques for the exact realization of OFDM （ orthogonal frequency division multiplexing） synchronization. In the practical applications, minimizing symbol timing synchronization error is also needed to be resolved. It is to improve the system property to resist muhipath and delay spread. In order to improve the precision of timing estimation, a new algorithm was proposed. Besides, a new training sequence was proposed, and it had good autocorrelation property. According to the structure of new sequence data, the algorithm was finished with thick and thin estimation. The Matlab simulation indicates that, in the influence of Gaussian noise and multipath channel, the MSE （mean square error） of the proposed algorithm is lower comparing with conventional algorithms, and the algorithm has good property to resist disturbance. And it provides good foundation for the frequency offset estimation.