基于求根MUSIC算法的低复杂度OFDMA频偏盲估计

A low complexity CFO blind estimation for OFDMA based on root-MUSIC algorithm

为了解决OFDMA系统频率同步困难的问题,针对子载波交织分配的OFDMA上行链路提出了一种基于多项式求根的多用户载波频偏盲估计算法.首先对接收信号的协方差矩阵进行特征值分解,并利用噪声子空间的特征向量构造一个多项式,然后求解最接近于单位圆的K个根来获得各个载波频偏.此算法的优点是能够在一个OFDMA符号块内同时估计出所有用户的载波频偏而无需信道状态、训练序列和导频等先验信息,同时在低信噪比下也具有良好的准确性.通过仿真表明,相对于一般的谱峰搜索方法,算法不仅减少了噪声对频偏估计的影响,而且有着更低的计算复杂度和更高的分辨率.

In order to overcome the problem of frequency synchronization in the orthogonal frequency division multiple access（OFDMA） system,a multiuser blind carrier frequency offset（CFO） estimation algorithm for OFDMA uplink was proposed based on searching polynomial root.Eigen-decomposition of the covariance matrix of the received signal was used,a polynomial was constructed by using the eigenvectors of the noise sub-space,and the roots nearest to the unit circle were extracted in order to estimate the CFOs of K users.The main advantages of this algorithm are that it obtains the CFOs of all users simultaneously in one OFDMA block without priori knowledge of the channel,training sequence or pilot signal and has good performance under low SNR condition.Simulations show that the algorithm not only achieves lower computational complexity and higher resolution than the general spectral peak searching methods,but also minimizes the noise influence on the frequency estimation.