基于ZC序列的OFDM同步及稀疏信道估计算法

Algorithm to estimate OFDM synchronization and sparse channels by ZC sequences

提出了一种能够在时域兼具完成正交频分复用(OFDM)同步和稀疏信道估计的算法.该算法使用的训练符号由两个重复的Zadoff-Chu(ZC)序列和一个ZC共轭序列组成.通过计算训练序列中两个度量函数的乘积实现了符号定时同步,不须要加权序列.通过在时域计算训练序列和本地序列的移位相关值完成了频偏同步,不须要进行FFT.在由ZC序列形成的矩阵中,依据互不相关(MIP)特性最小原则选出最优的压缩感知矩阵对时域稀疏信道进行观测,并采用正交匹配追踪(OMP)算法重构出了信道估计值.仿真结果对比表明本算法能够在时域以很高的精度完成同步和稀疏信道估计.

A new algorithm was proposed for the orthogonal frequency division multiplexing （OFDM） system, which can complete the synchronization and the sparse channel estimation together in the time domain. The training symbol used in the algorithm was composed by two repetition Zadoff-Chu （ZC） sequences and a conjugated ZC sequences. The symbol timing synchronization was achieved with computing the product of two metric functions in the training sequences without using weighted sequences. The frequency offset synchronization was solved in the time domain via calculating the shift correlation value between the training sequences and the local sequences without FFT （Fourier frequency transfer）. From the matrix formed by the ZC sequences, an optimal compressed sensing （CS） matrix was chosen according to minimizing the mutual incoherence property （MIP）. The sparse channel estimation was sensed by the optimal CS matrix and reconstructed via the orthogonal matching pursuit （OMP） algorithm. The simulation results were compared. Results show that the synchronization and the sparse channel estimation can be completed precisely by this algorithm.