论文部分内容阅读
针对最大熵方法(MEM)中的迭代计算量大与误差扩散的缺陷,以及借助旋转不变技术估计信号参数(ESPRIT)算法中的自相关函数估计会限制平坦快衰落信道预测精度的问题,设计了一种基于扩展Prony 算法的平坦快衰落信道预测算法.扩展Prony算法利用最小二乘法拟合及计算高次代数方程来求复根,不必估计自相关函数,并且能够批量计算,显著改善预测精度与运算效率.理论和仿真结果表明,相对于MEM和ESPRIT算法,扩展Prony算法具有更高的预测精度,在相同的预测精度下,具备3~5 dB的信噪比优势,此外该算法性能稳定,运算效率较高.“,”Aiming at the problem of large amount of computation and error diffusion in the maximum entropy method (MEM), and the estimation of the auto correlation function in ESPRIT algorithm, a fast fading channel prediction algorithm based on the extended Prony algorithm is designed. The extended Prony algorithm uses least squares fitting and polynomial equations to calculate complex roots, without estimation of autocorrelation function, and can significantly improve the prediction accuracy of volume calculation, and operation efficiency. Theoretical and simulation results show that, compared with the MEM and ESPRIT algorithm, the extended Prony algorithm has higher prediction accuracy, and has the advantages of 3~5 dB in the same prediction accuracy, in addition, the algorithm performance is stable, and the operation efficiency is high.