基于归一化误差函数的变步长分数间隔盲均衡算法研究

针对常数模算法收敛速度慢的特点,提出了一种快速收敛的盲均衡算法。该算法修改了传统常数模算法的误差函数,且采用可变步长和分数间隔采样,有效地提高了收敛速度。浅海水声信道模型仿真结果表明,采用归一化变步长分数间隔常数模盲均衡算法比传统的常数模算法和固定步长分数间隔算法收敛快,均衡性能有明显提高。

A study of variable step size blind fractionally spaced equalization based on normalized error function

According to the slow convergence speed of CMA,a fast blind equalization algorithm is presented in this paper. This algorithm quickens the speed of convergence and improves the algorithm performance by modulating the traditional error functions,adopting fractional space and variable step size. Results of computer simulation in shallow water acoustic channel indicate that by using this algorithm higher convergence rate and better performance are obtained compared to the traditional CMA algorithm and FSE-CMA.