论文部分内容阅读
传统恒模盲均衡技术由于在信号相位均衡中采用固定步长,致使相位点的收敛速度和收敛精度之间相互制约,其应用范围受到很大的限制。为使接收到的信号恢复效果达到最佳,对传统算法进行了改进,提出了一种以均方误差为判决依据,用时变步长代替固定步长的恒模盲均衡算法,同时对两种算法的均方误差曲线进行比较分析。实验结果表明,随着迭代次数的增加,改进算法恢复的星座图中相位点的收敛速度是原始算法的3倍,其迭代次数在2 000点以后便趋于稳定,使得相位之间的误差减少,信号的恢复效果明显。
The traditional constant modulus blind equalization technique has a very limited application range due to the use of a fixed step length in the signal phase equalization, which results in the restriction of the convergence speed and the convergence accuracy of the phase points. In order to optimize the received signal recovery effect, the traditional algorithm is improved. A constant modulus blind equalization algorithm based on the mean square error is proposed, which uses time-varying step instead of fixed step. At the same time, Algorithm mean square error curve comparative analysis. The experimental results show that the convergence rate of the phase points in the constellation chart recovered by the improved algorithm is three times that of the original algorithm with the number of iterations increasing, and the number of iterations tends to be stable after 2,000 points, so that the error between the phases is reduced , The signal recovery effect is obvious.