说明：  ①产生值为0、1的随机序列，长度为2000，BPSK调制，通过冲激响应为h[1,0.8,0.3]的信道传送，信道高斯白噪声20dB,用LMS算法均衡输出，均衡器抽头个数为8，步长为0.03，训练序列长度为200。 ②绘制出星座图显示理想信号、均衡器输入信号、均衡器输出信号。 ③绘制出重复20次实验的瞬时误差平方的统计平均收敛曲线。更改LMS算法步长和抽头个数,观察瞬时均方收敛曲线有什么变化，分析步长因子和抽头个数变化对LMS算法性能的影响。
(Generate a random sequence with values of 0 and 1, with a length of 2000, BPSK modulation, and transmission through a channel with an impulse response of h[1,0.8,0.3]. The channel Gaussian white noise is 20dB, the output is equalized by the LMS algorithm, and the equalizer taps The number is 8, the step size is 0.03, and the training sequence length is 200. Draw a constellation diagram to show the ideal signal, equalizer input signal, and equalizer output signal. Draw the statistical average convergence curve of the instantaneous error square of the 20 repeated experiments. Change the step length and the number of taps of the LMS algorithm, observe any changes in the instantaneous mean square convergence curve, and analyze the influence of the step factor and the number of taps on the performance of the LMS algorithm.)