恒模算法(CMA)是盲多用户检测中利用高阶统计量作为代价函数的自适应算法,但其步长恒定,不能随通信信道实际情况变化。在CMA和线性约束恒模算法(LCCMA)的基础上,将自适应步长引入到CMA中,得到自适应步长恒模算法(ASCMA),该算法能够根据通信信道情况自适应调节滤波器的步长,从而改善CMA算法的性能。通过MATLAB仿真表明,在白噪声通信信道环境中,ASCMA与LCCM相比,其输出信干比(SIR)、收敛速度等性能都有了较大的提高。 Constant modulus algorithm (CMA) is an adaptive algorithm using high-order statistics as a cost function in blind multi-user detection, but its step size is constant and can not change with the actual situation of the communication channel. Based on CMA and Linear Constrained Constant Modulus Algorithm (LCCMA), adaptive step size is introduced into CMA to get adaptive step size constant algorithm (ASCMA), which can adaptively adjust the filter according to the communication channel condition Step size, thereby improving the performance of the CMA algorithm. The simulation results of MATLAB show that in the white noise communication channel environment, the performance of ASCMA is much higher than that of LCCM in terms of output SIR and convergence speed.