一种改进的线性受限共轭梯度常模盲多用户检测算法

The Modified Linearly Constrained Conjugate Gradient CMA for Blind Multi-user Detection

提出了一种改进的线性受限共轭梯度常模算法(M-LCCGCMA),其核心是采用在最优自适应步长的方法对算法进行优化,并推导出步长的解析形式。将本文算法在加性白高斯和多径衰落信道的环境中进行了仿真。仿真结果表明,该算法的信干比性能和误码率性能均比现有的自适应步长常模算法要好。

This paper proposes a Modified linearly constrained condition conjugate gradient constant modulus algorithm based on the adaptive step size （M-LCCGCMA）, which uses the best adaptive step size to optimize the LC-CGCMA and its analytical form is de- duced. This algorithm is simulated and compared with the existing adaptive step size CMA in additive white Ganssian channel and multi- path fading channel environments. Simulation results show that the proposed M-LCCGCMA can give better SIR and BER improvement than the existing one.