An improved stochastic gradient constant modulus beam-forming algorithm

This paper analyzes the existing recursion formulas of SG-CMA, presents an improved SG-CMA with better convergence and stronger beam-forming gain. Considering the disadvantages of existing error functions, the algorithm makes some changes. It takes the sum of the signal modulus of each element as the expected value, which overcomes the fluctuation of signal modulus, and increases the application range and anti-noise capability. The signal modulus of an arbitrary array is built, then the SG-CMA and the improved SG-CMA are compared by simulation based on uniform linear array. The simulation results verify the effectiveness of the improved SG-CMA as well as its improvement on convergence and beam-forming gain.