Research on sparse adaptive beam-forming based on non-uniform norm for failure elements

In this paper, we present an Non-Uniform Norm constrained least mean square (NU-CNLMS) adaptive beam-forming algorithm which can improve the pattern for antenna array with failure elements by exploiting the sparsity property of it. The new algorithm modifies the cost function of classical constrained least mean squares algorithm in order to reach the optimum norm constraint for adaptive beam-forming system with different sparsity. So, we make a great trade-off between the sparsity and estimation bias generated by norm constraint. When employing this algorithm to antenna array, we can get a large aperture antenna array with few elements which is identical to the situation of failure elements existing. We investigate the effectiveness of new algorithm by simulation results which showed the performance of NU-CNLMS is superior to the other sparse adaptive beam-forming in terms of convergence speed and sparsity.