论文部分内容阅读
提出一种适用于混响环境中的波束形成算法,算法对现有的线性约束最小几何功率波束形成算法进行了广义上的优化,采用迭代复加权最小二乘法对线性约束最小几何功率波束形成算法的权向量进行求解推导,保证了期望方向上始终保持单位增益,且对其他方向上的能量起着明显而稳定的抑制作用,克服了传统的基于二阶统计矩的波束形成器不适用于混响信号的不足,并加快算法的收敛速度。仿真结果表明,这种算法在混响环境下相比线性约束最小几何功率约束形成算法更有效,鲁棒性更强。
A beamforming algorithm suitable for the reverberation environment is proposed. The algorithm optimizes the existing linear constrained minimum geometry power beamforming algorithm in a generalized way. The iterative complex weighted least squares method is applied to the linear constrained minimum geometric power beamforming algorithm Of the vector to solve the derivation, to ensure that the expected direction always maintain unity gain, and the energy in other directions played a clear and stable inhibitory effect, to overcome the traditional second-order statistical moments based beamformer is not suitable for mixing Lack of ringing signal, and speed up the convergence of the algorithm. The simulation results show that this algorithm is more efficient and robust than the linear constrained minimum geometric power constrained algorithm in reverberant environment.