遗传算法在传感器阵列多波束图设计中的应用

由于在水声系统中实际阵列模型往往与理想阵列模型失配,导致基于理想阵列模型设计的波束图在实际应用中严重畸变,从而不能有效地抑制旁瓣干扰,严重影响系统性能.针对常规的波束设计方法应用到实际任意形状阵列时不能有效收敛到期望的最佳波束图,提出了一种基于遗传算法的多传感器阵列波束设计方法,以实现对任意形状阵列的多波束优化设计,获取期望特性的波束图或最佳可得的波束函数.结合运用水池实验测量的16元十字阵的数据,给出了方法基本步骤与多波束设计结果.设计波束满足了应用需要,说明了方法的应用可行性和有效性,且具有很强的宽容性.提出的算法对于实际多传感器系统的多波束设计具有重要意义.

Application of Genetic Algorithm to Design of Multi-beam Pattern of Sensor Array

As the presence of the mismatch errors between the actual array model and the ideal model, the realized beam pattern distorts seriously in actual underwater acoustic systems. The distorted beam pattern can′t suppress the side-lobes effectively and would greatly degrade the performance of the sonar system. The conventional method can′t always converge to the desired optimum beam pattern, so this paper presents an efficient pattern synthesis method using the genetic algorithm, which can be used in beam pattern design for arbitrary arrays. Combining the practical experimental data of 16 element array, this paper lists the basic steps of the method and the designed results of multi-pattern. The designed beam patterns satisfy the need of application greatly. The practical application results indicated that the proposed beam-forming method is feasible, effective and tolerant. The method proposed can provide a significant reference on beam-patterns design in the actual multi-sensor system.