一种权系数部分更新的变步长自适应多径干扰对消算法  

A variable step and adaptive multipath interference cancellation algorithm based on partial-update of weight coefficients

作　　者：姜冰磊 冯西安[1] 

机构地区：[1]西北工业大学航海学院,西安710072

出　　处：《振动与冲击》2016年第13期85-89,共5页Journal of Vibration and Shock

基　　金：国家自然科学基金资助项目(61271414);陕西省自然科学基金资助项目(2014JM2-6090)

摘　　要：浅水环境中,多径干扰严重影响目标探测的性能,导致虚警概率增加。有效的多径干扰对消算法是浅水环境中目标检测的关键。自适应LMS算法能有效抑制多径干扰,但当干扰持续较长或空间分布范围较大时,需要使用高阶次滤波器,计算量很大,给实际应用带来困难。通过分析浅水环境下多径干扰的形成机理和分布特点,建立其理论模型,并提出阵列模型下的权系数部分更新的变步长自适应多径干扰对消算法。该算法采用权系数的周期性部分更新算法和基于改进Sigmoid函数的变步长算法,较好解决了自适应干扰对消算法中计算量与收敛速度之间的矛盾。理论分析与仿真结果表明,该算法能在有效地降低计算量的同时,仍保持较快的收敛速度,并具有良好的多径干扰对消效果。In shallow water environment,multipath interference affects performances of underwater target detections and increases false alarm probability. Thus,the algorithm about multipath interference cancellation is a key to target detections in shallow water environment. The adaptive LMS algorithm can suppress multipath interference effectively. But when the disturbance lasts too long or its space distribution range is too wide,the adaptive LMS algorithm needs filters with higher order and huge amount of calculation to bring difficulties to practical applications. Here,through analyzing the formation mechanism and distribution characteristics of multipath interference in shallow water,its theoretical model was established. Then,a variable step adaptive multipath interference cancellation algorithm based on partial-update of weight coefficients was proposed. This algorithm resolveed the contradiction between calculation amount and convergence speed with periodically updating weight coefficients and using the variable step method based on improved Sigmoid function.Theoretical analysis and simulation results showed that the proposed algorithm has a good capacity for multipath interference cancellation,it can effectively reduce calculation amount and keep a faster convergence speed.

关 键 词：多径干扰 自适应干扰对消 部分更新 变步长 

分 类 号：TP15[自动化与计算机技术—控制理论与控制工程]