自适应阵列处理典型算法的内在关系与统一框架

从经典维纳滤波器出发,以自适应干扰对消为纽带,推演出各种自适应阵列处理算法的等效原理。推导出降秩自适应滤波的统一算法结构,并提出线性约束正交投影算法。据此得出自适应阵列处理典型算法的内在关系,提出自适应阵列处理的统一框架,从而把单一约束和多线性约束(甚至静态方向图约束)、满秩和降秩、单级维纳滤波和多级维纳滤波有机统一起来,把对特殊干扰环境(相干干扰环境等)及期望信号特性(高阶累计量、循环平稳性等)下的处理视作预处理,构建起自适应阵列处理的统一理论基础。

On the relationship between adaptive array processing algorithms and the unified frame

Based on the wiener filter,various adaptive array processing algorithms are derived to be equivalent to each other by means of adaptive interference cancelling.The unified structure of reduced rank adaptive filtering algorithms is developed.Furthermore,linearly constrained orthogonal projection is proposed as a new reduced rank processing method with linear constraints.The unified frame of adaptive array processing is proposed based on the obtained algorithm system.Various circumstances,such as single or multiple linear constraints(even constraint on the quiescent beam pattern),full rank or reduced rank processing,one-stage or multistage wiener filter,can be consistent with the unified frame.Additional processing to special interference environments(e.g.coherent interferences) or special desired signal characteristics(e.g.high-order cumulants or cyclostationarity) can be regarded as the pre-processing.The unified frame is the universal theoretical basis of adaptive array processing.