基于噪声子空间特性的波束形成器设计

为高效、高性能地合成阵列接收信号, 提出基于噪声子空间特性的波束形成器设计。首先, 利用噪声的宽带分布特性, 对Capon空间谱的噪声区域做粗采样, 由此进行残留噪声补偿, 获得信号协方差矩阵的估计; 其次, 利用噪声子空间与信号子空间的正交性, 通过信号协方差矩阵分解构造出信号正交补投影算子; 最后, 利用噪声子空间与干扰子空间的正交性, 用该投影算子对观测信号进行处理, 获得干扰功率估计, 进而推导出干扰加噪声协方差矩阵和最终的波束形成器。仿真结果表明, 该方法仅耗费较少的快拍, 即可在大信噪比范围内实现较高信干噪比输出。

Beamformer design based on noise subspace characteristics

To efficiently synthesize array receiver signals with high performance, a beamformer design based on noise characteristics is proposed. Firstly, by utilization of the noise's wide-band distribution characteristic, the Capon spatial spectrum is coarsely sampled across the noise range, yielding a residual noise offset to estimate the signal covariance matrix. Secondly, on basis of the orthogonality between the noise subspace and the signal subspace, a signal orthogonal complement projector is constructed from the decomposition of this signal covariance matrix. Finally, in terms of the orthogonality between the noise subspace and the interference subspace, this projector is used to process the observed signals to acquire the estimation of the interference power, from which both the interference plus noise covariance matrix and the final beamformer can be derived. Simulation results show that the proposed method can achieve high signal to interference plus noise ratio across a wide signal to noise ratio range, only consuming small number of snapshots.