K分布杂波中距离扩展目标的Wald检测

Wald test of range-distributed targets in K-distributed clutter

将距离扩展目标建模为子空间信号,用球不变模型模拟K分布杂波,提出了广义Wald检测算法。该算法是对待检测距离单元进行非相参积累,对杂波的纹理分量而言具有CFAR特性。首先对各个待检测距离单元分别检测,其输出的统计量是"白化"后的信号向信号子空间投影的能量和其向与信号子空间正交的噪声子空间投影的能量的比值来计算的,然后将各个待检测距离单元输出的统计量进行累加,形成最终的检验统计量。为了验证其有效性,通过Monte Carlo仿真了该算法的检测性能,并与文献[5]提出的自适应Wald检测器进行比较。

The generalized Wald test algorithm is derived to implement range-distributed targets detection embedded in K-distributed clutter which is modeled as a spherically invariant random process （SIRP）. The range-distributed target is modeled as a subspace random signal. The algorithm performs an incoherent processing over all the cells under test and is shown that it ensures CFAR property with respect to the unknown statistics of the clutter texture component. First of all, every cell under test is computed by the ratio of the energy of the whitened data that lies in the transformed signal subspace to the energy of it that lies in the orthogonal subspace, the so-called noise subspace. And then, the final decision statistics is the summation of the ratio in every cell under test. Performances of the proposed detector are assessed by means of Monte Carlo simulation strategy. In particular, the simulation results highlight that the proposed detector has better detection performances compared with adaptive Wald receiver which was proposed in [ 5 ].