Principal component analysis (PCA) is a commonly used pattern analysis method that maps high-dimensional data into a lower-dimensional space maximizing the data variance, that results in the promotion of separability of data. Inspired by the principle of PCA, a novel type of learning discriminative matrix information geometry (MIG) detectors in the unsupervised scenario are developed, and applied to signal detection in nonhomogeneous environments. Hermitian positive-definite (HPD) matrices can be used to model the sample data, while the clutter covariance matrix is estimated by the geometric mean of a set of secondary HPD matrices. We define a projection that maps the HPD matrices in a high-dimensional manifold to a low-dimensional and more discriminative one to increase the degree of separation of HPD matrices by maximizing the data variance. Learning a mapping can be formulated as a two-step mini-max optimization problem in Riemannian manifolds, which can be solved by the Riemannian gradient descent algorithm. Three discriminative MIG detectors are illustrated with respect to different geometric measures, i.e., the Log-Euclidean metric, the Jensen--Bregman LogDet divergence and the symmetrized Kullback--Leibler divergence. Simulation results show that performance improvements of the novel MIG detectors can be achieved compared with the conventional detectors and their state-of-the-art counterparts within nonhomogeneous environments.
翻译:主要组成部分分析(PCA)是一种常用的模式分析方法,将高维数据映射到一个低维空间,使数据差异最大化,从而推动数据的分离。受五氯苯甲醚原则的启发,开发了一种新型的在不受监督的情景中学习歧视性矩阵信息几何(MIG)探测器,用于在非对等环境中的信号检测。Hermitian正偏向(HPD)矩阵可以用来模拟样本数据,而混杂的共变式矩阵则用一套HPD二级矩阵的几何平均值来估计,我们界定了一种预测,即将HRD矩阵映射成一个高维的多维多维元到一个低维和更具歧视性的矩阵,以便通过尽可能扩大数据差异,提高人平面矩阵的分离程度。在Riemannian 中,可使用Riemannian 元中双向微偏移梯度优化问题,这可以通过Riemannian 梯度测算法加以解决,而三种具有歧视性的MIGM探测器则用不同的几何测量度测量测量测量尺度,即Legal-Clodroal-Clibal-Cligal-Cligal-IGMmal-Smal-Smmleval-Syal-Syal-Smmmlational-Syal-Syal-Syal-Smmal-IG-IG-s-Smmal-Sy-s-S-Smmmal-Smalgal-Smlgal-Smal-Smalg-Symmal-S-S-S-Syal-S-S-Sy-Sy-s-s-IG-s-IG-S-Slg-S-S-S-S-Sm-Sm-Sm-Sm-Sm-Sm-Sm-Sl-Smal-Smmal-Smal-Smal-IG-IG-Sl-Smal-Smal-Smal-Smal-Smal-IG-IG-S-S-Smal-IG-IG-IG-Smal-C-C-Smal-