A new approach to assess the orthogonality of two-dimensional (2-D) separation systems based on conditional entropy is developed. It considers the quantitative distribution of peaks in the entire separation space such that the orthogonality obtained is independent of the number of peaks observed for each separation technique. Therefore, it can be used to compare the orthogonality of different 2...

This paper presents a radar target recognition method using kernel locally linear embedding (KLLE) and a kernel-based nonlinear representative and discriminative (KNRD) classifier. Locally linear embedding (LLE) is one of the representative manifold learning algorithms for dimensionality reduction. In this paper, LLE is extended by using kernel technique, which gives rises to the KLLE algorithm...

For frequency hypercubes of dimension d > 2, we discuss several generalizations of the usual notion of pairwise orthogonality. We provide some constructions for complete sets of orthogonal frequency hypercubes. 1. Strong orthogonality for frequency hypercubes In this paper we will examine strong forms of orthogonality for frequency hypercubes. The standard de nition requires that each ordered p...

The paper examines the impact of Rician fading on the orthogonality of dual-polarized MIMO channels in a measured urban macrocellular scenario. The measurements confirm the previous finding that the Rician K-factors of the crosspolarized channels are strongly correlated to those of the copolarized channels. However, the Rician K-factors do not in general give direct indications on the orthogona...

We study the worst case error of kernel density estimates via subset approximation. A kernel density estimate of a distribution is the convolution of that distribution with a fixed kernel (e.g. Gaussian kernel). Given a subset (i.e. a point set) of the input distribution, we can compare the kernel density estimates of the input distribution with that of the subset and bound the worst case error...

We study the worst case error of kernel density estimates via subset approximation. A kernel density estimate of a distribution is the convolution of that distribution with a fixed kernel (e.g. Gaussian kernel). Given a subset (i.e. a point set) of the input distribution, we can compare the kernel density estimates of the input distribution with that of the subset and bound the worst case error...

We present a framework for feature extraction and mode decomposition of spatiotemporal data generated by ergodic dynamical systems. Unlike feature extraction techniques based on kernel operators, our approach is to construct feature maps using eigenfunctions of the Koopman group of unitary operators governing the dynamical evolution of observables and probability measures. We compute the eigenv...