Hyperspectral Tensor Completion Using Low-Rank Modeling and Convex Functional Analysis.

Hyperspectral tensor completion (HTC) for remote sensing, critical for advancing space exploration and other satellite imaging technologies, has drawn considerable attention from recent machine learning community. Hyperspectral image (HSI) contains a wide range of narrowly spaced spectral bands hence forming unique electrical magnetic signatures for distinct materials, and thus plays an irreplaceable role in remote material identification. Nevertheless, remotely acquired HSIs are of low data purity and quite often incompletely observed or corrupted during transmission. Therefore, completing the 3-D hyperspectral tensor, involving two spatial dimensions and one spectral dimension, is a crucial signal processing task for facilitating the subsequent applications. Benchmark HTC methods rely on either supervised learning or nonconvex optimization. As reported in recent machine learning literature, John ellipsoid (JE) in functional analysis is a fundamental topology for effective hyperspectral analysis. We therefore attempt to adopt this key topology in this work, but this induces a dilemma that the computation of JE requires the complete information of the entire HSI tensor that is, however, unavailable under the HTC problem setting. We resolve the dilemma, decouple HTC into convex subproblems ensuring computational efficiency, and show state-of-the-art HTC performances of our algorithm. We also demonstrate that our method has improved the subsequent land cover classification accuracy on the recovered hyperspectral tensor.

Binary signaling design for visible light communication: a deep learning framework.

This paper develops a deep learning framework for the design of on-off keying (OOK) based binary signaling transceiver in dimmable visible light communication (VLC) systems. The dimming support for the OOK optical signal is achieved by adjusting the number of ones in a binary codeword, which boils down to a combinatorial design problem for the codebook of a constant weight code (CWC) over signal-dependent noise channels. To tackle this challenge, we employ an autoencoder (AE) approach to learn a neural network of the encoder-decoder pair that reconstructs the output identical to an input. In addition, optical channel layers and binarization techniques are introduced to reflect the physical and discrete nature of the OOK-based VLC systems. The VLC transceiver is designed and optimized via the end-to-end training procedure for the AE. Numerical results verify that the proposed transceiver performs better than baseline CWC schemes.