SkyLiTE: End-to-End Design of Low-altitutde UAV Networks for Providing LTE Connectivity

/in /by

Un-manned aerial vehicle (UAVs) have the potential to change the landscape of wide-area wireless connectivity by bringing them to areas where connectivity was sparing or non-existent (e.g. rural areas) or has been compromised due to disasters. While Google’s Project Loon and Facebook’s Project Aquila are examples of high-altitude, long-endurance UAV-based connectivity efforts in this direction, the telecom operators (e.g. AT&T and Verizon) have been exploring low-altitude UAV-based LTE solutions for on-demand deployments. Understandably, these projects are in their early stages and face formidable challenges in their realization and deployment. The goal of this document is to expose the reader to both the challenges as well as the potential offered by these unconventional connectivity solutions. We aim to explore the end-to-end design of such UAV-based connectivity networks particularly in the context of low-altitude UAV networks providing LTE connectivity. Specifically, we aim to highlight the challenges that span across multiple layers (access, core network, and backhaul) in an inter-twined manner as well as the richness and complexity of the design space itself. To help interested readers navigate this complex design space towards a solution, we also articulate the overview of one such end-to-end design, namely SkyLiTE– a self-organizing network of low-altitude UAVs that provide optimized LTE connectivity in a desired region.

Read more

Design and Comparison of Advanced Modulation Formats Based on Generalized Mutual Information

/in /by

Generalized mutual information (GMI) has been comprehensively studied in multidimensional constellation and probabilistic-shaped (PS) constellation together with different forward error correction (FEC) coding schemes. The simulation results confirm that GMI is an efficient and accurate tool to compare their post-FEC performance. In particular for uniformly geometric-shaped constellation, the pre-FEC Q-factor is highly correlated with GMI though the correlation is reduced at lower FEC coding rate. Furthermore, GMI can be used to design optimized constellation together with generalized pairwise optimization algorithm to mitigate the GMI loss in non-Gray-mapped constellation. The GMI-optimized 32QAM (opt32) shows ~0.5 dB signal-to-noise ratio improvement between 3 and 4 b/s GMI in both simulated and experimental results. Optimized two-dimensional 8 QAM is also designed to show the consistent GMI improvement over multi-dimensional 8 QAM-equivalent formats. In simulations, PS-64 QAM outperforms opt32 when a long sequence block is used in the distribution matcher.

Read more

A 4D Light-Field Dataset & CNN Architectures for Material Recognition

/in /by

We introduce a new light-field dataset of materials and take advantage of the recent success of deep learning to perform material recognition on the 4D light field. Our dataset contains 12 material categories, each with 100 images taken with a Lytro Illum, from which we extract about 30,000 patches in total. To the best of our knowledge, this is the first mid-size dataset for light-field images.

Read more

A Continuous Occlusion Model for Road Scene Understanding

/in /by

We present a physically interpretable 3D model for handling occlusions with applications to road scene understanding. Given object detection and SFM point tracks, our unified model probabilistically assigns point tracks to objects and reasons about object detection scores and bounding boxes. It uniformly handles static and dynamic objects, thus outperforming motion segmentation for association problems. It also demonstrates occlusion-aware 3D localization in road scenes.

Read more

WarpNet: Weakly Supervised Matching for Single-View Reconstruction

/in /by

Our WarpNet matches images of objects in fine-grained datasets without using part annotations. It aligns an object in one image with a different object in another by exploiting a fine-grained dataset to create artificial data for training a Siamese network with an unsupervised discriminative learning approach. The output of the network acts as a spatial prior that allows generalization at test time to match real images across variations in appearance, viewpoint and articulation. This allows single-view reconstruction with quality comparable to using human annotation.

Read more

Atomic Scenes for Scalable Traffic Scene Recognition in Monocular Videos

/in /by

We propose a novel framework for monocular traffic scene recognition, relying on a decomposition into high-order and atomic scenes to meet those challenges. High-order scenes carry semantic meaning useful for AWS applications, while atomic scenes are easy to learn and represent elemental behaviors based on 3D localization of individual traffic participants. We propose a novel hierarchical model that captures co-occurrence and mutual-exclusion relationships while incorporating both low-level trajectory features and high-level scene features, with parameters learned using a structured support vector machine. We propose efficient inference that exploits the structure of our model to obtain real-time rates.

Read more

Attribute2Image: Conditional Image Generation From Visual Attributes

/in /by

This paper investigates a novel problem of generating images from visual attributes. We model the image as a composite of foreground and background and develop a layered generative model with disentangled latent variables that can be learned end-to-end using a variational auto-encoder. We experiment with natural images of faces and birds and demonstrate that the proposed models are capable of generating realistic and diverse samples with disentangled latent representations. We use a general energy minimization algorithm for posterior inference of latent variables given novel images. Therefore, the learned generative models show excellent quantitative and visual results in the tasks of attribute-conditioned image reconstruction and completion. 

Read more