Semantic Mobile Base Station Placement

Location of Base Stations (BS) in mobile networks plays an important role in coverage and received signal strength. As Internet ofThings (IoT), autonomous vehicles and smart cities evolve, wireless net-work coverage will have an important role in ensuring seamless connectivity. Due to use of higher carrier frequencies, blockages cause communication to primarily be Line of Sight (LoS), increasing the importance of base station placement. In this paper, we propose a novel placement pipeline in which we perform semantic segmentation of aerial drone imagery using DeepLabv3+ and create its 2.5D model with the help ofDigital Surface Model (DSM). This is used along with Vienna simulator for finding the best location for deploying base stations by formulating the problem as a multi-objective function and solving it using Non-Dominated Sorting Genetic Algorithm II (NSGA-II). The case with and without prior deployed base station is considered. We evaluate the basestation deployment based on Signal to Interference Noise Ratio (SINR)coverage probability and user down-link throughput. This is followed by comparison with other base station placement methods and the bene-fits offered by our approach. Our work is novel as it considers scenarios where there is high ground elevation and building density variation, and shows that irregular BS placement improves coverage.

GIMP-ML: Python Plugins for using Computer Vision Models in GIMP

This paper introduces GIMP-ML, a set of Python plugins for the widely popular GNU Image Manipulation Program (GIMP). It enables the use of recent advances in computer vision to the conventional image editing pipeline. Applications from deep learning such as monocular depth estimation, semantic segmentation, mask generative adversarial networks, image super-resolution, de-noising and coloring have been incorporated with GIMP through Python-based plugins. Additionally, operations on images such as edge detection and color clustering have also been added. GIMP-ML relies on standard Python packages such as numpy, scikit-image, pillow, pytorch, open-cv, scipy. Apart from these, several image manipulation techniques using these plugins have been compiled and demonstrated in the YouTube playlist (https://www.youtube.com/playlist?list=PLo9r5wFmpD5dLWTyo6NOiD6BJjhfEOM5t) with the objective of demonstrating the use-cases for machine learning based image modification. In addition, GIMP-ML also aims to bring the benefits of using deep learning networks used for computer vision tasks to routine image processing workflows. The code and installation procedure for configuring these plugins is available at https://github.com/kritiksoman/GIMP-ML.