VCHILL 2: An Improved Implementation of Semantic Radar Data Distribution

Current radar data systems are organized primarily by date, so a user searching for radar files that satisfy some set of criterion must manually traverse a database of files, organized only by time of occurrence. This is a very tedious and inefficient process and often doesn't return the most meaningful of results.

This work presents the implementation of the front-end aspect of a new semantic radar data system known as VCHILL 2. This system allows users intuitive access to semantic meaning of radar data. This grants users the ability to search radar data by metadata, rather than manually traversing a database of radar files. Accessible metadata includes both low level information such as type of sweep and coverage, as well as higher level semantic attributes such as presence of a bow echo, or the cumulative distribution function of velocity. The system was developed using geospatial databases, HTML5 web servers, and a RESTful interface implemented in python. The front-end aspect of the system includes a user interface that in addition to connections to geospatial radar databases, cumulatively respond to user requests. The user interface was in particular, designed to be a RESTful interface which places constraints on data transmission between various aspects of the system. Due to this challenge, the system was constructed with a python framework that dynamically renders HTML5, receives/responds to user requests, and performs database searches using PostgreSQL. The python framework in essence, handles the key components that drive the front-end aspect of the VCHILL 2 system.

The front-end aspect of VCHILL 2 that the user interacts with is composed of visual HTML5, organizational CSS, and interactive Javascript. The visual and organizational components allow users to access the system within an internet interface and the interactive aspect allows for contact with the python framework that drives the system.

Combined, the front-end components of the system enables users the ability to display radar data, perform dynamic searches, and generate case files of relevant radar data through interaction with radar data and metadata databases. This functionality is unique to semantic radar systems.

This report will detail both the implementation of the front-end aspect of the system, as well as discussing the user interaction this methodology enables.