Computing Degree Show 2015

Cellular Geospatial Network Shapes

efficient generation and persistence using large datasets

Cellular geospatial network shapes are an effective visualisation tool to estimate the coverage of a given mobile network. Such shapes are formed using tower and device metadata sent by mobile devices over cellular networks. If this data is received and not processed immediately, a large backlog of records can accumulate over time.

The aim of this project is to incorporate an existing unprocessed dataset containing over 120 million entries of network status data into an accessible system. Through an agile-based approach, a system was developed to (1) process and persist network shapes within a database, (2) return shapes using a web-based Application Programming Interface, and (3) display shapes along with relevant metadata within a web-based user interface.

The final solution is a combination of four interconnected applications sitting on top of a SQL database back-end. The system utilises a wide variety of powerful, unique frameworks including Fluent NHibernate, AngularJS and the Google Maps JavaScript API. A typical user of the system can search for a network shape by a unique network identifier and view metadata for the polygonal shape vertices.


Student:
Emily McDonald

Website:
emilygmcdonald.com