Along with the increasing market penetration of mobile platforms in general, emphasis on applications providing location-specific information has also been increasing significantly. Using technology provided by third-geneartion and the upcoming fourth-generation mobile services which make localization of the device with almost one meter accurate possible, already became an organic part of data-transfer capabilities of such networks, it has become possible to create applications which can provide regionally optimised data for the user.
One of the most useful and typical exploitation of such features is meteorological data supply on-demand, where the user can be served by as specific data as possible. In this paper I propose a way to provide the user with an even more accurate, locally optimized data instead of the currently existing regional services.
Meteorological data sources, such as those encoded in METAR (Meteorological Airborne Report) format are freely available for the general public, refreshed every half hour, accessible via Internet through the FTP (File Transfer Protocol) server of NOAA (National Oceanic and Atmospheric Administration). Having these downloaded and interpolating the values for any given geographical location, by creating an average weighting of the surrounding measurement values based on the reciprocal distance of the user from all these points of reference, we can output the most likely, approximate value for that given location.
I developed the application for the Android platform, so it can handle localization and Internet-relay data communication totally independently of the actual hardware, while being one of the the most popular and widespread platform available globally today at the same time. Validation of the software, after implementation is carried out by the Android emulator, two case stdudies were implemented to check accuracy of information provided by the the software. Then, the implementd aplication run in a real hardware environment.