I am very new to GIS and I am diving into cartography and mapping because of an Android App that I am currently developing. I want to display environmental data, more precisely noise pollution data on a map in the App. But the data I received is quite complex and large and I am having trouble to properly simplify and export the data that it can easily be used with Open Street Maps on mobiles.
I have several versions of the data, one older shape file 23.5 MB, one newer, more accurate one with even 61.3 MB. When I export the files with QGis I get .kml of 66.4 MB and over 100 MB. I failed in simplifying geometry to reduce the file size since the size seems only to get smaller as the shapes stop looking too different from the original.
I heard of the concept to slice/intersect the shape, for example with shape files from district borders so the mobile device would only load that part of the large file when being in the respective district, with GPS on. I tried to do this with ArcGIS after having downloaded .kml of district borders, but this software seems too complex for me, unfortunately. Is this also possibly be done with QGIS?
What is the best way to boil a large layer of data down so it can easily be used for an Android app? I guess these large polygonal shapes can be expressed in a way simpler way to still get the values for a certain location?! The data, in this case a .kml of 66.4 MB imported via fusiontables onto Google Maps can be seen here: https://www.google.com/fusiontables/DataSource?snapid=S717313eWaJ
I hope some GIS expert has time to look at this and can guide me into the right direction.
Best Birgit
Edit1: Concerning tiling. I see that tiling works only for rastered images. The essence of using the .shp for the is the actual data that lies behind it - the app needs to be able to sort the data, e.g. show only values >x . In order for the device to know on which location which value has ben measured would tiling the shape only be a graphical solution?

