Accuracy of GPS Data. SharpGPS is Fine but not Smart Enough.
The task I’m currently work on is to increase the accuracy of GPS data. More concretely speaking I’m looking for approaches that would help to filter out the imprecise coordinates. It’s a well known problem that even if you don’t move, each second you get GPS coordinates from your reciever that deviate at up to 15 m. (GPS signal is quite good) from your current location. Thus such a simple task as to calculate the total distance that the user passed gets to a great challenge. Yes, you can ignore the coordinates that lay under a certain HODP threshold or to take into consideration only those coordinates that you get from a certain speed. Some of the heuristics I’m trying out at the moment. So if you have some suggestions I would highly appreciate your help. 
Looking in the internet for possible solutions, I didn’t find any helpful tips. But I came across a very nice API called SharpGPS. This is a GPS library for .NET 2.0 and .NET 2.0 Compact Framework. The current version is 0.5 alpha I liked the structural organization of the .NET Solution. The project comrises even some GUIs for testing the implemented features. This API can suite well for those developers who don’t want to invent bicycle once again and don’t want to waste their time with parsing NMEA protocol strings. For alpha 0.5 it’s a good result. I hope that in the future this API will contain some methods for increasing the quality of GPS precision and offer solutions for the problems I talked about at the beginning of the posting.