How to smooth geolocation data (with Kalman filter)?

I've prerecorded geolocation data streams produced by smartphones and would like to smooth the data as the streams produced by different devices contain varying degrees of noise. On the short term I'd just like to do quick, one-off analysis, assess the results and plan future actions (e.g. implement smoothing for location data streams in realtime). In addition to geolocation each datapoint includes heading, speed and accuracy. What would be the simplest / fastest way to do the analysis e.g R other tools? How do I build a multidimensional model that would incorporate all the mentioned data?