Improved OpenLR decoding using a stepwise increased deviation range

Exchanging information regarding traffic network related data, parties traditionally use a set of static references to the actual (road) network. Examples of such static references appear in TMC and DATEX II. These references use lists of road identifiers. The problem with this static road references approach is that the road network is continuously being changed which requires periodic administrative activity for all parties. Dynamic location referencing methods seek to reference locations without requiring a common definition of the road network. Examples of these methods are Agora-C, OpenLR, and Hidden Markov map matching. The challenge that these methods face is that each party can have a network definition that deviates in respect of the geographic location and of other descriptive information. Although the current OpenLR method takes possible deviations into account, its decoding methods can still result in false positives. This paper shows how the OpenLR decoding method can be improved by using a stepwise increased deviation range.