Package com.graphhopper.matching

Class HmmProbabilities

java.lang.Object
com.graphhopper.matching.HmmProbabilities
public class HmmProbabilities extends Object
Based on Newson, Paul, and John Krumm. "Hidden Markov map matching through noise and sparseness." Proceedings of the 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. ACM, 2009.

  • Constructor Details

    • HmmProbabilities

      public HmmProbabilities(double sigma, double beta)
      Parameters:
      sigma - standard deviation of the normal distribution [m] used for modeling the GPS error
      beta - beta parameter of the exponential distribution used for modeling transition probabilities

  • Method Details

    • emissionLogProbability

      public double emissionLogProbability(double distance)
      Returns the logarithmic emission probability density.
      Parameters:
      distance - Absolute distance [m] between GPS measurement and map matching candidate.

    • transitionLogProbability

      public double transitionLogProbability(double routeLength, double linearDistance)
      Returns the logarithmic transition probability density for the given transition parameters.
      Parameters:
      routeLength - Length of the shortest route [m] between two consecutive map matching candidates.
      linearDistance - Linear distance [m] between two consecutive GPS measurements.