Modeling of Time Geographical Kernel Density Function under Network Constraints

Time geography considers that the probability of moving objects distributed in an accessible transportation network is not always uniform, and therefore density function applied to quantitative time analysis needs consider actual constraints. Existing methods construct a kernel under constraints based on principle least effort each point shortest path between anchor points has same value. This, however, ignores attenuation effect with distance according first law geography. For this reason, article studies framework unity geography, it establishes mechanism for fusing extended traditional model point, thereby forming can approximate theoretical prototype Brownian bridge providing basis reducing uncertainty estimation space. Finally, empirical comparison taxi trajectory data shows proposed effective.