Parametric distance distributions for fixed access network analysis and planning

The access network displays the important particularity that the locations of the network components strongly depend on geometrical features such as road systems and city's architecture. This paper presents a model for access networks arising from stochastic geometry, which takes this geometric structure into consideration, and thus offers a relevant view on location-dependent characteristics such as point-to-point connections as shortest paths along the road system. Closed analytical formulas are derived for point-to-point distance distributions, explicitly depending on the morphology of this road system. Theoretical predictions are successfully compared to real data from fixed access networks deployed in dense urban areas.