No abstract

With the coming of 5G, the structure of cellular network becomes more and more complex, which provides a great challenge to network modeling. In order to model the real cellular network accurately, researchers proposed many models based on stochastic geometry, such as Strauss process (SP), Poisson hard-core process (PHCP), Thomas cluster process (TCP) and Marten cluster process (MCP). In this paper, we creatively divided the urban areas into three kinds by the indicator of their BS density, which are low-density, middle-density and high-density. Through simulation analysis, we found TCP is more accurate in modeling the low-density city and the middle-density city, while MCP is more accurate in modeling the high-density city.

No abstract