Modeling And Analysis Of Cellular Networks Based On Stochastic Geometry

With the advent of 5G era,the continuous growth of wireless network users and the emergence of new wireless network services,such as ultra-clear video,unmanned driving and virtual reality,have put forward higher requirements for the performance of wireless network.In the research of wireless network,the base station distribution modeling has an important influence on the performance analysis of wireless network.With the emergence of many non-traditional communication technologies,heterogeneous cellular network architecture has become an important way to improve network performance.Therefore,this paper mainly studies the fitting of cellular network base station distribution and the analysis of heterogeneous cellular network model.In the traditional cellular network modeling,hexagonal grid model is often used.However,this model cannot simulate the randomness of the base station distribution,and cannot truly model the base station distribution.Therefore,based on stochastic geometry theory,this paper selects real base station data of three cities to fit Hard core process,Strauss process,Matten cluster process and Thomas cluster process,and uses L function,coverage probability and other evaluation parameters to analyze the fitting degree of various point processes.The simulation results show that Matten cluster process is more suitable for fitting simulation of low density and medium density cities,and Thomas cluster process is more suitable for fitting simulation of high density cities.Based on the analysis of many non-traditional communication technologies,this paper chooses to introduce multi-cell cooperative technology into cellular network.Based on the above research conclusion,a two-tier heterogeneous cellular network model is constructed with the distribution of micro-base stations obeying the cluster process distribution,and the optimal power allocation scheme for network performance and energy efficiency is studied.The Lagrange-based energy efficiency optimization problem solution and greedy algorithm solution are derived.Based on the greedy algorithm,the user set of the greedy algorithm iteration is reduced by defining unstable users.The simulation results show that the new algorithm greatly improves the convergence speed of the greedy algorithm when the energy efficiency of the network is comparable.