Traffic modelling of an integrated 5G/Wi-Fi network with generally distributed user-dwell times

Abstract

A traffic modelling method is proposed for an integrated 5G/Wi-Fi network with generally distributed user-dwell times. The Wi-Fi traffic can move flexibly between 5G cells and Wi-Fi cells through overflow, vertical handoff, horizontal handoff, and take-back operations. Various user dwell times are modelled by general distributions and the channel holding times are derived in terms of the Laplace transforms of the dwell times. The various arrival rates and related performance measures are derived through an algorithm of fixed-point equations. Multi-dimensional Markov processes are used to model the Wi-Fi traffic and 5G traffic and derive the steady-state probabilities through stationary symmetric queues. The traffic modelling method is applicable to the next generation of cellular systems integrated with other types of networks such as sensor networks, intelligent vehicle networks and Internet of Things applications.