Integrating Stochastic Geometry and ON/OFF Traffic Models: Toward Spatio-Temporal Analysis of Wireless Networks With Heterogeneous Services

Abstract

Wireless networks (4G, 5G, etc.) of today are witnessing a continuously increasing level of heterogeneity in terms of supported services and applications. Spatio-temporal modeling of wireless networks plays a crucial role, not only in design and analysis but also for critical performance metrics. Therefore, in this work, we develop a joint model for the analysis of wireless networks wherein users can request for different services and can generate multiple flows. By assuming single flow per user for a given service request, we derive the packet arrival probability of the user by using ON/OFF traffic model. Moreover, we derive the probability mass function (pmf) for the number of active users under a base station (BS). By assuming a heterogeneous traffic scenario wherein a user can generate multiple flows for different services, we derive the pmf for the number of active users under a BS. To analyze the performance of wireless network, we derive critical performance metrics which include success probability, mean delay and stable region probability. For computationally efficient implementation of the derived expressions, we also develop a algorithm which takes significantly less time. We benchmark the derived expressions against state-of-the-art literature.