Approximate model of the session service process for mmWave/sub-THz 5G/6G cellular systems

The uncertainty in user locations in the coverage area of 5G/6G millimeter wave (mmWave, 30–100 GHz) and sub-terahertz (sub-THz, 100–300 GHz) base stations (BS) force researchers to utilize complex queuing systems for modeling the service process of arriving sessions. Such systems need to account not only for blockage events that change the amount of resource requirements but also for random resource requirements of sessions even when the requested rate is constant. In this paper, we will assess the approximation induced by the use of a model that captures only the mean session requirements. We evaluate the accuracy of the approximation by utilizing the use—and system-centric performance measures—the session drop probability and the mean system resource utilization. Our numerical results illustrate that the approximation accuracy mainly depends on the parameters of the traffic arrival process and system capacity. The radio part specifics including the BS and UE antenna gains as well as BS transmit power does not produce any noticeable impact on the approximation. Accounting for the operational regime of the base station (BS), in general, the simple approximate Erlang model can be utilized as a model providing a rough lower bound on the session drop probability and an upper bound on the resource utilization. Thus, it can be used in performance optimization of 5G/6G mmWave/sub-THz systems deployments, where the closed-form expressions can be included as a part of the objective function. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.