A Spatiotemporal Model for Hard-Deadline Multistream Traffic in Uplink IoT Networks

Abstract

Ultrareliable and delay-intolerant message delivery is one of the key components for Internet of Things (IoT) and cyber–physical systems to support real-time control and real-time interaction. In this article, we provide a spatiotemporal framework that captures the packet loss rate (PLR) for large-scale grant-free uplink IoT networks with multiple types of hard-deadline traffic. An independent and a prioritized packet scheduling schemes are proposed and investigated for efficiently realizing frequency diversity. Tools from stochastic geometry and queuing theory are utilized to account for the macroscopic coverage probability and microscopic PLR. The expressions of devices’ coverage, steady state distribution, and PLR of services are derived for both scheduling schemes. Detailed system-level simulations are used to identify network design guidelines. The results show that the independent scheme provides better network scalability and consumes lower average transmit power at the devices, while the prioritized scheme enhances the PLR performance of high priority service and requires lower peak transmit power of devices.