Rate Adaptation and Latency in Heterogeneous IoT Networks

Abstract: This paper studies the effect of rate adaptation in time slotted Internet of things (IoT) networks. For a given time slot duration and packets size, rate adaptation necessitates packet fragmentation to fit the time slot duration. Accounting for the quality and time resolution of the underlying traffic, this paper characterizes the tradeoff between transmission rate and packet latency in IoT networks. Using tools from stochastic geometry and queueing theory, a novel mathematical framework is developed for static and dynamic rate adaptation schemes. The results show that there is an optimal static rate that minimizes latency, which depends on the network parameters. Furthermore, the dynamic rate is shown to be resilient to different variations in the network parameters without sacrificing packet latency.