Latency in Mesh Networks

TL;DR

This study compares latency performance of Thread and Bluetooth Mesh wireless networks in static outdoor conditions, highlighting their strengths and limitations for low-latency applications like 5G and industrial automation.

Contribution

It provides empirical latency measurements for Thread and Bluetooth Mesh, revealing their performance differences and suitability for low-latency, dynamic environment applications.

Findings

Thread has lower latency than Bluetooth Mesh for larger payloads.

Thread meets some 5G latency targets, Bluetooth Mesh does not.

Both technologies require further optimization for dynamic environments.

Abstract

In this work, we consider latency in wireless networks, motivated by the need for reliable and low-latency wireless communication for applications such as vehicle-to-vehicle communication and industrial automation. The environment that these applications operate in is often highly dynamic, and, as a result, the wireless network technology used needs to be able to adapt to changing conditions, which makes mesh networking a good fit. Hence, in this work, we consider Thread and Bluetooth Mesh, two prominent mesh networking technologies. We measure one-way point-to-point latency using these technologies and show the cumulative probability distribution of the latency for varying payload size. We carry out the measurements under static conditions in an open-air environment. Our results show that Thread achieves lower latency than Bluetooth Mesh, especially for larger payloads. Furthermore, we show that Thread meets some of the latency targets set out for 5G. However, since Thread networks are slow to adapt to network changes, Thread may still not be suitable for application requiring low latency. Bluetooth mesh, on the other hand, does not meet any of the latency targets, although it is close when the payload is small (less than 11 bytes). Overall, there needs to be further optimization before these technologies can be used for reliable and low-latency communication in dynamic environments.