Empirical Comparison of the Energy Consumption of Cellular Internet of Things Technologies

Abstract

Today we are living in a society where an ever increasing number of devices or things are being connected to the Internet. Many of these devices are battery powered with the long term success of the business case justifying the deployment of these devices, dependent upon the ability of these devices to operate autonomously for extended periods without intervention to replace batteries. Utility companies delivering water and gas are seeking 10 year plus lifetimes from devices without a truck roll to replace batteries. The current industry understanding of the 5G technologies, NB-IoT and eMTC, is NB-IoT offers lower power consumption but there is little evidence for that belief. This paper uses standard load models to systematically investigate the energy demands of the two different radio access technologies under realistic usage patterns. The paper also investigates some key factors influencing energy consumption and whether a device could operate from an ideal 5 Wh battery for 10 years. In contributing to the field of knowledge in this area, we set out to compare 3GPP IoT radio access technologies using techniques aligned with existing 3GPP energy consumption evaluation methodologies. In adopting that approach, contrary to established industry opinions and messaging, we found when operating as an eMTC device, our radio module consumes less power than when it was operating as an NB-IoT device. Irrespective of it’s operating mode, our radio module was unable to achieve a 10 year battery life from an ideal 5 Wh battery using a typical IoT traffic model. Most interestingly, we observed minimal change in energy consumption whether the payload size was 1 byte or 1400 bytes.

Keywords

• Internet of Things
• LTE
• IoT
• eMTC
• NB-IoT
• LTE-M