Standards-based Internet of Things sub-GHz environmental sensor networks
Graeme M. Bragg · ePrints Soton (University of Southampton) · 2017
In recent years there has been shift in the use of wireless sensor networks from standalone systems that use bespoke methods of communication and data transfer to systems that use Internet standards and can interact more directly with the Internet. This has allowed wireless sensor networks to become a key enabler of the Internet of Things; however, the same is not true for environmental sensor networks as the focus of most existing research into Internet of Things wireless sensor networks has been on 2.4 GHz designs for indoor, urban and agricultural applications. In these applications, power, Internet connectivity and physical access are less of a challenge when compared to a typical environmental sensor network. Environmental sensor networks are used for monitoring natural processes and are generally deployed in harsh, remote environments where these factors are more of a concern. Sub-GHz radios are commonly used for communication due to their increased range and desirable propagation characteristics. Unlike wireless sensor networks, environmental sensor networks have been slow to adopt Internet standards and have continued to rely on bespoke methods of communication and data transfer, keeping their usability low. This has impeded the adoption of environmental sensor networks for earth sciences research. This thesis investigates whether the Internet standards that have helped to make wireless sensor network an important part of the Internet of Things can be applied to sub-GHz environmental sensor networks. It is demonstrated that 6LoWPAN can successfully be used with an 868 MHz network in a series of real-world deployments in the Highlands of Scotland that collected usable earth science data and facilitated research in other fields. Additionally, the suitability of these standards for real-world networks is assessed in terms of energy, throughput and latency performance and compared to a theoretical 2.4 GHz network. An publicly available open source Contiki radio driver for the CC1120 was developed as part of this work. Additionally, timing parameters for using ContikiMAC with 868 MHz radios were determines and shared with researchers at other institutions, facilitating further research into sub-GHz IoT ESNs by other researchers.