Towards Robust Synchronization in IoT Networks.

With the increasing deployment of large-scale distributed sensors and associated networks in the industrial IoT systems, highly reliable and precise clock synchronization gains ever increasing importance to ensure timeliness of the exchanged datagrams. While GPS (Global Position System), PTP (Precision Time Protocol) and NTP (Network Time Protocol) are prominent synchronization schemes, there are disadvantages in terms of accuracy achieved, hardware required, scalability, cost and associated failure modes. There is a need of robust, cost effective, scalable, easy to deploy, maintain and yet accurate synchronization method especially for IoT systems. This work focuses on implementation and validation of synchronization scheme using Kalman filter for IoT applications. The algorithm was implemented on a industrial platform for validation. The performance was assessed for its suitability in IoT based robust synchronization systems. The results show that the proposed approach and implementation is a promising candidate for synchronization in current and future IoT applications. The proof of concept confirms the feasibility of software only synchronization systems for IoT.