M2M Rendezvous Redundancy for the Internet of Things

Abstract

Machine to Machine communication will play an important role in enabling the Internet of Things. Devices in both home and industrial settings will cooperate to create smart spaces. Devices will communicate with each other autonomously, enabling them to make decisions to reach an optimal environmental goal. The nature of their deployment will involve these devices having to operate under extreme conditions, such as during a fire, flooding or following an explosion. Events such as these would usually cause a reduction in the number of devices that originally constituted the system. Individual device loss might be temporary or permanent. Devices might not be active at the time of failure and may be sleeping due to the low power requirements to extend device longevity. It is important that when devices wake they can obtain the information they require and if a device is destroyed that its last known state is preserved in the system. Preserving the last know state of a device also provides a useful source of information for post failure analysis as well as the continued operation of the system during the event. This paper details our rendezvous state redundancy protocol for the Internet of Things. We then validate the use of position relative topologies when creating distributed hash tables for redundancy in wireless mesh IoT sensor networks. We then provide an evaluation of the keying mechanism used to provide optimal redundancy for given failure patterns.