Time slotted channel hopping (TSCH) is the highly reliable and ultra-low power medium access control technology at the heart of the IEEE802.15.4e-2012 amendment to the IEEE802.15.4-2011 standard. TSCH networks are deterministic in nature; the actions that occur at each time slot are well known. This paper presents an energy consumption model of these networks, obtained by slot-based "step-by-step" modeling and experimental validation on real devices running the OpenWSN protocol stack. This model is applied to different network scenarios to understand the potential effects of several network optimization. The model shows the impact of keep-alive and advertisement loads and discusses network configuration choices.Presented results show average current in the order of 570 µA on OpenWSN hardware and duty cycles 1% in network relays in both real and simulated networks. Leaf nodes show 0.46% duty cycle with data rates close to 10 packets per minute. In addition, the model is used to analyze the impact on energy consumption and data rate by overprovisioning slots to compensate for the lossy nature of these networks.

No abstract

No abstract

Industrial low power networks are becoming the nexus of operational technologies and the Internet thanks to the standardization of networking layer interfaces. One of the main promoters of this shift is the IETF 6TiSCH WG, which addresses network management and IP integration of Time Synchronized Channel Hopping (TSCH) networks as those developed by the IEEE802.15.4 TG. The 6TiSCH WG is defining the operational interface and mechanism by which the network schedule can be distributed amongst the devices in the network. This operational sub-layer, called 6top, supports distributed scheduling and enables implementers to define the scheduling policy, only standardizing the distribution mechanism. This letter proposes a novel distributed scheduling policy based on the wellknown industrial control paradigm referred as Proportional, Integral and Derivative (PID) control. The proposed technique is completely decentralized, enabling each node to determine the number of cells to schedule to one another, according to its traffic demand. The mechanism is reactive to sudden or bursty traffic patterns, while staying conservative in over-provisioning cells.