Towards a modular IoT network model: Low power radio domain use-case in linear-topology wireless sensor network

摘要

The Internet of Things (IoT) is a new paradigm that envisions ubiquitous connectivity. It provides new ways in which distributed systems can be imagined with devices communicating with each other and also with humans. The data provided by such a large, heterogeneous network can be leveraged to provide enhanced services to society and solve problems previously deemed impossible. Most devices in IoT are resource constrained hence their performance metrics like throughput, latency and energy requirements are critical design parameters. There needs to be an efficient way in which deployments of IoT can be quickly modelled and their metrics predicted for further tuning and optimization. Currently, most modelling approaches concentrate on homogeneous mesh-like networks. The approaches also tend to model the network as a single phenomenon which is seldom the scenario in an IoT deployment. In this work we utilize a linear chain-like topology of wireless sensors utilizing ContikiMAC as a use-case. We introduce a modular network modelling approach to predict the metrics of the network deployment which is typically seen in pipeline monitoring, water quality monitoring, on-board freight train status monitoring among other scenarios. The model predicts behavior of nodes in a linear chain-like topology utilizing ContikiMAC protocol for multi-hop communication with a sink node.