A Stochastic Mobility Prediction Algorithm for finding Delay and Energy Efficient Routing Paths considering Movement Patterns in Mobile IoT Networks

Abstract

In Mobile IoT Networks, the network nodes are constantly moving in a field, causing interruptions in the communication paths and, thus, generating long delays at the time of building a communication path from a source IoT node to the gateway (destination node). Communication interruptions affect the delay performance in delay-sensitive applications such as health and military scenarios. In addition, these IoT nodes are equipped with batteries, whereby it is also necessary to accomplish energy consumption requirements. In summary, a gateway node should not receive messages or packets coming from the IoT nodes with undesired delays, whereby it is pertinent to propose new algorithms or techniques for minimizing the delay and energy consumption experimented in the IoT network. Due to IoT nodes are attached to humans, animals or objects, they present a specific movement pattern that can be analyzed to improve the path-building with the aim of reducing the end-to-end delay. Therefore, we propose the usage of a mobility prediction technique based on a Stochastic Model to predict nodes’ positions in order to obtain minimum cost paths in terms of energy consumption and delay in mobile IoT networks. Our stochastic model is tuned and evaluated under the Markov-Gauss mobility model, considering different levels of movement randomness in order to test how the capability prediction of our proposal can impact the delay and energy consumption in mobile IoT networks in comparison with others routing algorithms.