Machine Learning-based Mist Computing Enabled Internet of Battlefield Things

Abstract

The rapid advancement in information and communication technology has revolutionized military departments and their operations. This advancement also gave birth to the idea of the Internet of Battlefield Things (IoBT). The IoBT refers to the fusion of the Internet of Things (IoT) with military operations on the battlefield. Various IoBT-based frameworks have been developed for the military. Nonetheless, many of these frameworks fail to maintain a high Quality of Service (QoS) due to the demanding and critical nature of IoBT. This study makes the use of mist computing while leveraging machine learning. Mist computing places computational capabilities on the edge itself (mist nodes), e.g., on end devices, wearables, sensors, and micro-controllers. This way, mist computing not only decreases latency but also saves power consumption and bandwidth as well by eliminating the need to communicate all data acquired, produced, or sensed. A mist-based version of the IoTNetWar framework is also proposed in this study. The mist-based IoTNetWar framework is a four-layer structure that aims at decreasing latency while maintaining QoS. Additionally, to further minimize delays, mist nodes utilize machine learning. Specifically, they use the delay-based K nearest neighbour algorithm for device-to-device communication purposes. The primary research objective of this work is to develop a system that is not only energy, time, and bandwidth-efficient, but it also helps military organizations with time-critical and resources-critical scenarios to monitor troops. By doing so, the system improves the overall decision-making process in a military campaign or battle. The proposed work is evaluated with the help of simulations in the EdgeCloudSim. The obtained results indicate that the proposed framework can achieve decreased network latency of 0.01 s and failure rate of 0.25% on average while maintaining high QoS in comparison to existing solutions.