An energy and time-saving task scheduling algorithm for UAV-IoT collaborative system

Abstract

Unmanned Aerial Vehicles (UAVs) are capable of providing significant potential to the Internet of Things (IoT) devices through sensors, cameras, GPS systems, etc. Therefore, the smart UAV-IoT collaborative system has become a current hot research topic. However, there are issues like resource allocation, security, and privacy preservation, trajectory optimization, intelligent decision, energy harvesting, etc. that need extensive research and analysis. In this article, we propose an energy-efficient and time-saving task scheduling algorithm that divides the IoT devices into certain clusters based on physical proximity. By utilizing the algorithm, cluster heads can apply an Auto Regressive Moving Average (ARMA) model to predict intelligently the timestamp of the arrival of the next task and associated estimated payments. Based on the overall expected payment, a cluster head can smartly advise the UAV about its time of next arrival. Simulation results demonstrate that our proposed Energy and Time-saving Task Scheduling (ETTS) algorithm show significant improvement in terms of energy (around 67%) as well as a delay (around 36%) over the Task scheduling for Indoor Environment (TSIE) and Time Division Multiple Access-Workflow Scheduler (TDMA-WS). The improvement in delay arises from the saved time of retransmissions. ARMA model basically tries to ensure that processing capacity of an UAV doesn't remain unutilized or under-utilized. The energy that UAV invests to arrive at one particular clusterhead, should be reciprocated by a full or close-to-full task queue of a clusterhead.