Year
2024
Degree Name
Doctor of Philosophy
Department
School of Electrical, Computer and Telecommunications Engineering
Abstract
Internet of Things (IoT) networks require devices to handle a variety of tasks, including sensing, transmission, storage, and computation. The execution of these tasks requires efficient utilization of communication, computational, and energy resources. Further, tasks may have dependencies, which impose an order on task executions and allocation of resources. Moreover, they may require certain computation quality, which is determined by the amount of data collected and processed by a device/network.
To this end, this thesis addresses task allocation, computation and resource allocation problems in IoT networks. It first considers solar-powered devices that are capable of sharing their harvested energy. A key challenge is that devices are not aware of future energy arrivals and their energy level is coupled across time slots. Further, each task is required to be processed on a set of devices sequentially. The objective is to minimize the completion time of all tasks. To do so, a key problem is to optimize task scheduling and energy sharing decisions. In this respect, this thesis outlines a novel mixed integer linear program (MILP) to optimize the said quantities. Further, it outlines a heuristic algorithm that first schedules tasks based on their execution time and order before conducting energy sharing between devices.
Recommended Citation
Cui, Yuhan, Energy-Aware Task Allocation and Data Collection in Next Generation Wireless Networks, Doctor of Philosophy thesis, School of Electrical, Computer and Telecommunications Engineering, University of Wollongong, 2024. https://ro.uow.edu.au/theses1/1853
FoR codes (2020)
400602 Data communications
Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.