Date of Award

Fall 12-1-2017

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

College of Technology

Abstract

Internet of Things (IoT) is a fast-growing technological trend, which is expected to revolutionize the world by changing the way we do things. IoT is a concept that encourages all the electronic devices to connect to the internet and interact with each other. By connecting all these devices to the internet, new markets can be created, productivity can be improved, operating costs can be reduced and many other benefits can be obtained. In IoT architecture, often sensors and aggregators collect data and send to a cloud server for analyzing via the traditional cloud-server model. This client-server architecture is not adequate to fulfill the growing requirements of IoT applications because this model is subjected to cloud latency. This research proposed a distributed computing model called Distributed Shared Optimization (DSO) to eliminate the delay caused by cloud latency. DSO is based on swarm intelligence where algorithms are built by modeling the behaviors of biological agents such as bees, ants, and birds. Mobile Ad-hoc Network (MANET) is used as the platform to build distributed computing. The infrastructure-less and leader-less features of MANET make it the ideal candidate to build IoT with swarm intelligence. To test the theory, this research also built a simulation program and conducted multiple simulations on both DSO and client-server models. The simulation data was analyzed by descriptive statistics and One-Way ANOVA. This research found that there is a significant difference in computing time between DSO and client-server models. Further, Multiple-Regression technique was conducted on DSO simulation data to identify the effect sensors and data had towards DSO computing time.

Share

COinS