University of Illinois Urbana-Champaign

IoT virtualization for animal simulations

Li, Haoxiang

This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.

Permalink

https://hdl.handle.net/2142/107775

Description

Title
IoT virtualization for animal simulations
Author(s)
Li, Haoxiang
Contributor(s)
Caesar, Matthew
Issue Date
2020-05
Keyword(s)
  • Internet of Things
  • IoT
  • Virtual Reality
Abstract
IoT, also known as Internet of Things, is a popular topic nowadays. Devices of various scales, complexities, and functionalities are deployed for different tasks. However, there exist technical difficulties in many circumstances where physical deployments are challenging or require verifications beforehand. For example, it would be hard for a group of high school students, who have no prior knowledge of circuits, to physically implement even the simplest circuit design. Devices that would be massively deployed without human supervision, such as ocean quality sensors, will also require verification in a simulation. Professor Matthew has been conducting research on virtualizing IoT devices, where he aimed to provide a platform to construct virtual circuits and deploy them in the virtual world. This platform allows users to monitor simulated inputs and outputs through virtual consoles and change parameters and components of the circuits in a virtual setup. To drive the virtual world, there should be an engine that simulates different aspects of the real world, such as movement, weather, etc. Taking the example of monitoring animals, there should be simulators that generate both the animals and environment. To simulate animals, several attempts were made to accurately model different animals like zebra and lions. We initially used an FSM-like model with fixed states to describe different actions and conditions of an animal, then moved forward to a learning-based model using different machine learning techniques, like linear regressions and neural networks. With a learning-based, or black-box model, animals act with more flexibility, instead of only allowing to be in the several fixed states. Circuit emulation is also an important feature in the overall virtualization process. There were two steps to create a virtual device. The first step involved physical testing and modeling of the devices, and the second step involved mapping them into scripts that describe their behaviors. These two components are essential parts for the entire IoT virtualization project since they generate all the simulations in the virtual world, and they will be actively updated as we introduce more features to the project.
Type of Resource
text
Language
en
Permalink
http://hdl.handle.net/2142/107775

Owning Collections

Senior Theses - Electrical and Computer Engineering PRIMARY
The best of ECE undergraduate research