Electronic Thesis and Dissertation Repository

Thesis Format

Monograph

Degree

Master of Science

Program

Computer Science

Collaborative Specialization

Artificial Intelligence

Supervisor

Bauer, Michael A

Abstract

Smart cities are one of the most active research fields in the world, due to the various benefits and challenges associated with their implementation. A major challenge for smart cities is processing and transporting the huge volumes of data generated by the sensor network layer, which builds the fundamental physical layer. Ongoing research is needed to address the computational challenges arising in smart city environments, particularly to help ensure efficient operations around the sensor layer. To address this challenge, a novel physical layer framework is proposed that intelligently learns the behavior of the physical system to enable the agents to control the sensors effectively and achieve the predefined sensor(s) and environment objectives. In addition, a novel machine learning semi-supervised Bayesian learning algorithm is proposed to learn and predict the sensors' behaviors to efficiently manage the energy consumption in houses. The novel model and algorithm were used in various diverse simulations, where results demonstrated their effectiveness in managing the energy consumption in different house settings and smart city environments.

Summary for Lay Audience

As cities continue to expand and the urban population grows, there is an increased need to provide new solutions to help humans live a better life and to boost productivity in cities. One solution includes integrating diverse smart devices and tools in cities to provide a smart city environment. However, the bigger the city, the more smart devices and tools are required for the smart city environment, and the more data is produced. The huge volume of data has to be processed and transported through a very important layer in smart cities called the physical layer, creating a challenge that can affect the efficiency of smart cities. To address this challenge, a novel physical layer framework is proposed that intelligently learns the behavior of the physical system in a smart city, which enables the smart devices around the city to be controlled effectively, achieving the pre-determined environment objectives.

Furthermore, controlling the energy consumption is vital for having a sustainable smart city, due to the lack of resources and the rapid growth of the world’s population. A potential solution includes setting an energy consumption limit threshold for each house across the smart city, in order to try to manage the energy consumption. Another solution includes adding intelligence to the smart city, by utilizing a technique called Machine Learning (ML). ML is a collection of theories and methods that enable computers to learn from data and make predictions without being explicitly programmed. In this thesis, a novel ML-based algorithm is proposed that can learn and predict the sensors’ behaviors to efficiently manage the energy consumption in houses, and keep the energy consumption below the limit threshold for each house while keeping citizens fulfilled. The novel model and algorithm were used in various diverse simulations, where results demonstrated their effectiveness in managing the energy consumption in various different house settings and smart city environments.

Download

Share

COinS