Date of Award

7-2018

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Engineering and Sciences

First Advisor

William H. Allen

Second Advisor

Munevver Subasi

Third Advisor

Susan Earles

Fourth Advisor

Ivica Kostanic

Abstract

Home automation and control systems as basic elements of smart cities have played a key role in the development of our homes environments . They have a wide range of applications in many fields at home such as security and monitoring, healthcare, energy, and entertainment applications. The improvement of humans’ living standards make people keep trying to delegate many of their needs to a home automation system. Such a system has been built with capabilities of predicting what the user intends to do in smart home environment. However, there are many issues that need more investigation and solutions, such as: 1) many researches adopt a specific application without integrating different verities of applications in one environment, 2) there is no study tries to show the real effect or even evaluates the implementation of predicted actions that have been established via homes intelligent gateway, 3) there is an interoperability issue due to using different kinds of home applications that have different protocols for message context. In this proposal, we will describe a new approach of an intelligent self-adaptive system that can precisely monitor a stakeholder behaviors and analyze his/her actions trying to anticipate a stakeholder behavior in the future. In addition, we will evaluate the real effect of a predicted actions after implementing them by an intelligent gate-way in a simulated home environment. The principle behind a prediction process is presented by analyzing a sequence of users interaction events with heterogeneous, and distributed nodes in the environment using an intelligent gateway. Predicting next stakeholder action can be process using certain analytical algorithms. The main novelties in the proposed approach are threefold: I) Presents a novel visualization model for a home area network (HAN) based Cyper-Physical System’s design pattern, that may helps academic researcher, enterprise companies, and developer to facilitate the most important services of such smart environment for a better quality of services. II) Developing a learning technique which is embedded in an intelligent gateway to build a model of users behavior and interactions to balance the needs of multiple users within a smart home environment. III) The proposed system shows a high level concept of how we can design an intelligent selfadaptive system in home environment that has the capability to provide stakeholder with local services, and to support a use of IoT paradigm concurrently.

Share

COinS