Thesis (Ph. D.)--Ohio State University, 2007. / Title from first page of PDF file. Includes bibliographical references (p. 148-153).

No description available.

No description available.

We start by discussing fuzzy sets and the algebra of fuzzy sets. We consider some properties of fuzzy modeling tools. This is followed by considering the Mamdani and Sugeno models for designing fuzzy controllers. Various methods for using sets of data for desining controllers are discussed. This is followed by a chapter illustrating the use of genetic algorithms in designing and optimizing fuzzy controllers.Finally we look at some previous applications of fuzzy control in telecommunication networks, and illustrate a simple application that was developed as part of the present work.

Fuzzy control is a rule based type of control that aims to imitate the human's ability to express a control policy using linguistic rules, and to reason using those rules to control a system. Fuzzy control is nonlinear and not dependent on a precise mathematical description of the plant, and is therefore more easily applied to systems such as industrial processes that are hard to model. An overview is given of the fuzzy controller, along with descriptions of applications and theoretical approaches to designing and analyzing the controller. The self-organizing controller is able to generate or modify its rules in real time based on the system performance. It was tested to determine how well it was able to learn a quality control policy. The self-organizing controller was found to exhibit poor steady state performance, and to be equally likely to learn poor control as to learn good control. It was not found to eliminate the need for careful tuning of the controller parameters and gains. / Master of Science

Various established and emerging applications of RFID technology have been and are being implemented by companies in different parts of the world. However, RFID technology is susceptible to a variety of security and privacy concerns, as it is prone to attacks such as eavesdropping, denial of service, tag cloning and user tracking. This is mainly because RFID tags, specifically low-cost tags, have low computational capability to support complex cryptographic algorithms. Tag cloning is a key problem to be considered since it leads to severe economic losses. One of the possible approaches to address tag cloning is using an intrusion detection system. Intrusion detection systems in RFID networks, on top of the existing lightweight cryptographic algorithms, provide an additional layer of protection where other security mechanisms may fail. This thesis presents an intrusion detection mechanism that detects anomalies caused by one or more cloned RFID tags in the system. We make use of a Hybrid Fuzzy Genetics-Based Machine Learning algorithm to design an intrusion detection model from RFID system-generated event logs. For the purpose of training and evaluation of our proposed approach, part of the RFID system-generated dataset provided by the University of Tasmania’s School of Computing and Information Systems was used, in addition to simulated datasets. The results of our experiments show that the model can achieve high detection rates and low false positive rates when identifying anomalies caused by one or more cloned tags. In addition, the model yields linguistically interpretable rules that can be used to support decision making during the detection of anomaly caused by the cloned tags.

No description available.

published_or_final_version / Industrial and Manufacturing Systems Engineering / Doctoral / Doctor of Philosophy

published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

Call number: LD2668 .T4 1986 P68 / Master of Science / Computing and Information Sciences