Date of Award
Dissertation - Restricted
Doctor of Philosophy (PhD)
Mathematics, Statistics and Computer Science
Ahamed, Sheikh I.
Radio Frequency IDentification (RFID) is an automatic identification technology that uses radio frequency to identify objects. Securing RFID systems and providing privacy in RFID applications has been the focus of much academic work lately. To ensure universal acceptance of RFID technology, security and privacy issued must be addressed into the design of any RFID application. Due to the constraints on memory, power, storage capacity, and amount of logic on RFID devices, traditional public key based strong security mechanisms are unsuitable for them. Usually, low cost general authentication protocols are used to secure RFID systems. However, the generic authentication protocols provide relatively low performance for different types of RFID applications. We identified that each RFID application has unique research challenges and different performance bottlenecks based on the characteristics of the system. One strategy is to devise security protocols such that application specific goals are met and system specific performance requirements are maximized.
This dissertation aims to address the problem of devising application specific security protocols for current and next generation RFID systems so that in each application area maximum performance can be achieved and system specific goals are met. In this dissertation, we propose four different authentication techniques for RFID technologies, providing solutions to the following research issues: 1) detecting counterfeit as well as ensuring low response time in large scale RFID systems, 2) preserving privacy and maintaining scalability in RFID based healthcare systems, 3) ensuring security and survivability of Computational RFID (CRFID) networks, and 4) detecting missing WISP tags efficiently to ensure reliability of CRFID based system's decision. The techniques presented in this dissertation achieve good levels of privacy, provide security, scale to large systems, and can be implemented on resource-constrained RFID devices.