Doctor of Philosophy (PhD)


Computer Science

Document Type



The proliferation of wireless access and applications to the Internet and the advent of a myriad of highly evolved portable communication devices; creates the need for an efficiently utilized radio spectrum. This is paramount in the licensed and unlicensed radio frequency bands, that spawn an exponential growth in Dynamic Spectrum Access (DSA) research, Cognitive Radio (CR) and Cognitive Radio Networks (CRN) research. DSA research has given way to the paradigm shift toward CR with its dynamic changes in transmission schemas. This paradigm shift from a fixed and centralized frequency spectrum environment has morphed into a dynamic and decentralized one. CR provides wireless nodes the capability to adapt and exploit the frequency spectrum. The spectrum information obtained is scanned and updated to determine the channel quality for viability and a utilization/availability by the licensed (primary) user. To take advantage of the CR capabilities, previous research has focused on a Common Control Channel(CCC) for the control signals to be used for spectrum control. This utilization generates channel saturation, extreme transmission overhead of control information, and a point of vulnerability. The traditional designs for wireless routing protocols do not support an ad hoc multi-hop cognitive radio network model. This research focuses on a real world implementation of a heterogeneous ad hoc multi-hop Cognitive Radio Network. An overall model, coined Emerald, has been designed to address the architecture; the Medium Access Control layer, E-MAC; and the network layer, E-NET. First, a Medium Access Control(MAC) layer protocol is provided to avoid the pitfalls of a common control channel. This new design provides CRNs with network topology and channel utilization information. Spectrum etiquette, in turn, addresses channel saturation, control overhead, and the single point of vulnerability. Secondly, a routing model is proposed that will address the efficiency of an ad hoc multi-hop CRN with a focus on the Quality-of-Service(QoS) of the point-to-point as well as end-to-end communication. This research has documented weaknesses in spectrum utilization; it has been expanded to accommodate a distributed control environment. Subsets of the model will be validated through Network Simulator-2(NS/2) and MatLab© simulations to determine point-to-point and end-to-end communications.



Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Kannan, Rajgopal