Design and Analysis of Distributed Resource Allocation Techniques for Large-Scale Dynamic Spectrum Access Networks

Graduation date: 2016Access restricted to the OSU Community, at author's request, from September 1, 2015 - September 1, 2016The widespread use of wireless devices that we have recently been witnessing, such as smartphones, tablets, laptops, and wirelessly accessible devices in general, is causing an unprecedented growth in the required amount of the wireless radio spectrum. On the other hand, the spectrum resource has, for the last several decades, been allocated to spectrum users in static manners. These static allocation methods were found to be very inefficient. This inefficient use of the spectrum resource, coupled with its limited availability nature, has led to a shortage in the spectrum supply.\ud Consequently, the concept of dynamic spectrum access (DSA) has emerged as an alternative allocation approach with great potential for solving this shortage problem. In the DSA context, there are two types of users: primary users (PUs) and secondary users (SUs). While PUs have exclusive access rights to use their licensed spectrum bands at all time, SUs are allowed to use these bands only opportunistically. That is, prior to using any licensed band, SUs must first sense the band to make sure that it is vacant. When a PU returns while SUs are using its band, SUs must vacate immediately.\ud In the first part of this dissertation, we address the problem of resource management in DSA networks. Specifically, we develop resource and service management techniques to support SUs with certain QoS (Quality of Service) requirements in large-scale DSA networks. The proposed techniques empower SUs to seek and exploit spectrum opportunities dynamically and effectively, thereby maximizing the SUs' long-term received service satisfaction. Our proposed techniques are efficient in terms of optimality, scalability, distributivity, and fairness.\ud In the second part of this dissertation, we model, characterize, and analyze the key performance metrics of these DSA networks. Specifically, we use a continuous-time Markov process to derive and analyze the forced termination and blocking probabilities of SUs under two realistic limitations. First, we investigate the impact of the channel handoff agility limitation on the performance. Here, due to this agility limitation, which is imposed essentially by hardware restrictions, SUs can only switch to their immediate neighboring channels whenever they have to due to, for e.g., the return of a PU. We show that such a limitation has great impact on the achievable performances of DSA networks. Second, we study the impact of the adjacent channel interference on the performance of DSA schemes, which often rely on the use of guard bands to handle such an interference.\ud We model and analyze the impact of guard band deployment methods on the performance of DSA schemes.\ud Our study serves as design guidelines for choosing appropriate guard band deployment methods when designing DSA schemes