Abstract

The paradigm shift from static spectrum allocation to a dynamic one has opened many challenges that need to be addressed for the true vision of Dynamic Spectrum Access (DSA) to materialize. This dissertation proposes novel solutions that include: spectrum allocation, routing, and scheduling in DSA networks. First, we propose an auction-based spectrum allocation scheme in a multi-channel environment where secondary users (SUs) bid to buy channels from primary users (PUs) based on the signal to interference and noise ratio (SINR). The channels are allocated such that i) the SUs get their preferred channels, ii) channels are re-used, and iii) there is no interference. Then, we propose a double auction-based spectrum allocation technique by considering multiple bids from SUs and heterogeneity of channels. We use virtual grouping of conflict-free buyers to transform multi-unit bids to single-unit bids. For routing, we propose a market-based model where the PUs determine the optimal price based on the demand for bandwidth by the SUs. Routes are determined through a series of price evaluations between message senders and forwarders. Also, we consider auction-based routing for two cases where buyers can bid for only one channel or they could bid for a combination of non-substitutable channels. For a centralized DSA, we propose two scheduling algorithms-- the first one focuses on maximizing the throughput and the second one focuses on fairness. We extend the scheduling algorithms to multi-channel environment. Expected throughput for every channel is computed by modelling channel state transitions using a discrete-time Markov chain. The state transition probabilities are calculated which occur at the frame/slot boundaries. All proposed algorithms are validated using simulation experiments with different network settings and their performance are studied.

Graduation Date

2017

Semester

Fall

Advisor

Chatterjee, Mainak

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Computer Science

Degree Program

Computer Science

Format

application/pdf

Identifier

CFE0006890

URL

http://purl.fcla.edu/fcla/etd/CFE0006890

Language

English

Release Date

December 2017

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Share

COinS