Economics of White Space Network

This thesis focuses on the interaction between technologies and economics in the database-assisted white space network, which is a promising paradigm of dynamic spectrum sharing, and can effectively improve the spectrum utilization and alleviate the spectrum scarcity. Motivated by several recent business practices, we study three types of different business models: spectrum market, information market, and hybrid spectrum and information market.;In the spectrum market model, spectrum licensees, through spectrum brokers acted by databases, lease the under-utilized (licensed) TV channels to unlicensed white space devices (WSDs) for secondary utilization. We first propose a contract-theoretic spectrum reservation framework to tackle the licensed TV channels reservation problem under demand stochasticity and information asymmetry. Second, we consider a competitive market with multiple WSDs, and study their competition in both spectrum inventory and pricing. Based on this analysis, we propose two different spectrum wholesale pricing strategies that maximize the database manager's profit and the total network profit, respectively.;In the information market model, databases sells the advanced information regarding (unlicensed) TV channels to unlicensed wireless devices in order to enhance the secondary spectrum utilization performance. We first study a non-competitive information market with a single database. We propose a general framework for evaluating the value of interference information to WSDs. We show that the information market processes the unique property of positive externality, as more WSDs purchasing the information service will increase the value of the service to each buyer. We systematically characterize the market equilibrium and show that the market equilibrium increases with the initial market share. Then we derive the optimal information pricing plan that maximizes the database profit or revenue based on the market equilibrium analysis. Second, we propose and study an oligopoly competitive information market, where multiple databases compete for selling information to WSDs. Our numerical results show that, a large degree of positive network externality would improve the databases' revenues and the system performance. In the hybrid spectrum and information market, the geo-location database serves as both a spectrum market platform and an information market platform. We study the interactions among the database operator, the spectrum licensee, and unlicensed WSDs systematically, using a three-stage hierarchical model. Analyzing such a three-stage model is challenging due to the coexistence of both positive and negative network externalities in the information market. Despite of this, we are able to characterize how the network externalities affect the equilibrium behaviors of all parties involved.