Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Electrical & Systems Engineering

First Advisor

Saswati Sarkar


The secondary spectrum market where primaries (license holders) lease the secondaries (unlicensed users) in lieu of the financial remuneration can eliminate the inefficiencies of the "static" spectrum allocation policy. We redress some of the challenges that have inhibited the wide scale deployment of the secondary spectrum market.

We first consider a secondary spectrum market where the primaries quote their prices for their available channels at a single location. The transmission rates offered by the channels of primaries evolve randomly because of the fading and noise. The secondaries decide to buy among the channels based on the transmission rate and the prices. We formulate the problem as a non cooperative game with the primaries as players. Each primary selects a price based on its own channel state only, as it is unaware of the channel states of the other primaries. We show that under the unique NE strategy profile a primary prices its channel to render the channel which provides high transmission rate more preferable; this negates the perception that prices ought to be selected to render channels equally preferable to the secondary regardless of their transmission rates.

Next, we consider the setting where the secondary spectrum market operates over multiple locations. Each primary needs to select an independent set in a conflict graph and the price at each location. We consider two scenarios--i) the number of locations is small, and ii) the number of locations is large. We show that when the number of locations is small, in a symmetric NE strategy, each primary sells its channel to an independent set whose cardinality exceeds a certain threshold. The threshold also decreases as the transmission rate offered by the channel decreases. The symmetric NE is unique in a widely seen conflict graph-the linear conflict graph. In contrast, when the number of locations is large, a primary only sells its channel in the maximum independent set and the symmetric NE in not unique in the linear conflict graph.

Subsequently, we consider the setting where a primary owns a channel at a single location and can acquire the competitor's channel state information (C-CSI) by incurring a cost. Each primary now needs to decide whether to acquire the C-CSI or not and a price based on the information it has. We formulate the problem as a non cooperative game with two primaries as players and characterize the NE strategies. We first characterize the Nash Equilibrium (NE) of this game for a symmetric model where the C-CSI is perfect. We show that the payoff of a primary is independent of the C-CSI acquisition cost. We then generalize our analysis to allow for imperfect estimation and cases where the two primaries have different C-CSI costs or different channel availabilities. Our results show interestingly that the payoff of a primary increases when there is estimation error. We also show that surprisingly, the expected payoff of a primary may decrease when the C-CSI acquisition cost decreases when primaries have different availabilities.

Finally, we consider the setting where a primary allows multiple secondaries use the channel of a primary at a location. The interference must be limited at each primary-user terminal (primary-UT) in order to maintain a quality of service for each primary-UT. The secondary-base stations (secondary-BSs) are self-interested entities and only maximize their own utilities which makes it difficult to obtain a simple interference mitigation policy. We formulate the problem as a non cooperative coupled constrained concave game. We use the concept of the normalized Nash equilibrium (NNE) since it caters to the distributed setting. We develop a distributed algorithm which converges to the unique NNE for a large class of utility functions. In the distributed algorithm, the secondary-BSs do not need to exchange information among themselves, and the minimal cooperation from the primary-UTs. When the NNE is not unique or difficult to compute, we introduce the concept of WNNE which retains most of the properties of the NNE, but it can be computed easily compared to the NNE.