Joint spectrum pricing and admission control for heterogeneous secondary users

This paper solves the problem of long-term revenue maximization of a spectrum database operator, through joint pricing of spectrum resources and admission control of secondary users. A unique feature that we consider is the stochastic and heterogeneous nature of secondary users' demands. We formulate the problem as a stochastic dynamic programming problem, and consider the optimal solutions under both static and dynamic prices. In the case of static pricing, we constrain the prices to be time-independent while allowing the admission control policies to be time dependent. We show that in most cases a stationary (time independent) admission policy is in fact optimal in this case. We further look at the general case of dynamic pricing, where both the prices and admission control policies can be time dependent. We show that the flexibility of dynamic pricing can significantly improve the operator's revenue (by more than 30%) when secondary users have high demand elasticities.