Competition with Dynamic Spectrum Leasing

Dynamic spectrum leasing can greatly enhance the spectrum efficiency and encourage more flexible services in the spectrum market. This paper presents a detailed analytical study of the strategic interactions of two competing secondary network operators (duopoly) who need to make optimal investment (leasing) and pricing decisions while taking secondary endusers' heterogeneous wireless characteristics into consideration. The operators need to determine how much to lease from the spectrum owner, and compete to sell the spectrum to secondary users to maximize their individual profits. We model the system as a three-stage multi-leader dynamic game. Both the operators' equilibrium investment and pricing decisions turn out to have nice threshold properties. Each secondary user receives a fair equilibrium resource allocation that only depends on the leasing cost of the operators and is independent of other users' channel conditions and transmission powers. To further understand the impact of competition, we compare the duopoly equilibrium result with the coordinated case where the two operators cooperate to maximize the total profit. We show that the Price of Anarchy of the two operators' total profit is 82% with symmetric leasing costs, i.e., the maximum loss of the total profit due to competition is no larger than 18%. We also show that competition always leads to better payoffs for users compared with the coordinated case.