Secondary Pricing of Spectrum in Cellular CDMA Networks

We study secondary pricing of spectrum in wireless cellular networks employing CDMA at the physical layer. We consider a primary license holder who aims to lease its spectrum within a certain geographic subregion of its own network. Such a transaction has two contrasting economic implications for the seller: On the one hand the seller obtains a revenue due to the exercised price, or rent, of the region. On the other hand, the seller incurs a cost due to (i) reduced spatial coverage of its network and (ii) possible interference from the leased region into the retained portion of its network. We formulate an optimization problem with the objective of profit maximization, and characterize its solutions based on a reduced load approximation that can be shown to be asymptotically exact. The form of optimal prices suggests charging the buyer per admitted call, in proportion with the interference it generates. The charged amount balances the corresponding loss of revenue incurred by the seller due to the influence of an admitted call. We numerically argue that this pricing approach yields better profit compared to some other simplistic techniques.