Dynamic spectrum allocation for uplink users with heterogeneous utilities

Dynamic allocation of spectrum prior to transmission is an important feature for next generation wireless networks. In this work, we develop and analyze a model for dynamic spectrum allocation, that is applicable for a broad class of practical systems. We consider multiple service providers (SPs), in the same geographic region, that share a fixed spectrum, on a non-interference basis. This spectrum is allocated to their customer end users for transmission to the SPs. Assuming that a user can obtain service from all the SPs, this work develops an efficient algorithm for spectrum allocation. The quality of service depends on system parameters such as number of users and SPs, the channel conditions between the users and SPs and the total transmit power of each user. The SPs have different efficiencies of reception. We adopt a user utility maximization framework to analyze this system. We develop the notion of spectrum price that enables a simple distributed spectrum allocation with minimal coordination among the SPs and users. Given the user utility functions and the system parameters, we characterize the spectrum price and the users´ optimal bandwidth allocations. Our work provides theoretical bounds on performance limits of practical operator to user based dynamic spectrum allocation systems and also gives insights to actual system design.