The objective of the proposed research is to develop the foundations for resource allocation in femtocell networks, and evaluate the performance of the solutions using real testbeds. Wireless resource allocation in femtocell-based networks is particularly challenging due to four main reasons: (1) interference between femtocells and between femto- and macro-cells; (2) variable and unpredictable delays in signaling and data communication between each femtocell and the operator's network; (3) limited bandwidth for signaling over a wireless channel to a large number of femtocells; and, (4) involvement of three parties with differing priorities: the operator, the users, and the femtocell owners. The specific research tasks in this project are: (1) For legacy systems, develop distributed solutions for both downlink and uplink scheduling and handover under co-deployment of macro- and femto-cells that do not require any changes to existing hardware and standards; (2) Develop distributed, adaptive and self-organizing solutions (unconstrained by legacy requirements) for resource allocation by using a powerful tool from statistical physics, called Glauber dynamics; and (3) Design mechanisms to facilitate offline truthful auctions in practical settings involving the end-users, the femtocell owners, and the operator. The solutions will be implemented and evaluated on two testbeds under development at University of Michigan and at Ohio State University. The proposed research has the potential to significantly impact the cellular industry and end-users. Given the high penetration rates of cell-phones worldwide, this research has an immediate and widespread impact.
