Optimal mobile assisted offloading and network price differenciation

Abstract

This thesis studies two problems related to cellular resource provisioning. The first problem deals with mobile-assisted data offloading, and the second problem analyzes network partitioning for differential pricing and Quality of Service. In the first part, we study an offloading mechanism for cellular networks in which mobiles with good cellular links can act as hotspots and assist other mobiles. We study throughput-optimal offloading and also a fair offloading strategy which we call proportional increment offloading. We show that the former problem can be reduced to a capacitated facility location problem (CFLP), whereas the latter can be solved by solving a sequence of CFLPs. We propose a belief propagation-based algorithm to solve CFLP, a well-known NP-complete problem. We also study a max-min fair offloading strategy, which maximizes the minimum throughput among all mobiles. Further, we propose a simple incentivizing scheme to entice mobiles to behave as hotspots. We primarily consider Point Coordination Function (PCF) based Wi-Fi access for offloading but also discuss Distributed Coordination Function (DCF) based access and related issues. We perform extensive simulation to evaluate the performance of the proposed algorithms and the effectiveness of mobile-assisted offloading. In the second part, we first study the effect of partitioning the network into two subnetworks. In the case of equal partitions, the operator’s revenue increases with network capacity up to a certain threshold but decreases beyond that. Further, partitioning does not help for very high values of capacity. We then show that optimal partitioning alleviates this issue. Finally, we analyze the benefit of partitioning the network into more than two identical subnetworks.