Multicast in Wireless Networks

Abstract

Multicast is an efficient communication means of transmitting the same content to multiple receivers while minimizing network resource usage. However, wireless networks are very diverse and they have their own characteristics. Multicast has not been studied extensively for the networks different from traditional wired networks.

Our thesis is to prove that multicast is an effective means in wireless networks. By efficiently using network resources through schedulers, multicast can be an effective communication means. Therefore, we study multicast scheduler adaptation to the wireless networks where existing multicast schemes are not applied directly. Among the wireless networks, we consider multicast in cellular data networks and disruption tolerant networks.

First, we propose two proportionally fair multicast scheduling algorithms at the air interface in the downlink direction to adapt dynamic channel states in cellular data networks: Multicast Proportional Fairness (MPF) and Inter-Group Proportional Fairness (IPF) algorithms. Our algorithms take into account reported data rate requests from users and the average throughput of each user inside a cell and use this information to select an appropriate data rate for each group. We prove that MPF and IPF algorithms are proportionally fair among all users and among groups inside a cell respectively. Through simulations, we demonstrate that these algorithms achieve good balance between throughput and fairness among users and groups.

Second, we study joint optimization of link scheduling, routing and replication for disruption-tolerant networks (DTNs). We define a new notion of optimality for DTNs, called "snapshot optimality" which uses only contemporarily available knowledge. We then present a new efficient approximation algorithm, called Distributed Max-Contribution (DMC) based only on locally and contemporarily available information. Through a simulation study based on real GPS traces, we show that DMC demonstrates near-optimal performance.

By proposing an efficient multicast schedulers to cellular data networks and DTNs, we prove that multicast is an effective means of communication for wireless networks.