Search

BitTorrent is one of the Internet΄s most efficient content distribution protocols. It is known to perform very well over the wired Internet where end-to-end performance is almost guaranteed. However, in wireless ad hoc networks, many constraints appear as the scarcity of resources and their shared nature, which make running BitTorrent with its default configuration not lead to best performances. To these constraints it adds the fact that peers are both routers and end-users and that TCP-performance drops seriously with the number of hops. We show in this work that the neighbor selection mechanism in BitTorrent plays an important role in determining the performance of the protocol when deployed over a wireless ad hoc network. It is no longer efficient to choose and treat with peers independently of their location. A first solution is to limit the scope of the neighborhood. In this case, TCP connections are fast but there is no more diversity of pieces in the network: pieces propagate in a unique direction from the seed to distant peers. This prohibits peers from reciprocating data and leads to low sharing ratios and suboptimal utilization of network resources. To recover from these impairments, we propose an enhancement to BitTorrent which aims to minimize the time to download the content and at the same time to enforce cooperation among peers. Our solution considers a restricted neighborhood to reduce routing overhead and to improve throughput, while establishing few connections to remote peers to improve diversity of pieces. With the help of extensive NS-2 simulations, we show that these enhancements to BitTorrent significantly improve the file completion time while fully profiting from the incentives implemented in BitTorrent to enforce fair sharing. [ABSTRACT FROM AUTHOR]

AOMP (Ad-hoc Overlay Multicast Protocol) is a novel approach for application-layer multicast in ad-hoc networks. We introduce in this paper a new algorithm that exploits a few properties of IP-routing to extract underlying topology information. The basic idea is to match nodes΄ path to the source in order to detect near neighbors in the physical topology. Then, in a dynamic and decentralized way, we construct a minimum cost mobility-aware delivery tree, connecting nodes that are close to each other. We design a tree improvement algorithm in order to enhance the global performance of AOMP during data distribution. Our simulations results show that, compared to previously proposed application-layer multicast structures, AOMP yields trees with lower cost and traffic redundancy. In addition, it performs well in terms of packet losses, especially in case of node mobility. [ABSTRACT FROM AUTHOR]

Multicast is an efficient paradigm for transmitting data from a sender to a group of receivers. The IEEE 802.11 wireless LANs standards specify how to send multicast frames with no ACK and using one of the Basic Service Set (BSS) rates. This situation has led many researchers to design techniques aiming to improve reliability of a multicasting mechanism. The Leader-Base Protocol (LBP) is one such mechanism proposed in the literature that is the most promising approach. The main idea behind the design of the LBP mechanism is to reduce the probability of collision of the feedback messages sent by the multicast group members. However, the LBP mechanism falls short by not considering the varying conditions characterizing the wireless channels. In this paper, we introduce a novel auto rate selection multicast mechanism for multi-rate wireless LANs, namely ARSM (Auto Rate Selection for Multicast), capable of adapting the data transmission to the varying conditions of the channel. Our simulation results show that our new scheme outperforms the IEEE 802.11 and the LBP mechanisms. [ABSTRACT FROM AUTHOR]

We propose the so-called backup tree algorithm to compute a set of n-1 backup multicast delivery trees from the default multicast tree for application level multicast. For each backup multicast tree exactly one link of the default multicast tree is replaced by a backup link from the set of available links. The backup tree algorithm calculates the n-1 trees with a complexity of O (m log n). [ABSTRACT FROM AUTHOR]

In this paper, we present efficient ways of encoding multicast trees. Multicast tree encodings provide a convenient way of performing stateless and explicit multicast routing in networks and overlays. We show the correspondence of multicast trees to theoretical tree data structures and give lower bounds on the number of bits needed to represent multicast trees. Our encodings can be used to represent multicast trees using both node identifiers and link indexes and are based on balanced parentheses representation of tree data structures. These encodings are almost space optimal and can be read and processed efficiently. We evaluate the length of these encodings on multicast trees in generated and real topologies. [ABSTRACT FROM AUTHOR]