Alok Nandan

SPAWN: a swarming protocol for vehicular ad-hoc wireless networks

Future vehicular networks are expected to deploy short-range communication technology for inter-vehicle communication. In addition to vehicle-to-vehicle communication, users will be interested in accessing the multimedia-rich Internet from within the vehicular network. This motivates a compelling application of Co-operative Networking in the Vehicular Ad-Hoc network where the Ad Hoc network extends and complements the Internet. The broadcast nature of the wireless medium drives us to explore different design paradigms from the ones used in typical wired settings.A new paradigm in content delivery on the Internet using peer-peer swarming protocols is emerging [1,2]. We propose SPAWN, a simple cooperative strategy for content delivery in future vehicular networks. We study the issues involved in using such a strategy from the standpoint of Vehicular Ad-Hoc networks. Several enhancements to a popular swarming protocol (BitTorrent) are discussed including a gossip mechanism that leverages the inherent broadcast nature of the wireless medium, and a piece-selection strategy that uses proximity to exchange pieces quicker. Preliminary results show that SPAWN increases the perceived performance of the network, resulting in faster downloads for popular files.

Enhancing TCP performance in networks with small buffers

TCP performance can be significantly affected when the buffer capacity at routers is small. This is possible when either many flows share the network or the bandwidth-delay product is large (e.g. satellite links). The behavior of various versions of TCP with respect to buffer capacity issues has not been studied in much detail. We investigate the behavior and performance of different TCP variants under small buffer capacity conditions. We recognize TCP pacing as a potential solution. However, instead of using TCPs sending rate as the dictating metric, we make use of the bandwidth-share estimate (BSE), maintained by TCP Westwood, to set the pacing interval. We call this newly proposed protocol paced-Westwood. We also show the need to scale BSE further to mitigate the effects of positive feedback in BSE. For this, we propose a further enhancement that we call /spl alpha/-paced Westwood that uses a scaling parameter /spl alpha/ to enforce convergence of BSE and the pacing interval. The proposed /spl alpha/-paced Westwood uses its BSE to space the packet bursts during the slow-start phase, resulting in a superior throughput in the troublesome low buffer capacity cases. With the help of simulations, we show that our enhanced TCP Westwood outperforms both unpaced as well as paced TCP NewReno under low buffer capacity networks.

CapProbe: a simple and accurate capacity estimation technique for wired and wireless environments

The problem of estimating the capacity of an Internet path is one of fundamental importance. Due to the multitude of potential applications, a large number of solutions have been proposed and evaluated. The proposed solutions so far have been successful in partially addressing the problem, but have suffered from being slow, obtrusive or inaccurate. In this work, we evaluate CapProbe, a low-cost and accurate end-to-end capacity estimation scheme that relies on packet dispersion techniques as well as end-to-end delays. The key observation that enabled the development of CapProbe is that both compression and expansion of packet pair dispersion are the result of queuing due to cross-traffic. By filtering out queuing effects from packet pair samples, CapProbe is able to estimate capacity accurately in most environments, with minimal processing and probing traffic overhead. In fact, the storage and processing requirements of CapProbe are orders of magnitude smaller than most of the previously proposed schemes. We tested CapProbe through simulation, Internet, Internet2 and wireless experiments. We found that CapProbe error percentage in capacity estimation was within 10% in almost all cases, and within 5% in most cases.

The use of a proxy on board the satellite to improve TCP performance

High errors and high delays create well-known problems with TCP. One established approach to solve this problem involves dividing TCP connections into segments, or splitting the connection. This paper takes this approach one step further by exploring the use of a TCP proxy on board a satellite for the purpose of enhancing end-to-end TCP performance. We show that this approach yields a number of important advantages, especially for small, mobile terminals, multi-segment systems, and multicast applications. As part of this architecture, we introduce a method of TCP backpressure using the advertised window, which optimizes memory usage on the satellite. Using simulation, we demonstrate that performance may improve by as much as three-fold.

GhostShare - reliable and anonymous P2P video distribution

P2P networks have emerged as a powerful multimedia content distribution mechanism. However, the widespread deployment of P2P networks are hindered by several issues, especially ones that influence end-user satisfaction, including privacy protection. In this paper, we propose GhostShare, a P2P network built on the pastry substrate, to distribute video content. The primary design goals of GhostShare are anonymity and load balancing for participating peers. We present simulation results that prove the effectiveness of GhostShares load balancing mechanism and provide an analysis of the anonymity scheme.

Optimizing neighbors by objective functions in peer-to-peer networks

Many distributed hash table topologies, such as Pastry, allow flexible choosing of a peers neighbors while maintaining routing consistency. Traditionally, such flexibility has been used to only optimize the overlay only for latency. In this paper, we create a set of objective functions that allow a peer to select neighbors for its routing table which minimize ping time, maximize bandwidth, or attempt to do both. In conjunction with a novel algorithm for quickly finding peers that maximize a given objective function without settling to a local maximum in the identifier space, we show through simulation that routing tables optimized in a greedy fashion by each node can have significant impact on end-to-end latency and capacity, such as reducing end-to-end delay by over 50 percent.

On index load balancing in scalable P2P media distribution

Peer-Peer (P2P) technologies have recently been in the limelight for their disruptive power in particular they have emerged as a powerful multimedia content distribution mechanism. However, the widespread deployment of P2P networks are hindered by several issues, especially the ones that influence end-user satisfaction, including reliability. In this paper, we propose a solution for an efficient and user-oriented keyword lookup service on P2P networks. The proposed mechanism has been designed to achieve reliability via index load balancing and address the scalability issues of extremely popular keywords in the index. The system performance have been analytically derived as well implemented using the OpenDHT framework on PlanetLab.