Reconnaissance of Golcuk 1999 Earthquake Damage Using Satellite Images

This study utilizes remotely sensed pre- and post-disaster images in order to detect any change specifically associated with structural and major regional damage caused by natural disasters such as a strong earthquake. Currently, postdisaster reconnaissance is conducted using mostly ground-truth teams. Although by using this method high-resolution assessment can be provided, it is very time consuming, costly and sometimes not practical. This study investigates an unsupervised near near-time method of reconnaissance using a pair of coregistered remotely sensed images of the same scene acquired at different times as input. The output is an image (mostly binary) in which “changed” pixels are separated from “not-changed” ones. The approach is fully deterministic using principal component analysis, normalization and further setting a threshold to classify pixels. For verification, a set of multispectral images from Golcuk-Turkey 1999 earthquake was used. It is concluded the program generates very similar results as obtained by ground-truth teams.     

Characterizing files in the modern Gnutella network

The Internet has witnessed an explosive increase in the popularity of Peer-to-Peer (P2P) file-sharing applications during the past few years. As these applications become more popular, it becomes increasingly important to characterize their behavior in order to improve their performance and quantify their impact on the network. In this paper, we present a measurement study on characteristics of available files in the modern Gnutella system. We develop two new methodologies to capture accurate snapshots of available files in a large-scale P2P system. These methodologies were implemented in a parallel crawler that captures the entire overlay topology of the system where each peer in the overlay is annotated with its available files. We have captured more than 50 snapshots of the Gnutella system that span over 1 year period. Using these snapshots, we conduct three types of analysis on available files: (1) Static analysis, (2) Topological analysis, and (3) dynamic analysis. Our results reveal several interesting properties of available files in Gnutella that can be leveraged to improve the design and evaluation of P2P file-sharing applications. This paper extends and supplants the earlier version of this paper presented at MMCN 2006 [1]. This material is based upon work supported by the National Science Foundation (NSF) under Grant No. Nets-NBD-0627202, CAREER Award CNS-0448639, and an unrestricted gift from Cisco Systems. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF or Cisco.     

Layered quality adaptation for Internet video streaming

Streaming audio and video applications are becoming increasingly popular on the Internet, and the lack of effective congestion control in such applications is now a cause for significant concern. The problem is one of adapting the compression without requiring video servers to reencode the data, and fitting the resulting stream into the rapidly varying available bandwidth. At the same time, rapid fluctuations in quality will be disturbing to the users and should be avoided. We present a mechanism for using layered video in the context of unicast congestion control. This quality adaptation mechanism adds and drops layers of the video stream to perform long-term coarse-grain adaptation, while using a TCP-friendly congestion control mechanism to react to congestion on very short timescales. The mismatches between the two timescales are absorbed using buffering at the receiver. We present an efficient scheme for the distribution of available bandwidth among the active layers. Our scheme allows the server to trade short-term improvement for long-term smoothing of quality. We discuss the issues involved in implementing and tuning such a mechanism, and present our simulation results     

On performance evaluation of swarm-based live peer-to-peer streaming applications

During recent years, swarm-based peer-to-peer streaming (SPS) mechanisms have become increasingly popular for scalable delivery of live streams over the Internet. The performance of SPS mechanisms depends on the overall effect of several factors including the connectivity of the overlay, the details of packet scheduling scheme and environment settings (e.g., peer and source bandwidth). Prior studies often presented overall performance of their proposed techniques in terms of delivered quality to all peers at a particular setting without demonstrating their inherent performance bottlenecks. Therefore, it is difficult to determine whether and how the reported performance of an SPS mechanism might change as a function of available resources or the connectivity of the overlay. In this paper, we present a simple yet effective methodology for performance evaluation of SPS mechanisms. Our methodology leverages an organized view of an overlay coupled with a two-phase notion of content delivery in SPS mechanisms to derive a set of metrics that collectively capture the behavior of each phase of content delivery. Therefore, the collection of our metrics can be viewed as the “signature of content delivery” of a given SPS mechanism. We also present the signature of a well-performing SPS mechanism that can be used as a reference for assessment of other mechanisms. To demonstrate the ability of our proposed evaluation methodology in identifying performance bottlenecks of SPS mechanisms and their underlying causes, we conduct two case studies: (1) assessing the performance of a set of candidate packet scheduling schemes; and (2) examining the effect of overlay localization on the performance of SPS mechanisms. In addition to illustrating the use of our methodology through examples, our case studies shed an insightful light on the performance bottlenecks in our target scenarios.     

Incorporating contribution-awareness into mesh-based Peer-to-Peer streaming systems

While Peer-to-Peer streaming has become increasingly popular over the Internet during recent years, the proper allocation of available resources among peers in a resource constraint environment, remains a challenging problem. In a resource constraint environment, the allocated resources and thus delivered quality to individual peers should be proportional to their contribution to the system, i.e., resource allocation should be contribution aware. This in turn results in fairness among peers and encourages active contribution from participating peers which is essential for scalability of P2P systems. However, contribution-aware resource allocation is challenging due to the distributed and dynamic nature of resources in P2P systems. In this paper, we present a tax-based contribution-aware scheme for live mesh-based P2P streaming approaches. In our proposed scheme, individual peers use a tax function to determine their number of parent peers (i.e., their share of resources) based on the number of their child peers (i.e., peers’ contributed resources) and the aggregate available resources in the system. We examine the behavior of a commonly used tax function, and describe how the contribution aware scheme can leverage the tax function. Through extensive simulations we demonstrate the ability of our proposed scheme to properly allocate available resources among participating peers over a wide range of scenarios. We show that the amount of resources (i.e., bandwidth) is divided across peers proportional to their contribution and in our default simulation setting the median delivered quality to high bandwidth peers with high contribution is improved by 100%. We believe that our results shed an insightful light on the dynamics of resource utilization and allocation in the context of live mesh-based P2P streaming.     

Tsunami: A parasitic,indestructible botnet on Kad

While current botnets rely on a central server or bootstrap nodes for their operations, in this paper we identify and investigate a new type of botnet, called Tsunami, in which no such bottleneck nodes exist. In particular, we study how a Tsunami botnet can build a parasitic relationship with a widely deployed P2P system, Kad, to successfully issue commands to its bots, launch various attacks, including distributed denial of service (DDoS) and spam, at ease, as well as receive responses from the bots. Our evaluation shows that in a Kad network with four million nodes, even with only 6 % nodes being Tsunami bots, Tsunami can reach 75 % of its bots in less than 4 min and receive responses from 99 % of bots. We further propose how we may defend against Tsunami and evaluate the defense solution.     

On Unbiased Sampling for Unstructured Peer-to-Peer Networks

This paper presents a detailed examination of how the dynamic and heterogeneous nature of real-world peer-to-peer systems can introduce bias into the selection of representative samples of peer properties (e.g., degree, link bandwidth, number of files shared). We propose the metropolized random walk with backtracking (MRWB) as a viable and promising technique for collecting nearly unbiased samples and conduct an extensive simulation study to demonstrate that our technique works well for a wide variety of commonly-encountered peer-to-peer network conditions. We have implemented the MRWB algorithm for selecting peer addresses uniformly at random into a tool called ion-sampler. Using the Gnutella network, we empirically show that ion-sampler yields more accurate samples than tools that rely on commonly-used sampling techniques and results in dramatic improvements in efficiency and scalability compared to performing a full crawl.     

Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems

In recent years, peer-to-peer (P2P) file-sharing systems have evolved to accommodate growing numbers of participating peers. In particular, new features have changed the properties of the unstructured overlay topologies formed by these peers. Little is known about the characteristics of these topologies and their dynamics in modern file-sharing applications, despite their importance. This paper presents a detailed characterization of P2P overlay topologies and their dynamics, focusing on the modern Gnutella network. We present Cruiser, a fast and accurate P2P crawler, which can capture a complete snapshot of the Gnutella network of more than one million peers in just a few minutes, and show how inaccuracy in snapshots can lead to erroneous conclusions-such as a power-law degree distribution. Leveraging recent overlay snapshots captured with Cruiser, we characterize the graph-related properties of individual overlay snapshots and overlay dynamics across slices of back-to-back snapshots. Our results reveal that while the Gnutella network has dramatically grown and changed in many ways, it still exhibits the clustering and short path lengths of a small world network. Furthermore, its overlay topology is highly resilient to random peer departure and even systematic attacks. More interestingly, overlay dynamics lead to an ldquoonion-likerdquo biased connectivity among peers where each peer is more likely connected to peers with higher uptime. Therefore, long-lived peers form a stable core that ensures reachability among peers despite overlay dynamics.     

Adaptive receiver-driven streaming from multiple senders

This paper presents the design and evaluation of an adaptive streaming mechanism from multiple senders to a single receiver in Peer-to-Peer (P2P) networks, called P2P Adaptive Layered Streaming, or PALS. PALS is a receiver-driven mechanism. It enables a receiver peer to orchestrate quality adaptive streaming of a single, layer-encoded video stream from multiple congestion-controlled senders, and is able to support a spectrum of noninteractive streaming applications. The primary challenge in the design of a streaming mechanism from multiple senders is that available bandwidth from individual peers is not known a priori, and could significantly change during delivery. In PALS, the receiver periodically performs quality adaptation based on the aggregate bandwidth from all senders to determine: (i) the overall quality (i.e number of layers) that can be collectively delivered by all senders, and more importantly (ii) the specific subset of packets that should be delivered by individual senders in order to gracefully cope with any sudden change in their bandwidth. Our detailed simulation-based evaluations illustrate that PALS can effectively cope with several angles of dynamics in the system including: bandwidth variations, peer participation, and partially available content at different peers. We also demonstrate the importance of coordination among senders and examine key design tradeoffs for the PALS mechanism. Nazanin Magharei is currently a PhD student in the Computer Science Department at the University of Oregon. She received her BSc degree in Electrical Engineering from Sharif University of Technology, Iran in 2002. Her research interests include Peer-to-Peer streaming and multimedia caching. Reza Rejaie is currently an Assistant Professor at the Department of Computer and Information Science at the University of Oregon. From October 1999 to March 2002, he was a Senior Technical Staff member at AT&T Labs-Research in Menlo Park, California. He received a NSF CAREER Award for his work on P2P streaming in 2005. Reza has served on the editorial board of IEEE Communications Surveys & Tutorials, as well as the program committee of major networking conferences including INFOCOM, ICNP, Global Internet, ACM Multimedia, IEEE Multimedia, NOSSDAV, ICDCS, and MMCN. Reza received his MS and PhD degrees in Computer Science from the University of Southern California (USC) in 1996 and 1999, and his BS degree in Electrical Engineering from the Sharif University of Technology (Tehran, Iran) in 1991, respectively. Reza has been a member of both the ACM and IEEE since 1997.