network behavior is described as an overlay network because the peer protocols While peer-to-peer (P2P) applications have had a rapid ascent and wide impact, caite.info Overlays 9 8 pdf. Abstract. This chapter reviews core concepts of peer-to-peer (P2P) networking. It A model differentiating P2P infrastructures, P2P applications, and P2P communities caite.info Clarke, I. (). Freenet's Next Generation Routing Protocol. Request PDF on ResearchGate | P2P Networking and Applications | Peer-to- Peer (P2P) networks enable users to directly share digital content (such as audio, .
|Language:||English, Spanish, Dutch|
|Genre:||Science & Research|
|ePub File Size:||15.60 MB|
|PDF File Size:||18.58 MB|
|Distribution:||Free* [*Regsitration Required]|
Peer-to-Peer (P2P) networks enable users to directly share digital content (such as audio, video, and text files) as well as real-time data (such as telephony. The Morgan Kaufmann Series in Networking. Series Editor, David Clark, M.I.T.. P2P Networking and Applications. John F. Buford, Heather Yu, and Eng Lua. Purchase P2P Networking and Applications - 1st Edition. Print Book & E-Book. DRM-free (EPub, PDF, Mobi). × DRM-Free Easy - Download and start reading.
In this chapter, we have presented a search protocol for unstructured P2P networks that solves several problems e. The query forwarded by the peer 5 is also blocked by the peer 10 and the peer 11 as both of them know that the peer 5 is malicious. This leads to a major conflict between the requirements of anonymity of the users to protect their privacy and an increasing need to provide robust access control, data integrity, confidentiality and accountability services. Free Shipping Free global shipping No minimum order. The trust value of a peer lies in the interval [0, 1].
Last but not the least I express my thanks to my friends for their cooperation and support. Neighbor selection by peer P for forwarding the query string c2, f4. The community edges and the connectivity edges are drawn using solid and dotted lines respectively.
The peers that receive the query for forwarding are shaded. The breadth first search BFS tree for the search initiated by peer 1. Topology adaptation based on outcome of the search in Fig. Malicious nodes are shaded in gray color. Identity protection of the requesting peer i from the supplier peer k by use of trusted peer j. Protecting data handle using trusted node. Peer i and k are the requester and the supplier peer respectively.
Peer j is the trusted peer of the requester peer i AR for various percentages of malicious peers in the network. EAR of honest peers for various percentages of malicious peers in the network. EAR for various percentages of malicious peers in the network with and without the trust management module.
QMR for various percentages of malicious peers in the network. Closeness centrality for various percentages of malicious peers in the network. An illustrative content distribution among peers Simulation parameters The goal of a P2P system is to aggregate resources available at the edge of Internet and to share it co-operatively among the users. The file sharing P2P systems have particularly become popular as a new paradigm for information exchange among large number of users in the Internet.
These systems are more robust, scalable, fault-tolerant and they offer better availability of resources than the traditional systems based on the client- server model. Depending on the presence of a central server, the P2P systems can be classified as centralized or decentralized . In the decentralized architecture, both the resource discovery and the resource download happen in a distributed manner. The decentralized P2P architectures may further be classified as structured or unstructured networks.
In unstructured P2P networks, however, the placement of the contents is unrelated to the topologies of the networks. The unstructured P2P networks perform better than their structured counterparts in dynamic environments. However, they need efficient search mechanisms and they also suffer from numerous problems such as: The malicious peers often use these networks to carry out content poisoning and to distribute harmful programs such as Trojan Horses and viruses .
Distributed reputation based trust management systems have been proposed by the researchers to provide protection against the malicious content distribution in a distributed environment . HISTORY During the past few years, in the area of wireless communications and networking, a novel paradigm named the IoT which was first introduced by Kevin Ashton in the year , has gained increasingly more attention in the academia and industry . By embedding short-range mobile transceivers into a wide array of additional gadgets and everyday items, enabling new forms of communication between people and things, and between things themselves, IoT would add a new dimension to the world of information and communication.
Unquestionably, the main strength of the IoT vision is the high impact it will have on several aspects of every-day life and behavior of potential users. From the point of view of a private user, the most obvious effects of the IoT will be visible in both working and domestic fields.
The real power of the IoT lies in the universal connectivity among all devices and objects. It should be possible for the service requestors to understand what the service providers have to offer by semantic modeling. This is a key issue for stepping towards ubiquitous services, where the new or modified services may appear at any time, and towards device networks that are capable of dynamically adapting to the context changes as may be imposed by the application.
This calls for a middleware which will interface between the devices and the applications. Since the devices need to communicate with each other, there is a need for a naming and addressing scheme, and a mechanism for search and discovery. Moreover, since each device is mapped to an identity through naming and addressing , there are serious security and privacy concerns.
In a massively distributed system like the IoT, several agent platforms will exist each having a set of agents running and registered with a directory facilitator DF. Problem, however, arises when the agents from different platforms will have to search the remote DFs and interact with the agents located on the remote platforms.
In these scenarios, the agents will have to use resource discovery protocols which are similar to the file searching protocols in a purely unstructured P2P network. Hence, efficient searching in unstructured peer-to-peer network has a direct contextual relevance to the resource discovery in IoT applications.
We broadly divide these protocols into three categories: The primary objective of the schemes under the general searching category is to enhance the search efficiency - i. The secure searching schemes attempt to incorporate security into the searching mechanisms by defending against various possible attacks on the peers and the overall network. The privacy-preserving searching mechanisms protect peer i.
In the following subsections, we briefly discuss some of the currently existing schemes under each of these three categories of search. Adamic et al. However, such strategies lack scalability and do not perform well in a network having large number of peers.
Condie et al. The peers connect to those peers from whom they are most likely to download the authentic files . The peers add or remove their neighbors based on local trust and connection trust which are decided based on the transactions history.. This punishment strategy is relaxed in the reciprocal capacity-based adaptive topology protocol RC-ATP , wherein a peer connects to others which have higher reciprocal capacities .
While the RC-ATP scheme provides better network connectivity than the APT scheme and it also reduces the cost due to the inauthentic downloads, it has a large overhead due to the topology adaptation. Kamvar et al. In , which is known as Gnutella v0. The peers are categorized into two types: The leaf-peers have connections with their respective ultra-peers, while the ultra-peers have connections with their own leaf-peers as well as with the other ultra-peers.
The leaf-peers can initiate lookup requests, receive lookup responses and respond to requests for which they have exact answers. An ultra-peer forwards the lookup requests to other the ultra-peers or the leaf-peers to which the ultra-peer is connected, if it exactly knows which leaf-peer has answers to the requests. At the ultra-peer level of the hierarchy, a flooding mechanism is used for forwarding the lookup requests.
Hsiao et al. In the proposed algorithm, each peer creates and maintains a constant number of overlay connections with other peers in a distributed manner. Tang et al.
In , a fully distributed protocol named distributed cycle minimization protocol DCMP has been presented that minimizes duplicate messages by eliminating any possible redundant cycles of the messages.
Lin el al. Huang-Fu et al. Balfe et al. The authors have argued that the central problem in securing P2P network lie in the fact that these networks do not have any stable verifiable peer identity verification mechanism. This leads to a major conflict between the requirements of anonymity of the users to protect their privacy and an increasing need to provide robust access control, data integrity, confidentiality and accountability services. The authors have shown how the trusted computing group TCG protocols for direct anonymous attestation DAA can be used to enforce the use of stable, platform- dependent pseudonyms so that spoofing attacks can be prevented.
The proposed scheme also uses the DAA protocol to build entity authentication using pseudonyms for establishing secure communication channels between any given pair of peers. A large number of studies have been carried out on the reputation and trust management in both the unstructured and the structured P2P networks.
The EigenTrust scheme proposed by Kamvar et al. The scheme utilizes a novel normalization process in which the trust ratings of a peer are averaged and normalized. However, the normalization may lead to partial loss of important information on the original distribution and variance of trust function.
One easy way to preserve the privacy of the users in network communication is to deploy some fixed servers or proxies for this purpose. For example, in the Publius system , the identity of a publisher is protected by encrypting the data communicated in the network, and managing the key distribution among k servers by using the mechanism of threshold cryptography .
Some anonymity schemes based on the use of a trusted third party server have been presented in . Lu et al. The proposition is based on selection of a trusted peer as the proxy during the data acquirement. The requester peer sends the request and receives the data through the proxy without revealing its identity. Since the real identity of the requester is never revealed during the communication, the privacy of the requester node is protected.
However, in an structured P2P network, the selection and maintenance of the trusted peers for each peer is difficult due to the dynamic nature of the network topology and the autonomy of the peers. Hence, the scheme of the searching protocols for P2P networks existing in the literature. First, in the proposed protocol, the links in the original overlays are never deleted in order to avoid network partitioning.
This claim is validated by the simulation results presented in Section 5. The protocol presented in this chapter takes the advantage of semantic communities formation to improve the quality of service QoS of search by reducing the search time and increasing the rate of authentic file downloads.
On the other hand, the central module of the proposed protocol is a robust trust management framework, which is responsible for securing the searching process and protecting the privacy of the peers and their data. Finally, unlike the APT and the RC-ATP protocols, the proposed protocol punishes the malicious peers by blocking all the queries which originate from these peers.
Network topology: The topology of a P2P network plays an important role in the formation of trust among its peers and for efficient operation of a search protocol in the network. Following the work in  and , in the current proposition, the P2P network has been modeled as a power law graph. In the network environment, a certain percentage of the peers are randomly chosen to act as malicious peersThe network links are categorized into two types: The connectivity links are the edges of the original power law network which provide seamless connectivity among the peers.
On the other hand, the community links are added probabilistically between the peers who know each other, and have already interacted with each other before. A community link may be deleted when the perceived trustworthiness of a peer falls in the perception of its neighbors.
The formal procedure of computing trust of a peer is discussed later in this section. However, informally, it may be said that the value of the trust metric of a peer i as computed by another peer j increases when the peer j has some positive experience while interacting with the peer i i.
A negative experience i. A limit is put on the additional number of edges that a peer can acquire to control the bandwidth usage and the query processing overhead in the network. This increase in network load is measured relative to the initial network degree corresponding to the connectivity edges.
Let final degree x and initial degree x be the initial and the final degree of a node x. Content distribution: The dynamics of a P2P network are highly dependent on the volume and the variety of the files that each peer chooses to share. Hence a model reflecting the real-world P2P networks is required. It has been observed that the peers are, in general, interested in a subset of the contents in the P2P network .
Also, the peers are often interested only in the files from a few content categories. Some categories of files are more popular than the others. It has been shown that the Gnutella content distribution follows the zipf distribution .
In the proposed scheme, the files are assigned to the peers at the network initialization phase as follows: Finally, the peer i is assigned files F according to its content categories and interest levels in those categories.
Each distinct file f c,r is uniquely identified by the content category c to which it belongs and its popularity ranking r within that category .
Table 4. In the proposed scheme, we assume that there are 32 content categories. It must be noted that the number of content categories can be any positive integer n.
However, for the evaluation of the performance of the proposed protocol, we have used 32 content categories. Each content category is characterized by its popularity rank.
As already discussed earlier, the peers are assumed to be interested in a subset of the total available contents in the network. Accordingly, each peer initially chooses a number of content categories and shares files only in those categories.
In the proposed protocol, each peer randomly chooses between three to six content categories. The files belonging to more popular categories are shared more in numbers. Table 1 shows an illustrative content distribution among 5 peers in a network. The category C1 is more replicated as it is the most popular category.
Peer 1 P1 shares files of three categories: C1, C2, C3. As explained earlier, P1 shares maximum number of files in category C1, followed by category C2 and so on.
On the other hand, Peer 3 P3 shares maximum number of files in category C2 as it is the most popular among the categories of files chosen by it. Query initiation model: The authors in  have shown that the peers usually query for the files which are available in the network, and which belong to the content categories of their interests. However, the number of queries a peer issues may vary from peer to peer. The query initiation is modelled as Poisson arrival process. Trust management engine: A trust management engine is designed which helps a peer to compute the trust ratings of the other peers based on the past transactions, as well as on the recommendations of its neighbor.
For computing the trust values of the peers, a method similar to the one proposed in  is followed. The framework employs a beta distribution for reputation representation, updates and integration. The first-hand information and the second-hand information recommendation from neighbors are combined to compute the reputation value of a peer.
The weight assigned by a peer i to a second-hand information received from a peer k is a function of the reputation of the peer k as maintained in the peer i. For each peer j, a reputation Rij is computed by a neighbor peer i. The reputation is embodied in the Beta model which has two parameters: The reputation of the peer j as maintained by the peer i is computed using 2.
After receiving this new information, the peer i combines it with its current assessment Rij to obtain a new reputation Rijnew as shown in 4. To prevent against bad-mouthing and ballot-stuffing attacks , the peers assign higher weights to the first-hand observations i. To incorporate these issues while updating the reputation values using the second-hand information, the Dempster-Shafer theory  and the belief discounting model  are employed. The use of these models leads to the derivation of the expressions in 5 and 6.
To make the trust management system robust against sleeper attack  , where a peer behaves honestly for a sufficiently long time to acquire a good reputation and then starts misbehaving and exploiting the system, the proposed system assigns more weights to the recent observations for computing the aggregate reputation metrics of a peer.
As mentioned earlier in this section, the trust value of a peer is computed as the statistical expected value of its reputation. The trust value of a peer lies in the interval [0, 1]. In the implementation of the proposed protocol, we have used an LRU least recently used data structure which is maintained in each peer to keep track of the most recent transactions the peer had with maximum of 32 peers.
However, the choice of the number of peers whose transaction history is maintained in each peer is a tuneable parameter, which can be increased or decreased based on the memory and the computing capabilities of the peers. Identity of the peers: The public key serves as the identity of the peer.
The identities are persistent and they enable two peers that have exchanged keys to locate and connect to one another whenever the peers are online. The DHT entries for the peer i are signed by the peer i and encrypted using its public key. Each entry is indexed by a 20 byte randomly generated shared secret, which is agreed upon during the first successful connection between the two peers.
Node churning model: In P2P networks, a large number of peers may join and leave at any time. This activity is termed as node churning. To simulate node churning, prior to each generation a set of consecutive searches , a fixed percentage of nodes are chosen randomly as inactive peers.
These peers neither initiate nor respond to a query in that generation, and they join the system later with their LRU data structure cleared. The clearing of the LRU data structure ensures that these peers do not have any historical information about their past transactions with other peers in the network. Since in a real world network, even in presence of churning, the approximate distribution of content categories and files remain constant, the contents of the peers undergoing churn are exchanged with the peer remaining in the network, so that the content distribution model of the network remains unchanged.
Threat model: The malicious peers adopt various strategies threat models to conceal their behavior so that they can effectively disrupt the activities in the network, and yet go undetected. The proposed protocol considers two threat models.
The peers which share good quality files enjoy better topological positions after topology adaptation. In the threat model A, the malicious peers attempt to circumvent this effect by providing good files occasionally with a probability - known as degree of deception- to lure other peers to form communities with them.
In the threat model B, a group of malicious peer joins the system and provides good files until the connectivity of the peers reaches a maximum value - the edge limit. The peers then start acting maliciously by spreading fake contents in the network. Search for books, journals or webpages All Webpages Books Journals.
View on ScienceDirect. Hardcover ISBN: Morgan Kaufmann. Published Date: Page Count: Sorry, this product is currently out of stock. Flexible - Read on multiple operating systems and devices. Easily read eBooks on smart phones, computers, or any eBook readers, including Kindle. When you read an eBook on VitalSource Bookshelf, enjoy such features as: Access online or offline, on mobile or desktop devices Bookmarks, highlights and notes sync across all your devices Smart study tools such as note sharing and subscription, review mode, and Microsoft OneNote integration Search and navigate content across your entire Bookshelf library Interactive notebook and read-aloud functionality Look up additional information online by highlighting a word or phrase.
Institutional Subscription. Instructor Ancillary Support Materials. Free Shipping Free global shipping No minimum order. Uses well-known commercial P2P systems as models, thus demonstrating real-world applicability. Discusses how current research trends in wireless networking, high-def content, DRM, etc. Provides online access to the Overlay Weaver P2P emulator, an open-source tool that supports a number of peer-to-peer applications with which readers can practice.
English Copyright: Huawei Technologies. Powered by. You are connected as. Connect with: Use your name: Thank you for posting a review!
We value your input. Share your review so everyone else can enjoy it too. Your review was sent successfully and is now waiting for our team to publish it.