Improved privacy preserving p2 p multimedia
- 1. Improved Privacy-Preserving P2P Multimedia Distribution Based on Recombined Fingerprints ABSTRACT: Anonymous fingerprint has been suggested as a convenient solution for the legal distribution of multimedia contents with copyright protection whilst preserving the privacy of buyers, whose identities are only revealed in case of illegal re- distribution. However, most of the existing anonymous fingerprinting protocols are impractical for two main reasons: 1) the use of complex time-consuming protocols and/or homomorphic encryption of the content, and 2) a unicast approach for distribution that does not scale for a large number of buyers. This paper stems from a previous proposal of recombined fingerprints which overcomes some of these drawbacks. However, the recombined fingerprint approach requires a complex graph search for traitor tracing, which needs the participation of other buyers, and honest proxies in its P2P distribution scenario. This paper focuses on removing these disadvantages resulting in an efficient, scalable, privacy-preserving and P2P- based fingerprinting system. EXISTING SYSTEM: Most fingerprinting systems can be classified in three categories, namely symmetric, asymmetric and anonymous schemes.
- 2. In symmetric schemes, the merchant is the one who embeds the fingerprint into the content and forwards the result to the buyer; hence, the buyer cannot be formally accused of illegal re-distribution, since the merchant also had access to the fingerprinted content and could be responsible for the re- distribution. In asymmetric fingerprinting, the merchant does not have access to the fingerprinted copy, but he can recover the fingerprint in case of illegal re- distribution and thereby identify the offending buyer. In anonymous fingerprinting, in addition to asymmetry, the buyer preserves her anonymity (privacy) and hence she cannot be linked to the purchase of a specific content, unless she participates in an illegal re-distribution. DISADVANTAGES OF EXISTING SYSTEM: Developing a practical system using this idea appears difficult, because public-key encryption expands data and substantially increases the communication bandwidth required for transfers. Homomorphic encryption constrains the type of mathematical operations which can be performed on the content for embedding, making it difficult to use the more advanced and robusttechniques in the data hiding literature.
- 3. In addition, the application of this idea in a distributed scenario (such as P2P networks) is not simple, since embedding would have to be performed by peer buyers, requiring a complex and supervised protocol. PROPOSED SYSTEM: The content is divided into several ordered fragments and each of them is embedded separately with a random binary sequence. The binary sequence for each fragment is called segment and the concatenation of all segments forms the whole fingerprint. The merchant distributes different copies to a reduced set of M seed buyers. The fingerprints of these buyers are such that their segments have low pair- wise correlations. The buyers other than the seed ones engage on P2P transfers of the content in such a way that each new buyer obtains fragments from at least two other buyers. The total number of buyers is N _ M. The communication between peer buyers is anonymous through an onion routing-like protocolusing a proxy. The fingerprint of each new buyer is built as a recombination of the segments of its parents.
- 4. Proxies know the pseudonyms of source and destination buyers and they have access to the symmetric keys used for encrypting the multimedia content. A transaction record is created by a transaction monitor to keep track of each transfer between peer buyers. These records do not contain the embedded fingerprints, but only an encrypted hash of them. The fingerprints’ hashes are encrypted in such a way that the private key of at least one parent is required for obtaining their cleartext. The real identities of buyers are known only by the merchant. The transaction monitor records buyers’ pseudonyms. In case of illegal re-distribution, a search is required through the distribution graph. The search starts from the seed buyers and is directed by a correlation function between the traced fingerprint and the fingerprints of the tested buyers. These tested buyers must co-operate with a tracing authority to compute the correlation between their fingerprint and the one extracted from the illegally re-distributed file. The fingerprints’ hashes recorded in the transaction monitor are enough to prevent buyers from cheating in this step. At each step of the traitor tracing protocol, the buyer with maximum correlation is chosen as the most likely ancestor of the illegal re-distributor. This criterion is mostly right, but some incorrect choices may occur during the search process, requiring the exhaustion of a subgraph and backtracking.
- 5. The search ends when perfect correlation is found between the fingerprint of the tested buyer and that of the illegally re-distributed file. If a buyer refuses to take a correlation test, the hash recorded in the transaction monitor can be used as evidence against her. ADVANTAGES OF PROPOSED SYSTEM: This paper reviews the main features of the proposal suggested, highlights its main drawbacks, and suggests several significant improvements to achieve a more efficient and practical system, especially as traitor tracing is concerned, since it avoids the situations in which illegal redistributors cannot be traced with the proposal. Furthermore, better security properties against potentially malicious proxies are obtained. Although the system proposed in this paper uses publickey encryption in the distribution and traitor tracing protocols, it must be taken into account that this encryption is only applied to short bit strings, such as the binary fingerprints and hashes, not to the content. The fragments of the content are encrypted using symmetric cryptography, which is much more efficient.
- 6. SYSTEM ARCHITECTURE: SYSTEM REQUIREMENTS: HARDWARE REQUIREMENTS: System : Pentium IV 2.4 GHz. Hard Disk : 40 GB. Floppy Drive : 1.44 Mb. Monitor : 15 VGA Colour.
- 7. Mouse : Logitech. Ram : 512 Mb. SOFTWARE REQUIREMENTS: Operating system : Windows XP/7. Coding Language : ASP.net, C#.net Tool : Visual Studio 2010 Database : SQL SERVER 2008 REFERENCE: David Megıas, Member, IEEE, “Improved Privacy-Preserving P2P Multimedia Distribution Based on Recombined Fingerprints”, IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, VOL. 12, NO. 2, MARCH/APRIL, 2015.