Research On P2P Collusive Behavior For Copyright Analysis

As a typical representative of network resource sharing, P2P (Peer-to-Peer)technology has already become the main way. Users can share music, movies, gamesand other files via P2P technology, in which, any peer can become a resource providerto achieve easier, resource dissemination and sharing. Also, the rampant piracyseriously in P2P networks seriously damages the interests of the company of digitalmedia. To be note that, the collusive piracy has become a major threat of P2Pcopyright, which phenomenon has drawn wide concern in both academy and industry.The study of P2P copyright issues can be classified into the following two ways:The first way is to use protocol-level signature-like methods to automatically identifyand prevention; the second way belongs to an application-level regulation, whichneeds to find the relationship between illegal content and IP and then makenon-real-time user warning or punishment. It is worth noting that the first way has toembed a specific personal license agreement (such as PAP) into the standard P2Pprotocol, The second approach aims not to affect and change the opening and sharingof standard P2P protocols. So, from the industrial realization, the second approach ismore practical and easier to accept. This paper follows the second way, for piracyprevention, we give a deep study on CBF-based collusive behavior analysis. Theinnovations can be generalized as follows:1. We proposed a MCBFC model to storage P2P behavior’s followingrelationships, which is a cyclic structure containing time tab and N CBF. And theCounter of each of CBF will be connected to a linked list, so that the followingrelations can be hashed into each corresponding linked list.2. We proposed a following relation analyzing approach based on contentfeedback. In this paper, we proposed a model of PFCF and analysis the location ofdeploying of PFCF model in P2P. Based on MCBFC, we can effectively detect theresource and user, through mining user’s sequential behavior, the probability offollowing can be achieved, which can prevent P2P copyright.3. We do a lot of experiments on comparison of CBF and MCBFC, whichcompare in collision rate, space allocation, time complexity. Our MCBFC structurehas better storage efficiency than the CBF. Moreover, under certain cost, MCBFC canachieve effective support on the following analysis and MCBFC storage structure can store user behavior history data for analysis.