| Abstract [eng] |
Anonymity has become a very important factor in protecting business or private information. Over the past few years P2P applications have become a significant part of our virtual activities such as voice communication, messaging or data exchange. Most of the P2P network users are concerned that all their actions in the network and their identity can be easily revealed by other network users. For that reason a variety of tools have been proposed to ensure the anonymity of P2P network. These tools and techniques are designed with three main purposes: to protect the identity of the sender, to protect the identity of data retrieval initiator and to protect data content which is being sent across the network. After analyzing the principles of anonymous networks and systems, it has been found that in various publications the final results of analysis are included without explicit mathematical model. The main problem was formed – how to measure the anonymity of various P2P networks mathematically. Anonymity evaluation methodology was developed and adapted to P2P networks in order to measure anonymity. This methodology enables to calculate anonymity degree of both structured and unstructured P2P systems. Anonymity degree of structured P2P system is always lower than in unstructured system because in structured system nodes know the topology of the whole network and can simply get information about the possible routes. The proposed mathematical model allows to evaluate entropy of P2P system with random tunnel length. This model allows to calculate the total anonymity of P2P systems more precisely. Anonymity evaluation method and results obtained by it are important for the development of anonymous P2P systems based on random routing principles. |
