New algorithm examines the origins of Internet attacks
With the sparse Interference Algorithm (PDF) to the Swiss scientist Pedro Pinto and his colleagues at the Ecole Polytechnique Federale de Lausanne efficient search for the Urspüngen of Internet threats. So far, using institutions like the U.S. NSA brute-force methods to search in complex networks according to the source of an epidemic threat (malware, worms, Trojans, Internet rumors) - scanning all is in question node or address spaces, a rake-and time-consuming process.
Sparse Interference (in German as "scattered influence") into account, inter alia, the time factor in an avalanche occurring dissemination of information. Satisfy the algorithm according to the Wissenschafttler few clues by which can be identified with great certainty the origin, for example, a fraudulent e-mail or posting of a terrorist. It suffices therefore to monitor for example, only a few participants in a chat-round, a social network or a limited number of phones to reliably identify the originator. With a monitoring of randomly selected 25 percent of the participating nodes is apparent already a hit rate of 90 percent, rich in specific nodes selected for this success is already a choice of 5 percent.
They have fed the program a trial basis with the data from the published data exchange of 9/11-Terroristen, while three suspects have been released - which is as a "mastermind" of the attack have been identified, Pinto. The algorithm was suitable even for predicting the source and the future spread of a biological warfare agent or an infectious disease on the basis of existing data and existing distribution channels. The scientists' work was published in early August in the journal Physical Review Letters.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment