庞又儇邑.PDFVIP

A General Mo del for Authentic Data Publication  Chip Martel , Glen Nuckolls, Michael Gertz, Prem Devanbu, April Kwong Stuart G. Stubblebine Department of Computer Science CertCo University of California 55 Broad Street { Suite 22 Davis, CA 95616 USA New York, NY 10004 fmartel jnuckolls jgertz jdevanbu jkwonga g@cs . stuart@ Abstract Query answers from on-line databases can easily b e corrupted by hackers or malicious database publishers. Thus it is imp ortant to provide mechanisms which allow clients to trust the results from on-line queries. Authentic publication is a novel scheme which allows untrusted publishers to securely answer queries from clients on b ehalf of trusted o-line data owners. Publishers validate answers using compact, hard-to-forge veri cation objects (V O s), which clients can check eÆciently. This approach provides greater scalability (by adding more publishers) and b etter security (on-line publishers dont need to b e trusted). To make authentic publication attractive, it is imp ortant for the V O s to b e small, eÆciently com- putable and veri able. This has led researchers to develop several dierent data representations for eÆcient V O computation. In this pap er, we develop a novel, data mo del called Search DA Gs. Within this mo del, we develop a generalized algorithm for the construction of V O s. We show that the V O s thus constructed are secure, compact and eÆci