Hiroaki Kikuchi

Secure instant messaging protocol preserving confidentiality against administrator

We study a security enhancement of instant messaging service. After the requirements and the assumptions are addressed, a modified Diffie-Hellman protocol suitable to instant messaging is presented. The main feature of the proposed protocol is to prevent malicious administrator from intercepting plain message and applying data mining techniques to obtain privacy of end users. From the viewpoint of communication and computational costs, the proposed protocol are evaluated and the comparison of some existing protocols is given. In addition, a trial implementation of secure instant messaging system is demonstrated.

Secure Multiparty Computation for Comparator Networks

We propose a multiparty protocol for comparator networks which are used to compute various functions in statistical analysis, such as the maximum, minimum, median, and quartiles, for example, through sorting and searching. In the protocol, all values which are inputted to a comparator network and all intermediate outputs are kept secret assuming the presence of an honest majority. We also introduce an application of the protocol for a secure (M + 1)-st price auction.

Multi-Bit Embedding in Asymmetric Digital Watermarking without Exposing Secret Information

A new method of multi-bit embedding based on a protocol of secure asymmetric digital watermarking detection is proposed. Secure watermark detection has been achieved by means of allowing watermark verifier to detect a message without any secret information exposed in extraction process. Our methodology is based on an asymmetric property of a watermark algorithm which hybridizes a statistical watermark algorithm and a public-key algorithm. In 2004, Furukawa proposed a secure watermark detection scheme using patchwork watermarking and Paillier encryption, but the feasibility had not tested in his work. We have examined it and have shown that it has a drawback in heavy overhead in processing time. We overcome the issue by replacing the cryptosystem with the modified El Gamal encryption and improve performance in processing time. We have developed software implementation for both methods and have measured effective performance. The obtained result shows that the performance of our method is better than Frukawas method under most of practical conditions. In our method, multiple bits can be embedded by assigning distinct generators in each bit, while the embedding algorithm of Frukawas method assumes a single-bit message. This strongly enhances capability of multi-bit information embedding, and also improves communication and computation cost.

De Morgan bisemilattice of fuzzy truth value

A fuzzy (linguistic) truth value is a truth value specified by a fuzzy set over a closed interval [0, 1]. Logical operations, defined according to the extension principle, do not satisfy all identities to be lattice because there are subnormal truth values and non-convex truth values that violate the absorption laws. In 2000, Brzozowski proposed de Morgan bisemilattice, which is generalized algebra of de Morgan lattice in order for applications in multi-valued simulations of digital circuits. This paper studies the algebraic structure of fuzzy truth values from a view point of Brozozowskis (2000) de Morgan bisemilattice. The main results include a necessary and sufficient condition of a subset of fuzzy truth values to be de Morgan bisemilattice.

Closures of fuzzy linguistic truth value with regards to the extension principle

We studied non-convex or sub-normal linguistic truth values that has not been considered so far. The logical operations based on the extension principle are closed into V/sub N/, V/sub C/ and V/sub R/ and arbitrary intersection of any of these is also closed. There are infinite number of subsets outside of the convex truth value, that are classified by the ranks. We show that a set of linguistic truth values of a common rank is not closed.

A characterization of Kleenean functions

The paper deals with Kleenean sanctions defined as fuzzy logic functions with constants. Kleenean functions provide a means of handling conditions of indeterminate truth value (ambiguous states) which ordinary classical logic (binary logic) cannot cope with. The paper clarifies a necessary and sufficient condition for a function to be a Kleenean function. The condition is provided with a set of two conditions, and it is shown that they are independent of each other.

Privacy preserving Web-based questionnaire

This paper proposes a secure protocol for Web-based questionnaire that preserves the privacy of responders and ensures the robustness. The proposed protocol is based on secure electronic voting protocol proposed by Cramer et al. confidentiality and efficiency, which are common requirements for both applications. This paper points out an issue particular in the secure questionnaire, that is, a requirement to deal with a multiple-choice question, and shows that the communication overhead grows exponentially with the number of choices n. To address the issue, the proposed protocol reduces the cost from /spl Theta/(2/sup n/) to /spl Theta/(n). The performance based on the experimental implementation is also shown.

Dual RSA accumulators and its application for private revocation check

This paper points out the privacy issue in the OCSP (Online Certificate Status Protocol), namely, the OCSP responder learns confidential information-who sends a message to whom. To preserve the privacy of the OCSP requester, this paper presents a cryptographic protocol for the authenticated dictionary, namely, an untrusted directory provides a verifiable answer to a membership query for a given element. In the protocol, a user is able to retrieve whether or not a target element belongs to a database that the directory has without revealing which element he/she wishes to know against the untrusted directory. The protocol requires linear exponentiations to the number of elements in the database, but achieves a constant size communication complexity between a user and a directory. The privacy of query is assured under the /spl Phi/-hiding assumption introduced by Cachin.

Webpage clustering - automated classification into jointly classified groups

A directory service for Webpages is widely used over the Internet. Webpages, however, are not always disjointly partitioned into subcategories, i.e., there are some pages to be belonging to multiple groups. To classify pages, conventional clustering algorithms are not adequately applied. This paper proposes a new algorithm of fuzzy clustering specialized for Webpage directory.

Applications of fuzzy logic functions to knowledge discovery in databases

This chapter is a summary of knowledge discovery algorithms that take an input of training examples of target knowledge, and output a fuzzy logic formula that best fits the training examples. The execution is done in three steps; first, the given mapping is divided into some Q-equivalent classes; second, the distances between the mapping and each local fuzzy logic function are calculated by a simplified logic formula; and last, the shortest distance is obtained by a modified graph-theoretic algorithm. After a fundamental algorithm for fitting is provided, fuzzy logic functions are applied to a more practical example of classification problem, in which expressiveness of fuzzy logic functions is examined for a well-known machine learning database.