Sangho Ha

Mapping XML documents to the object-relational form

In e-commerce, partners doing business need to exchange many forms of documents as a means of communication between them through the Internet, such as product catalogs, purchase orders, invoices and so on. XML, which is a new standard adopted by World Wide Web Consortium to complement HTML for data exchange on the Web, has been used as a language for describing those documents used in e-commerce such as in xCBL. For XML to be used effectively in e-commerce, we need a system to store XML data, retrieve the data if a request is given, create a required document based on the data, and send the created document to the partners. In this paper, we are concerned with storing XML data. We suggest a method to store XML documents in the object-relational form as an alternative to get advantages from the two approaches of object-oriented and relational table forms. We then implement it on the Oracle 8i database.

A Robust Printed Image Watermarking Based on Iterative Halftoning Method

In this paper, we propose a noble watermark embedding and extraction method for printed images. Watermark bits are hidden at pseudo-random locations within a printed halftone image during halftoning process. To remove the artifacts and distortion due to the inserted watermark bits, iterative error minimizing technique is used. During each iteration the pattern which decreases the halftone error is selected as the next pattern where the halftone error is defined as the difference of HVS-filtered printed halftone image and grayscale image. The inserted watermarks are extracted from the scanned images after a series of image processing operation which is described in detail. In order to be robust to cropping and rotation, the watermark is inserted periodically in halftone images. Experiments using real scanned images show that the proposed method is feasible method to hide the large amount of data within a halftone image without noticeable distortion and the watermark is robust to cropping and rotation.

Compilation of a Functional Language for the Multithreaded Architecture: DAVRID

Multithreading has been much focused on as one of the strongest parallel instruction execution paradigms for massively parallel processing. In this paper, we describe the compilation processes of parallel programming language Id- for a multithreaded architecture DAVRID(DAtaflow Von Neumann RISC hybrlD). Two fundamental issues in parallel processing, tolerance to communication latency and inexpensive synchronization, are solved by a compiler-controlled multithreading. Our compiler features a simple mechanism for handling closures, and a loop unfolding technique for handling parallel and sequential loops in separate, which can greatly contribute to parallel execution of loops.

A massively parallel multithreaded architecture: DAVRID

MPAs (massively parallel architectures) should address two fundamental issues for scalability: synchronization and communication latency. Dataflow architectures cause problems of excessive synchronization costs and inefficient execution of sequential programs while they offer the ability to exploit massive parallelism inherent in programs. In contrast, MPAs based on the von Neumann computational model may suffer from inefficient synchronization mechanism and communication latencies. DAVRID (Dataflow von Neumann, RISC Hybrid) is a massively parallel multithreaded architecture. By combining the advantages of the von Neumann model and the dataflow model, DAVRID preserves good single thread performance and tolerates latency and synchronization costs. We describe the DAVRID architecture and evaluate it through simulation results over several benchmarks.<<ETX>>

A Watermarking Method for Halftone Images Based on Iterative Halftoning Method

This paper proposes a new watermarking method to embed invisible data in halftone images. Halftone images are binary images appearing in printed matters such as books. The embedded data in halftone images can be used for copyright protection. The data hiding is integrated into halftoning operation. Watermark bits are hidden at pseudo-random locations within a halftone image during halftoning process. In order to minimize the distortions due to the inserted watermark, the proposed method tries to find the best neighborhood pattern using iterative binary search method. In order to cope with unintentional attacks such as cropping and distortions, the watermark is inserted periodically in halftone images. Experiments show that the proposed method can hide the relatively large amount of data within a halftone image without causing noticeable artifacts and the watermark is robust to crop and geometrical distortion.

Partitioning a lenient parallel language into sequential threads

Multithreading is attractive in a large-scale parallel system since it allows split-phase memory operations and fast context switching between computations without blocking the processor. The performance of multithreaded architectures depends significantly on the quality of multithreaded codes. In this paper, we describe an enhanced thread formation scheme to produce efficient sequential threads from programs written in Id/sup -/, a lenient parallel language. This scheme features graph partitioning based only on long latency instructions, a combination of multiple switches and merges introducing a generalized switch-and-merge, thread merging, and redundant arc elimination using thread precedence relations. Simulation results show that our scheme reduces control and branch instructions effectively.<<ETX>>

An Efficient Object Hierarchy Construction for Viewing Internet Products on Mobile Phones

With the advances of wireless technologies and mobile computing, m-commerce is being realized for many kinds of mobile devices. Service contents for m-commerce are usually newly written to meet specific characteristics of the target mobile devices. To avoid these formidable efforts, it is very important to effectively exploit the Internet merchant information currently served for e-commerce. However, bringing the Internet contents to mobile devices is far from straightforward due to the many limitations of mobile devices such as little memory, small displays, low processing speeds, and so forth. In this paper, assuming that the Internet merchant information is written in XML, we will suggest four methods to construct the object hierarchy for viewing the documents on the mobile phones. We then will compare them by experiment in terms of response times and memory space.

A watermarking method for halftone images based on direct binary search

Watermarking techniques previously developed deal with on-line digital data. In this paper we deal with printed images rather than on-line data. Printed image watermarking is performed by exploiting the printing process itself i.e. halftoning. It hides watermark data at pseudo-random locations within a halftone image. The proposed method tries to find the best halftone image which has the least perceived halftoning error by iterative direct binary search method. The artifacts due to the inserted watermark are removed by finding the best neighborhood pixel pattern. By integrating data hiding into halftoning operation, relatively large amount of data can be hidden within a halftone image while the visual quality of final halftone image is maintained. Experimental result shows that the proposed method generates watermarked halftone images with good visual quality.