This special issue of the journal includes a collection of excellent papers that had been accepted by The 4th International Conference on Ubiquitous Information Management and Communication. Among all the accepted papers presented at the conference, the best papers were recommended for publication. The set of recommended papers went through another round of review by independent reviewers. Finally, five papers were selected by the guest editors for inclusion in this special issue. These selected papers actually cover a wide spectrum of research problems in the area of ubiquitous information management and communication. The first three papers address research issues on the user-level application of ubiquitous computing in an internet environment and the other two papers consider more fundamental algorithmic and performance issues underneath, in order to support the development of useful applications.
   The first paper in this special issue is contributed by Masayuki Terada, Kozo Noaki and Kimihiko Sekino from NTT DOCOMO, Japan, with title “Sound Marketization of User-Generated Content based on a Fair Micro-Billing Scheme”. This paper addresses the user-generated content market problem from the interesting economic point of view, rather than software and information technology. In a commercial world, contributions by individuals should be recognized and paid for. However, in the existing user-generated content market, many user-generated content items are distributed for free, due to the lack of a proper mechanism to compensate the creators effectively. Various charging models exist, but they are not good enough to activate this market. Direct monetization models are risky for users with possibility of purchasing undesirable content, while indirect monetization models in general could not fairly reward the content creators. In this paper, the authors study why a sound market is difficult to achieve, and propose a smartcard-based fair micro-billing scheme towards a sound market by extending a direct monetization model. The proposed scheme is designed to achieve fairness in monetizing user-generated content. It allows users to pay for content in a “bit by bit” manner, so as to benefit both content creators and users, in significantly reducing the content quality risk for users and the “freeride” risks for creators. Secure and flexible implementation with current smartcards based on digital signatures is also presented in this paper.
   While the first paper studies fair sales of digital content from the market point of view, the second paper studies the security problems in real service providers from the information technology point of view. This paper, entitled “Vulnerabilities and Countermeasures of Commercial Online Music Streaming Services in Korea”, is authored by Sangsik Lee, Donghyun Choi, Dongho Won and Seungjoo Kim from Sungkyunkwan University, Korea. It investigates into the security problems with online music streaming services in Korea, provided by the three major providers, namely, Dosirak, Cyworld and Naver. Attacks that lead to an infringement of copyright with the three providers are studied. It is found that Dosirak stores user authentication information via cookies, which can be easily altered by an adversary to impersonate a paid user. For Cyworld and Naver, the two providers transfer music and authentication information without encryption. The playing of a song can be changed from the preview mode to the full mode by altering a particular field in the message, or by modifying the JavaScripts installed at the client machine for song playing. Countermeasures to protect copyrighted material from these vulnerabilities, namely, cookie, message and JavaScript alteration are proposed in this paper. Encryption and hash-based message authentication code are adopted for safe-guarding from cookie and message alteration. Compiled programs applying anti-reverse engineering techniques such as code obfuscation and anti-debugging are installed on client machines to combat the JavaScript alteration threat.
    In the third application paper entitled “Distributed Cooking Recipe Recommendation and Adaptation”, Qing Li, Wei Chen and Lijuan Yu from City University of Hong Kong, Hong Kong and Beijing Institute of Technology, China, propose a distributed recipe recommendation mechanism that utilizes social information for adaptive recipe recommendation in a peer-to-peer network. Noticing that only the basic recipes available in the network may not be sufficient to suit diverse users’ tastes and preferences about taste, smell, texture, temperature and so forth, they propose a recipe flavor model and derive new recipes based on existing ones along two aspects: ingredient and cooking styles/patterns, while satisfying two important user criteria: being interested in the recommended recipes and being able to utilize (e.g., cook) the recipes. The cooking procedure of a recipe is modeled as a graph. A distributed algorithm to generate communities sharing common preferences of recipe based on social network analysis and reputation at individual member, social, and community level is presented. A recipe similarity measure, a filtering algorithm to generate candidates of cooking recipes for recommendation, and a recipe adaptation method to better match users’ preferences are also proposed. Recipe requests are routed through the peer-to-peer network based on cooking based, flavor based, and hybrid recommendation approaches. Simulation experiments using a multi-agent system are conducted for the performance comparison with four methods. Finally, a prototype system for cooking recipe recommendation is displayed.
     The fourth paper moves down to the system level of software development in considering a remote storage system, authored by Masato Oguchi, Reika Higa, Kosuke Matsubara, Takao Okamawari and Saneyasu Yamaguchi from Ochanomizu University, Softbank Telecom, Kogakuin University and NTT DOCOMO, Japan. This paper is titled “Performance Improvement of iSCSI Remote Storage Access through Optimization of Multiple Layers”. iSCSI is one ofthe most popular protocols to access remote storage area network running over IP, but it experiences performance degradation under a high latency environment due to the complicated protocol stack. Sequential write is one of the major operations suffering from severe performance degradation. iSCSI remote storage access involves multiple hierarchical protocols, making it difficult for network engineers to identify the performance bottleneck. System tools which can monitor the complicated hierarchical protocols are developed, and to be precise, a kernel monitor tool, a packet monitor tool and data analysis tools have been developed. With the new tools, the authors have identified the cause of the iSCSI performance degradation problem. In particular, packet transmission from initiator to target is analyzed and temporary transmission suspension is observed. The impact on window size and TCP socket buffer are analyzed. This is followed by analysis in the kernel code on socket access. A software module is inserted for optimization of parameters after identifying the cause of data transmission suspension. After implementing the optimization steps, the throughput for iSCSI access is measured again and the performance is found to have improved drastically, close to the theoretical limit.
    This special issue is concluded by the contribution from Chin-Chen Chang, Kuo-Nan Chen, Zhi-Hui Wang and Ming-Chu Li, being a collaborative work between Taiwan and China, from three universities, i.e., Feng Chia University, National Chung Cheng University and Dalian University of Technology. In this final paper titled “Lossless Information Hiding in the VQ Index Table”, a reversible data hiding scheme based on a vector quantization index table is proposed. This scheme has a high hiding capacity for secret data and a satisfactory compression rate, and yet flexible with adjustable hiding capacity and compression rate. The approach proceeds by building a vector quantization index table and rearranging the codebook according to frequency. The codebook is then clustered into groups according to an adjustable parameter, providing spaces to hide the secret data in the index table. The original image is compressed using the vector quantization index table by replacing each block with an index in the codebook. Each index is then processed to embed a “digit” of the secret data in a base dependent on the number of groups selected. For common cases, the index is embeddable without incurring extra data. Only if the index is un-embeddable would there be some data overhead. This approach results in a low overhead in embedding the secret data. Experiments are conducted with six common images with varying parameter. The amount of secret data that can be conveyed is quite high, while the image compression rate is still good, as indicated by the data bit rate. A comparison with another recent algorithm is made and the proposed approach is performing better with higher hiding capacity and lower data bit rate.
    In summary, the collection of papers in this special issue covers the different aspects along the theme of the conference, in building useful software under an internet communication environment. We have to express our greatest gratitude to Prof. Won Kim, who had established the International Conference on Ubiquitous Information Management and Communication and brought about this special issue. We would also like to congratulate the contributing authors for their fine work.
    Finally, we are extending our big hands to the reviewers who have done an excellent job in helping to select the papers and providing all the constructive comments for the authors to improve their papers. Last but not least, we would like to thank the journal publisher and their staffs for their dedicated efforts in making this publication possible. With the collaborative efforts from all the parties, we are proud to have successfully compiled this special issue, with solid contributions from the International Conference on Ubiquitous Information Management and Communication.

[PDF]