• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.4
• /
• pp.381-392
• /
• 2003

Component-based Development(CBD) based on the software reuse has been more attractive from software companies that want to enhance software productivity. However, since component identification process is mainly dependent on domain expert´s intuition and experience, it was very difficult to develop tools for supporting the component identification process. In this paper, we propose a systematic procedure of identifying reusable component by using object dependencies and object usages and provide a design and implementation of its supporting tool. In object-oriented domain models. there exists several diagrams which are described in different viewpoints. From these diagrams, object dependency and object usages are extracted and merged into an object dependency network, which is a basis for performing a comfonent identification algorithm. Finally, through a case study of internet banking system, we evaluate the applicability of the proposed identification process and tool.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.1
• /
• pp.37-48
• /
• 2004

Important Java network application services have been rapidly increased along with the growth of the Internet. So, it is desirable for such applications to serve transparently, continuously and safely even if the network is temporally partitioned or certain hosts running those services are crashed down. To satisfy such requirements, many group communication systems have been developed. However, existing Java-based group communication systems do not support both the extended virtual synchrony and various types of message delivery such as FIFO, causal, total and safe delivery service. In this paper, we present the design and implementation of a Java group communication system, named JACE, supporting various types of message delivery between group members and the extended virtual synchrony model. The JACE system consists of a number of protocol modules which can be stacked on top of each other in a variety of ways. In addition, using the JACE system, we have developed an experimental UDDI registry for discovering and publishing information about Web services.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.3
• /
• pp.356-363
• /
• 2000

Java programming language is designed to run effectively on internet and distributed network environments. However, because it has a deficit to be executed by the interpreter method on each platform, to execute Java programs efficiently the code generation system which transforms Bytecode into SPARC code as target machine code must be developed. In this paper, we implement a code generation system which translates Bytecode into SPARC code using the retargetable code generating techniques. For the sake of code expander, we wrote a Bytecode table describing a rule of SPARC code generation from Bytecode, and implemented the information extractor transforming Bytecode to suitable form during expanding of source code from class file. The information extractor determines constant pool entry of each Bytecode instruction operand and then the code expander translates the Bytecode into SPARC code accoring to the Bytecode table. Also, the retargetable code generation system can be systematically reconfigured to generate code for a variety of distinct target computers.

• The Journal of the Korea Contents Association
• /
• v.11 no.12
• /
• pp.639-648
• /
• 2011

Since the emergence of the smartphone and tablet PC whose wireless internet accessibility has improved, personal computing environment has shifted from PCs to mobile web based mobile devices, and therefore the use of information and the communication method have been changing rapidly. Under the circumstance, domestic universities have actively been establishing mobile websites for smartphone users. Such mobile websites, however, focused on diverse functions and a showy design rather than on usability or accessibility, and were developed simply for the purpose of public relations. For this reason, students are now faced with difficulties in using the websites. Therefore, this paper proposes a novel web interface with the consideration of usability and accessibility in order for users to easily use and assess proper information. The matters to be taken into account in developing a mobile website are an intuitive interface design, user experiences, a user customized interface and compliance with web standards.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.12
• /
• pp.998-1002
• /
• 2009

In IPTV multicast architecture, the IGMP(Internet Group Management Protocol) is used for access networks. This protocol supports the functionality of join or leave for a specific multicast channel group. But, malicious attackers can disturb legitimate users being served appropriately. By using spoofed IGMP messages, attackers can hi-jack the premium channel, wasting bandwidth and exhausting the IGMP router's resources. To prevent the message spoofing, we can introduce the packet-level authentication methods. But, it causes the additional processing overhead to an IGMP processing router, so that the router is more susceptible to the flooding attacks. In this paper, we propose the two-level authentication scheme in order to mitigate the IGMP flooding attack.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.320-332
• /
• 2007

Recently, new interests on digital contents and UCC(User Created Content)s are growing fast through the development of internet. However, there have been many side-effects on those interests. The representative problems are perversing illegal copies and the distributions for personal valuable digital contents to unauthorized anonymous users. These decrease creation of good digital contents by interfering with the growth of information technology industry and the content provider#s creative will. To resolve these problems, in the paper, we propose a novel auction system for multimedia contents and bidding processes. The system applies the concept of used goods onto digital contents. Especially, the auction system is based on mobile environment. Finally, the performance evaluation shows that those main auction process algorithms indicate the time complexity of logarithm scale for insertions and searches. Therefore, the performance of the system is not significantly influenced by the amount of contents even though the volume of contents in the system is increasing.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.300-310
• /
• 2007

The Session Initiation Protocol (SIP) is one of the major protocols used in call-setup over IP telephony. The SIP-signaled calls use many-sided states according to a request of user. In this paper, we suggest seven states and some events that help developers to design and implement new applications efficiently. And they enable an object-oriented design of the system. If you design the call-setup procedure only by the processing model suggested in RFC 3261 over commercial network, a fatal error may occur in the system because of heavy data traffic or unpredicted exception cases. However, according to the suggested seven states, if they are predefined events in the current system state, the standardized processing routine is executed. Otherwise, they can be processed by the exception routine in system. All event processing routines are designed and implemented using Finite State Machine (FSM).

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.6
• /
• pp.614-618
• /
• 2008

As the multimedia data increase a lot with the rapid development of the Internet, an efficient retrieval technique focusing on individual users is required based on the analyses of such data. However, user modeling services provided by recent web sites have the limitation of text-based page configurations and recommendation retrieval. In this paper, we construct the user preference model with a Bayesian network to apply the user modeling to video retrieval, and suggest a method which utilizes probability reasoning. To do this, context information is defined in a real office environment and the video scripts acquired from established cameras and annotated the context information manually are used. Personal information of the user, obtained from user input, is adopted for the evidence value of the constructed Bayesian Network, and user preference is inferred. The probability value, which is produced from the result of Bayesian Network reasoning, is used for retrieval, making the system return the retrieval result suitable for each user's preference. The usability test indicates that the satisfaction level of the selected results based on the proposed model is higher than general retrieval method.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.614-618
• /
• 2016

As information on the internet and the data from smart devices are growing, the amount of stream data is also increasing in the real world. The stream data, which is a potentially large data, requires online learnable models and algorithms. In this paper, we propose a novel class of models: predictive convolutional neural networks to be able to perform online learning. These models are designed to deal with longer patterns as the layers become higher due to layering convolutional operations: detection and max-pooling on the time axis. As a preliminary check of the concept, we chose two-month gathered GPS data sequence as an observation sequence. On learning them with the proposed method, we compared the original sequence and the regenerated sequence from the abstract information of the models. The result shows that the models can encode long-range patterns, and can generate a raw observation sequence within a low error.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.583-590
• /
• 2016

Recently, the rapidly increasing interest in IoT in almost all areas of casual human life has led to wide acceptance of speech recognition as a means of HCI. Simultaneously, the demand for speech recognition systems for mobile environments is increasing rapidly. The server-based speech recognition systems are typically fast and show high recognition rates; however, an internet connection is necessary, and complicated server computation is required since a voice is recognized by units of words that are stored in server databases. In this paper, we present a novel method for recognizing the Korean vowel 'ㅗ', as a part of a phoneme based Korean speech recognition system. The proposed method involves analyses of waveform patterns in the time domain instead of the frequency domain, with consequent reduction in computational cost. Elementary algorithms for detecting typical waveform patterns of 'ㅗ' are presented and combined to make final decisions. The experimental results show that the proposed method can achieve 89.9% recognition accuracy.