• The KIPS Transactions:PartD
• /
• v.19D no.1
• /
• pp.29-38
• /
• 2012

Software globalization is becoming more important worldwide but little is known about how globalization technologies are actually carried out in the Korean software industry. In this paper, we have designed Globalization Quality Management (GQM) framework and applied the framework to the industry domestically for the first time. GQM provides a structured and effective way for software organizations to adopt globalization practices and evaluate the results. GQM consists of three main components: 1) software quality management process, 2) globalization support model and 3) globalization assessment model. This framework also supports both plan-driven and iterative/incremental development methods. On the basis of the GQM, a survey study was conducted among software engineering professionals, gathering the data from 31 IT companies across 7 large-scale projects in Korea. The result indicate that the evaluation score for globalization capability is 2.47 and global readiness is 2.55 out of 5 points. Also It said that internationalized product design (32.9%) and global/local product requirements analysis (28%) are needed to be taken care of first.

• Journal of KIISE:Information Networking
• /
• v.29 no.1
• /
• pp.86-96
• /
• 2002

In video conferencing system using H.263 encoding and decoding time is as important as compression rate is. To reduce encoding time, a number of methods were proposed. We use a method of them that reduces the computational complexity in motion estimation. The complexity is determined by three factors, such as a cost function, a search range parameter, and a motion search algorithm. In fact, it takes a lot of time to encode the video data on account of the cost function factor. That's the reason that we use the factor to reduce encoding time. In this paper, we tried to reduce total encoding time by reducing the number of calling the cost function. In case of a little moving, our algorithm enabled faster motion searching than TSS(Three Step Search) and NTSS(New TSS). Here, we called the algorithm by an ITSS(Improved TSS) that improves a shortcoming of NTSS requiring more checkpoints than TSS. For an experimentation, our algorithm was compared to other algorithms using PSNR, file size and SAD call times.