• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.8
• /
• pp.1674-1683
• /
• 2003

The performance of a packet data multiplexer with stop­and­wait ARQ protocols under Markovian interruption is considered in this work It is assumed that the input process, into the system is Poisson process, and that the output channel is divided into a series of time slots and a data packet can be transmitted in a slot time. In this system the round­trip propagation delay is defined to be the frame time. It is modeled that the output channel can be blocked by some Markevian interruption, whose state change between the blocking and non­-blocking states is given by Markov process. The overall system has been analyzed by constructing a relationship, taking the Markovian interruption into account, about the buffer behavior between the successive frames of slots. The validity of this analytical results has been verified by computer simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.8
• /
• pp.1505-1512
• /
• 2009

This paper propose a new frame level rate control algorithm for improving video quality and decreasing quality variation of an entire video sequence in a very low bit rate environment. In the proposed scheme, the allocated bits to a GOP are distributed to each frame properly according to the frame characteristics as well as the buffer status and the channel bandwidth. The H.264 standard uses various coding modes and optimization methods to improve the compression performance, which makes it difficult to control the generated traffic accurately. In this paper, proper prediction models for low bit rate environments are lust proposed, and a target distortion is determined using the models. According to the target distortion, the bit budget is allocated to each frame. It is shown by experimental results that the new algorithm can generate the PSNR performance better than that of the existing rate control algorithm.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.3_4
• /
• pp.205-212
• /
• 2003

In this paper, we propose a simple and efficient hardware-supported antialiasing algorithm and its rendering scheme. The proposed method can efficiently reduce the required memory bandwidth as well as memory size compared to a conventional supersampling when rendering 3D models. In addition, it can provide almost the same high quality scenes as supersampling does. In this paper, we have introduced the RUF (Recently Used Fragment) buffer that stores some or whole parts of a fragment or two more the merged results of fragments that recently used in color calculation. We have also proposed a color calculation algorithm to deteriorate the image quality as referencing the RUF buffer. Because of the efficiency presented in the proposed algorithm, the more number of sampling points increases the more memory saving ratio we can gain relative to the conventional supersampling. In our simulation, the proposed method can reduce the amount of memory size by 31% and the memory bandwidth by 11% with a moderate pixel color difference of 1.3% compared to supersampling for 8 sparse sampling points.

• Journal of Korea Multimedia Society
• /
• v.13 no.3
• /
• pp.349-358
• /
• 2010

Adaptability is an important issue in video streaming over mobile environments, since the clients may request videos with great differences in their workload. In this paper, we propose the issues in limited bandwidth scheduling for efficient MPEG-4 video stream transmission over a mobile or wireless network. In the phase of admission control, the amount of bandwidth allocated to serve a video request is the mean bandwidth requirement of its requested video. The dynamic allocation of bandwidth in the phase of scheduling depends on the playback buffer levels of the clients with an objective to make it more adaptive to the playback situation of individual clients. In the proposed RTA scheduling protocol, more bandwidth may be allocated temporarily to the client whose buffer level is low. By employing the buffer level based scheduling policy, this protocol attempts to maximize the real-time performance of individual playback while minimizing the impact of transient overloading. Extensive simulation experiments have been performed to investigate the performance characteristics of the RTA protocol as comparing with BSBA protocol. This RTA protocol shows the better performance by transferring more frames than BSBA protocol.Computer simulations reveals that the standard deviation of the bit rate error of the proposed scheme is 50% less than that of the conventional method.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.820-826
• /
• 2016

There is a growing requirement for big data processing which extracts valuable information from a large amount of data. The Hadoop system employs the MapReduce framework to process big data. However, MapReduce has limitations such as inflexible and slow data processing. To overcome these drawbacks, SQL query processing techniques known as SQL-on-Hadoop were developed. Apache Tajo, one of the SQL-on-Hadoop techniques, was developed by a Korean development group. External merge sort is one of the heavily used algorithms in Tajo for query processing. The performance of external merge sort in Tajo is influenced by two parameters, sort buffer size and fanout. In this paper, we analyzed the performance of external merge sort in Tajo with various sort buffer sizes and fanouts. In addition, we figured out that there are two major causes of differences in the performance of external merge sort: CPU cache misses which increase as the sort buffer size grows; and the number of merge passes determined by fanout.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06d
• /
• pp.109-111
• /
• 2012

본 논문에서는 동적으로 현재화면의 최적밝기 정보를 구해 이 값을 이용하여 LCD 장치를 제어하고, 이를 통해 소모전력을 감소시키는 기법을 제안한다. 논문에서 제안된 기법은 프레임버퍼 이미지를 통해 얻은 현재 화면의 이미지 정보를 분석하여 계산된 현재 화면의 최적의 휘도 정보를 찾고 이 값에 맞는 LCD 장치의 최적 Backlight를 계산하는 과정을 거쳐 얻어진 최종 Backlight 레벨로 LCD의 Backlight를 제어하는 방법으로 소모 전력을 절감하는 기법이다. 이 기법을 적용하여 실험을 통해서 약 10%정도의 개선효과를 보였음을 밝힌다.

• Annual Conference of KIPS
• /
• 2015.04a
• /
• pp.588-590
• /
• 2015

본 논문에서는 웹 브라우저의 사용자 응답성을 개선하기 위해 뉴 메모리를 활용하는 기법을 제안한다. 프레임버퍼에 저장된 기본 웹 페이지의 실행 화면을 뉴 메모리에 캡쳐하고, 이를 웹 브라우저 실행 시 그대로 복원함으로써 사용자가 체감하는 웹 브라우저의 응답성을 크게 개선한다. 또한 제안하는 기법을 뉴 메모리 에뮬레이션 보드에 직접 구현하여 그 성능을 검증하고자 한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.229-231
• /
• 1998

본 논문에서 제안하는 매체접근제어 프로토콜은 활성 단말의 개수를 통계적 특성에 의하여 예측하고 이를 상향 요구 슬롯의 액세스 확률에 적용하여 요구슬롯의 수율(Throughput)을 높이도록 하였으며, 단말에세 슬롯을 할당하는 방식으로 고정할당법과 동적 할당법을 동시에 사용하였다. 고정 할당법은 모든 단말에게 최송한 제공될 수 있는 서비스 품질(QoQ: Quality of Server)을 보장하는 방식이며 동적할당법은 음성 단말에 대하여 버퍼의 상태에 따라 지정된 서비스 품질을 제공하기 위하여 각 단말에게 동적으로 슬롯을 할당하는 방법이다. 제안된 프로토콜에 대해 시뮬레이션을 통하여 음성 및 데이터 단말의 수율과 패킷 전송 지연을 구하고 음성 단말의 패킷 손실률을 분석하여 동적 할당법의 효율에 대하여 알아본다.

• Journal of the Korea Computer Graphics Society
• /
• v.14 no.1
• /
• pp.17-25
• /
• 2008

We proposed the method for photomosaic generation using a programmable GPU. We design vertices to generate a photomosaic through a graphics pipeline and suggest a texture representation of an image database whice is used for tile. Both the source image and the tiles are stored to texture, which are matched by a vertex shader and drawn by a fragment shader. This is much faster than several techniques which achieve the best match for each tile.

• Journal of Korea Multimedia Society
• /
• v.2 no.1
• /
• pp.80-88
• /
• 1999

In this paper, we propose a Special Memory for Graphics(SMGRA) which accelerates memory access time for graphics operations. The SMGRA has a rectangular array memory architecture which has already proposed by Whelan to process pixels in the rectangle area simultaneously, but the SMGRA should improve address decoding time and reduce the number of address pins by using address multiplexing scheme. The SMGRA has a Z-value comparator in the DRAM which is to convert read-modify-write Z buffer into single-write only operation that improves approximately 50% frame buffer access bandwidth.