• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04d
• /
• pp.130-132
• /
• 2003

화상회의 시스템에서 원활한 동영상의 전달을 위해서는 주어지는 처리시간에 대하여 그에 상응하는 품질을 보장하는 프레임 스케줄링 방법이 필수적이다 영상 복원 작업에 주어지는 처리시간이 증가함에 따라 그 실행 결과로 얻어지는 영상의 품질이 함께 증가할 것이 요구된다. 본 논문에서는 영상회의 시스템에서의 압축된 영상의 복원 과정에서 프로세서의 처리시간에 따르는 결과 영상의 품질에 대한 QoS함수를 도출하고 이를 IRIS 스케줄링 방안에 적용하는 온-라인 스케줄링 알고리즘을 제안한다. 본 논문에서 제시하는 알고리즘에 대한 모의실험의 결과는 주어진 시스템 성능에 H하여 영상회의에 참여한 모든 화자 영상의 총체적 화질을 향상시킬 수 있음을 보여준다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.1
• /
• pp.12-21
• /
• 2003

In this paper, we propose an on-line scheduling algorithm with the goal of maximizing the total reward of IRIS (Increasing Reward with Increasing Service) real-time tasks that have reward functions and arrive dynamically into the system. We focus on enhancing the performance of scheduling algorithm, which W.: based on the following two main ideas. First, we show that the problem to maximize the total reward of dynamic tasks can also be solved by the problem to find minimum of maximum derivatives of reward functions. Secondly, we observed that only a few of scheduled tasks are serviced until a new task arrives, and the rest tasks are rescheduled with the new task. Based on our observation, the Proposed algorithm doesn't schedules all tasks in the system at every scheduling print, but a part of tasks. The performance of the proposed algorithm is verified through the simulations for various cases. The simulation result showed that the computational complexity of proposed algorithm is$O(N_2)$ in the worst case which is equal to those of the previous algorithms, but close to O(N) on the average.

• The KIPS Transactions:PartA
• /
• v.10A no.3
• /
• pp.181-188
• /
• 2003

We propose a heuristic on-line scheduling algorithm for the IRIS (Increasing Reward with Increasing Service) tasks, which has low computation complexity and produces total reward approximated to that of previous on-line optimal algorithms. The previous on-line optimal algorithms for IRIS tasks perform scheduling on all tasks in a system to maximize total reward. Therefore, the complexities of these algorithms are too high to apply them to practical systems handling many tasks. The proposed algorithm doesn´t perform scheduling on all tasks in a system, but on (constant) W´s tasks selected by a predefined task selection policy. The proposed algorithm is based on task selection policies that define how to select tasks to be scheduled. We suggest two simple and intuitive selection policies and a generalized selection policy that integrates previous two selection policies. By narrowing down scheduling scope to only W´s selected tasks, the computation complexity of proposed algorithm can be reduced to O(Wn). However, simulation results for various cases show that it is closed to O(W) on the average.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05d
• /
• pp.600-604
• /
• 2002

본 연구에서는 시스템 과부하로 인해 충분한 실행시간을 가지지 못한 태스크의 중간 결과도 수용 가능한 범위 내에서 인정하는 멀티미디어 응용을 위한 동적 태스크 스케줄링 알고리즘을 제안하며, 스케줄링 알고리즘의 시간 복잡도보다는 실행 오버헤드론 줄이는데 초점을 둔다. 시뮬레이션 결과 태스크 도착율과 스케줄링 윈도우 크기에 따라 스케줄링 성능이 달라 진다는 것을 확인하고, 적절한 스케줄링 윈도우 크기를 결정할 수 있는 방안에 대해 논의한다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.7
• /
• pp.387-396
• /
• 2004

In this paper, we propose an algorithm that distributes processor time to the tasks decoding encoded frames with a goal maximizing total QoS (quality of services) of video conference system. An encoded frame has such a characteristic that the QoS of recovered frame image also increases as the processor time given for decoding the frame gets to increase. Thus, the quality of decoded image for each frame can be represented as a QoS function of the amount of service time given to decode. In addition, every stream of video conference has close time-dependency between continuous frames belonging to the same stream. Based on the time-dependency and QoS functions, we propose an on-line frame scheduling algorithm which does not schedule all frames in the system but just a few frames while maximizing total QoS of video streams in the conference. The simulation results show that, as the system load gets to increase, the proposed algorithm compared to the existing EDF algorithm can reduce the quality of decoded frame images more smoothly and show the movements of conference attendees more naturally without short cutting.