• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2012년도 춘계학술발표대회
• /
• pp.607-608
• /
• 2012

Data aggregation is a fundamental problem in wireless sensor networks which attracts great attention in recent years. Delay and energy efficiencies are two crucial issues of designing a data aggregation scheme. In this paper, we propose a distributed, energy efficient algorithm for collecting data from all sensor nodes with the minimum latency called Delay-aware Power-efficient Data Aggregation algorithm (DPDA). The DPDA algorithm minimizes the latency in data collection process by building a time efficient data aggregation network structure. It also saves sensor energy by decreasing node transmission distances. Energy is also well-balanced between sensors to achieve acceptable network lifetime. From intensive experiments, the DPDA scheme could significantly decrease the data collection latency and obtain reasonable network lifetime compared with other approaches.

• ETRI Journal
• /
• 제42권5호
• /
• pp.721-733
• /
• 2020

Recently, there is an increasing demand for ultra-low-latency (ULL) services such as factory automation, autonomous driving, and telesurgery that must meet an end-to-end latency of less than 10 ms. Fifth-generation (5G) New Radio guarantees 0.5 ms one-way latency, so the feasibility of ULL services is higher than in previous mobile communications. However, this feasibility ensures performance at the radio access network level and requires an innovative 5G network architecture for end-to-end ULL across the entire 5G system. Hence, we survey in detailed two the 3rd Generation Partnership Party (3GPP) standardization activities to ensure low latency at network level. 3GPP standardizes mobile edge computing (MEC), a low-latency solution at the edge network, in Release 15/16 and is standardizing time-sensitive communication in Release 16/17 for interworking 5G systems and IEEE 802.1 time-sensitive networking (TSN), a next-generation industry technology for ensuring low/deterministic latency. We developed a 5G system based on 3GPP Release 15 to support MEC with a potential sub-10 ms end-to-end latency in the edge network. In the near future, to provide ULL services in the external network of a 5G system, we suggest a 5G-IEEE TSN interworking system based on 3GPP Release 16/17 that meets an end-to-end latency of 2 ms.

• 한국인터넷방송통신학회논문지
• /
• 제18권2호
• /
• pp.55-62
• /
• 2018

VANET은 급격히 부상하고 있는 서비스 영역으로 수 ms 수준의 엄격한 latency 요구사항을 가진다. 이러한 수준의 낮은 latency를 보장하기 위해 latency가 보장되면서 coordinator에 의한 scheduling 없이 join/leave가 자유로운 Autonomous TDMA(ATDMA)가 제안된 바 있다. 본 연구에서는 hidden node, 채널 페이딩, 노드 밀집도 변이 등이 존재하는 non-perfect decoding 환경에서 동작하도록 ATDMA를 확장한 ATDMA revision(ATDMA-R)을 고안하였다. 또한 시뮬레이션을 통해 ATDMA와 ATDMA-R의 성능을 비교하여 ATDMA 대비 ATMDMA-R이 다양한 실제적 환경에서 강인성을 보이는 것을 확인하였다.

• 전자공학회논문지
• /
• 제51권11호
• /
• pp.83-93
• /
• 2014

최근 수년간 전기적 상호 연결 (electrical interconnect, EI) 기반 네트워크-온-칩 (Network-on-Chip, NoC) 에 대한 연구가 활발히 진행되고 있는 가운데, 궁극적으로 금속 배선은 대역폭, 응답 시간(latency), 전력 소모 등에서 물리적 한계에 직면할 것으로 예상된다. 실리콘 포토닉스(silicon photonics) 기술 발전으로 광학적 상호 연결(optical interconnect, OI)을 결합한 하이브리드 광학 네트워크-온-칩(Hybrid Optical NoC, HONoC)이 이러한 문제를 극복하기 위한 유망한 해결책으로 부각되고 있다. 한편 시스템-온-칩(System-on-Chip, SoC)은 높은 에너지 효율을 위하여 이기종 멀티 코어(Heterogeneous multi-core)로 구성되고 있어서 정형화된 토폴로지 기반 NoC 아키텍처의 확장이 필요하다. 본 논문에서는 타깃 애플리케이션 트래픽 특성을 고려한 에너지 및 응답 시간 최적화 하이브리드 광학 네트워크-온-칩의 토폴로지 설계 기법을 제안한다. 유전자 알고리즘을 이용하여 구현하였고, 실험 결과 평균 전력손실은 13.84%, 평균 응답 시간은 28.14% 각각 감소하였다.

• 한국통신학회논문지
• /
• 제32권3C호
• /
• pp.205-215
• /
• 2007

This paper presents low latency and/or computation algorithms of iterative codes of turbo codes, turbo product codes and low density parity check codes for use in wireless broadband communication systems. Due to high coding complexity of iterative codes, this paper focus on lower complexity and/or latency algorithms that are easily implementable in hardware and further accelerate the decoding speed.

• 전자통신동향분석
• /
• 제32권5호
• /
• pp.74-84
• /
• 2017

In wireless access networks, it is extremely important to provide high quality of real time and interactive services, including voice and video traffic. Furthermore, low latency communication is shifting toward new paradigm which enhances user's high quality of experience, meeting the requirements for specific applications such as tactile internet, remote-control robot and machines, and mission critical application. In this paper, we introduce the approaches to achieve the extremely low latency service. The approaches include the core requirements and the key technologies providing low latency communication maintaining high reliability in wireless access networks.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권2호
• /
• pp.17-22
• /
• 2022

Due to the recent advances in high-speed communication technologies as well as the easy production of high-quality video contents, IPTV is becoming increasingly popular. Meanwhile, as the number of IPTV channels increases, channel search time to find the desired channel keeps increasing. In this paper, we discuss how to improve the channel search latency in IPTV, and introduce caching and prefetching techniques that are widely used in memory management systems. Specifically, we adopt memory replacement, prefetching, and caching techniques in IPTV channel search interfaces and show the effectiveness of these techniques as the number of channels are varied.

• 한국통신학회:학술대회논문집
• /
• 한국통신학회 2004년도 하계종합학술발표회논문초록집
• /
• pp.178-178
• /
• 2004

• 한국정보통신학회논문지
• /
• 제24권6호
• /
• pp.751-758
• /
• 2020

사물인터넷을 비롯한 다양한 형태의 통신을 활용한 응용분야가 생겨남에 따라 통신의 형태가 다양해지고 있다. 어떠한 응용은 빠른 피드백을 요구하기도 하고, 다른 응용은 수신여부와는 상관없이 데이터를 계속 보내기도 한다. 한 방향으로 데이터를 전송하는 것은 반대 방향 응답에 의해 확인될 수 있다. 이러한 정보 교환은 양방향 통신을 형성한다. 송신자는 수신자의 응답에 따라 다음 동작을 결정하므로 빠른 응답을 제공하는 것이 중요하다. 본 논문에서는 셀룰러 네트워크에서 양방향 통신의 전송지연 성능을 향상시키기 위한 기지국 협력 전송 기법을 제안한다. 양방향 트래픽뿐만 아니라 동일한 방향의 트래픽을 가진 협력 기지국을 활용하는 양방향 통신 기법을 고안하였다. 제안한 방식을 통하여 지연 시간 성능이 향상됨을 확인하였다.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제15권5호
• /
• pp.570-576
• /
• 2015

Existing flash storage devices such as universal flash storage and solid state disk support command queuing to improve storage I/O bandwidth. Command queuing allows multiple read/write requests to be pending in a device queue. Because multi-channel and multi-way architecture of flash storage devices can handle multiple requests simultaneously, command queuing is an indispensable technique for utilizing parallel architecture. However, command queuing can be harmful to the latency of fsync system call, which is critical to application responsiveness. We propose a dynamic queue depth adaptation technique, which reduces the queue depth if user application is expected to send fsync calls. Experiments show that the proposed technique reduces the fsync latency by 79% on average compared to the original scheme.