• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.9
• /
• pp.1892-1897
• /
• 2012

In this paper, efficient CNU(Check Node Update) algorithms are analyzed for high speed LDPC decoding in DVB-S2 standard. In aspect to CNU methods, there are some kinds of CNU methods. Among of them, MP (Min Product) method is quite often used in LDPC decoding. However MP needs LUT (Look Up Table) that is critical path in LDPC decoding speed. A new SC-NMS (Self-Corrected Normalized Min-Sum) method is proposed in the paper. NMS needs only normalized scaling factor instead of LUT and compensates the overestimation of MP approximation. In addition, SC method is proposed. It gives a faster convergence toward a decoded codeword. If a message change its sign between two iterations, it is not reliable and to avoid to propagate noisy information, its module is set to 0. The performance of SC-NMS has a little degrade compare to MP by 0.1 dB, however considering computational complexity and decoding speed, SC-NMS algorithm is optimal method for CNU algorithm.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.7
• /
• pp.110-120
• /
• 2016

A power-efficient neuromodulator is needed for implantable systems. In spite of their stimulation signal's simplicity of wave shape and waiting time of MCU(micro controller unit) much longer than execution time, there is no consideration for low-power design. In this paper, we propose a novel of low-power algorithm based on the characteristics of stimulation signals. Then, we designed and implement a neuromodulation software that we call NMS(neuro modulation simulation). In order to implement low-power algorithm, first, we analyze running time of every function in existing NMS. Then, we calculate execution time and waiting time for these functions. Subsequently, we estimate the transition time between active mode (AM) and low-power mode (LPM). By using these results, we redesign the architecture of NMS in the proposed low-power algorithm: a stimulation signal divided into a number of segments by using characteristics of the signal from which AM or LPM segments are defined for determining the MCU power reduces to turn off or not. Our experimental results indicate that NMS with low-power algorithm reducing current consumption of MCU by 76.31 percent compared to NMS without low-power algorithm.

• Journal of Internet Computing and Services
• /
• v.13 no.2
• /
• pp.33-40
• /
• 2012

Recently, according to increasing the network traffic in the IMS, the role of Network Management System (NMS) is very important because of limited network resource. NMS can perform two kinds of routing ways with the capability of static or dynamic routing. The way of A dynamic routing is more efficient than static routing one because it can make the flow of traffic changeful among nodes in the IMS. Therefore, in this paper, we suggest a management function of NMS, using a dynamic routing algorithm for managing the CSCFs in the IMS. And then we analyze the algorithm by measuring the performance of PoC, one of the prominent application services to be deployed in the IMS.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.11
• /
• pp.1248-1255
• /
• 2012

LDPC decoder architectures are generally classified into serial, parallel and partially parallel architectures. Conventional method of LDPC decoding in general give rise to a large number of computation operations, mass power consumption, and decoding delay. It is necessary to reduce the iteration numbers and computation operations without performance degradation. This paper studies horizontal shuffle scheduling(HSS) algorithm and self-correction normalized min-sum(SC-NMS) algorithm. In the result, number of iteration is half than conventional algorithm and performance is almost same between sum-product(SP) and SC-NMS. Finally, This paper implements high-speed LDPC decoder based on FPGA. Decoding throughput is 816 Mbps.

• Journal of Korea Multimedia Society
• /
• v.24 no.3
• /
• pp.416-422
• /
• 2021

Due to the large differences in human activity within classes, the large similarity between classes, and the problems of visual angle and occlusion, it is difficult to extract features manually, and the detection rate of human behavior is low. In order to better solve these problems, an improved Faster R-CNN-based detection algorithm is proposed in this paper. It achieves multi-object recognition and localization through a second-order detection network, and replaces the original feature extraction module with Dense-Net, which can fuse multi-level feature information, increase network depth and avoid disappearance of network gradients. Meanwhile, the proposal merging strategy is improved with Soft-NMS, where an attenuation function is designed to replace the conventional NMS algorithm, thereby avoiding missed detection of adjacent or overlapping objects, and enhancing the network detection accuracy under multiple objects. During the experiment, the improved Faster R-CNN method in this article has 84.7% target detection result, which is improved compared to other methods, which proves that the target recognition method has significant advantages and potential.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.18-23
• /
• 2001

멀티미디어 통신 등 서비스 광대역화를 위하여 PSTN, PCN 및 B-lSDN 등과 같은 기존의 망과 연동되게 될 IMT-2000 통신망은 기존의 통신망들이 서로 다른 하드웨어와 운영체제 등 상이한 플랫폼 환경 하에서 개발되어 있는 관계로 망관리 시스템을 구축하는 과정에서 통합 및 유지보수 상의 여러 문제점이 대두되게 된다. 이와 같은 문제점을 해결하기 위하여 본 논문에서는 I-Farmer 모델(Intelligent-Farmer Model)에 기반한 ITS(ISM-Node to SIB) 알고리즘을 제안한다. ITS 알고리즘은 I-Farmer 모델로 설계한 에이전트 시스템 중 Entity Node 및 ILB / OLB 컴포넌트를 AIN의 GFP(Global Functional Plane)내의 SIB 및 시스템 응용프로그램 생성 컴포넌트인 지능망 ASIB로 변형(transformation)해주는 알고리즘이다.

• 한국IT서비스학회:학술대회논문집
• /
• 2002.11a
• /
• pp.444-450
• /
• 2002

실세계의 사물을 다중측면에 의하여 분석한 후 개체노드, 측면개체노드, 균일성 개체 및 다중화 추상화라는 여러 개념을 이용하여 디자인할 수 모델인 I-Farmer 모델의 구현 방안을 제시하였다. 컴포넌트는 컴포지션에 의한 소프트웨어 블록 재사용성을 제공하며 컴포넌트는 이들의 수행하는 기능에 대한 명세를 나타내는 인터페이스 명세 모델이 반드시 필요하게 된다. 본 논문에서는 I-Farmer 모델링 방법론에 의하여 디자인되어진 시스템을 컴포넌트 기반개발의 인터페이스 명세 모델로의 변형(transformation) 메카니즘인 ITI 알고리즘 및 CORBA / JAVA RMI COM IDL과의 상호관련성을 제시한다.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.59 no.1
• /
• pp.53-57
• /
• 2010

Low-density parity-check (LDPC) codes with belief-propagation (BP) algorithm achieve a remarkable performance close to the Shannon limit at reasonable decoding complexity. Conventionally, each iteration in decoding process contains two steps, the horizontal step and the vertical step. In this paper, an efficient implementation of the adaptive offset min-sum (AOMS) algorithm for decoding LDPC codes using the single-step method is proposed. Furthermore, the performances of the AOMS algorithm compared with belief-propagation (BP) algorithm are investigated. The algorithms using the single-step method reduce the implementation complexity, speed up the decoding process and have better efficiency in terms of memory requirements.

• Smart Structures and Systems
• /
• v.15 no.3
• /
• pp.735-749
• /
• 2015

The major difficulty of using Bayesian probabilistic inference for system identification is to obtain the posterior probability density of parameters conditioned by the measured response. The posterior density of structural parameters indicates how plausible each model is when considering the uncertainty of prediction errors. The Markov chain Monte Carlo (MCMC) method is a widespread medium for posterior inference but its convergence is often slow. The differential evolution adaptive Metropolis-Hasting (DREAM) algorithm boasts a population-based mechanism, which nms multiple different Markov chains simultaneously, and a global optimum exploration ability. This paper proposes an improved differential evolution adaptive Metropolis-Hasting algorithm (IDREAM) strategy to estimate the posterior density of structural parameters. The main benefit of IDREAM is its efficient MCMC simulation through its use of the adaptive Metropolis (AM) method with a mutation strategy for ensuring quick convergence and robust solutions. Its effectiveness was demonstrated in simulations on identifying the structural parameters with limited output data and noise polluted measurements.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.3B
• /
• pp.214-219
• /
• 2011

The IMS (IP Multimedia Subsystem) is used for providing multimedia services to network provider. The IMS is composed of many nodes, depending on their function. According to the status (fault, performance, etc.) of each node, we need to provide dynamic routing algorithm. In this paper, we propose a dynamic routing algorithm for management of the IMS nodes using SNMP based on FCAPS (Fault, Configuration, Accounting, Performance and Security) in the IMS network.