• Electronics and Telecommunications Trends
• /
• v.33 no.5
• /
• pp.121-134
• /
• 2018

The Von Neumann based architecture of the modern computer has dominated the computing industry for the past 50 years, sparking the digital revolution and propelling us into today's information age. Recent research focus and market trends have shown significant effort toward the advancement and application of artificial intelligence technologies. Although artificial intelligence has been studied for decades since the Turing machine was first introduced, the field has recently emerged into the spotlight thanks to remarkable milestones such as AlexNet-CNN and Alpha-Go, whose neural-network based deep learning methods have achieved a ground-breaking performance superior to existing recognition, classification, and decision algorithms. Unprecedented results in a wide variety of applications (drones, autonomous driving, robots, stock markets, computer vision, voice, and so on) have signaled the beginning of a golden age for artificial intelligence after 40 years of relative dormancy. Algorithmic research continues to progress at a breath-taking pace as evidenced by the rate of new neural networks being announced. However, traditional Von Neumann based architectures have proven to be inadequate in terms of computation power, and inherently inefficient in their processing of vastly parallel computations, which is a characteristic of deep neural networks. Consequently, global conglomerates such as Intel, Huawei, and Google, as well as large domestic corporations and fabless companies are developing dedicated semiconductor chips customized for artificial intelligence computations. The AI Processor Research Laboratory at ETRI is focusing on the research and development of super low-power AI processor chips. In this article, we present the current trends in computation platform, parallel processing, AI processor, and super-threaded AI processor research being conducted at ETRI.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.4
• /
• pp.67-75
• /
• 1998

In general, Evoluationary Algorithm(EAs) are refered to as methods of population-based optimization. And EAs are considered as very efficient methods of optimal sytem design because they can provice much opportunity for obtaining the global optimal solution. This paper presents a co-evolution scheme of artifical neural networks, which has two different, still cooperatively working, populations, called as a host popuation and a parasite population, respectively. Using the conventional generatic algorithm the host population is evolved in the given environment, and the parastie population composed of schemata is evolved to find useful schema for the host population. the structure of artificial neural network is a diagonal recurrent neural netork which has self-feedback loops only in its hidden nodes. To find optimal neural networks we should take into account the structure of the neural network as well as the adaptive parameters, weight of neurons. So we use the genetic algorithm that searches the structure of the neural network by the co-evolution mechanism, and for the weights learning we adopted the evolutionary stategies. As a results of co-evolution we will find the optimal structure of the neural network in a short time with a small population. The validity and effectiveness of the proposed method are inspected by applying it to the stabilization and position control of the invered-pendulum system. And we will show that the result of co-evolution is better than that of the conventioal genetic algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.1B
• /
• pp.125-135
• /
• 2010

Recently, the problem of global warming has issued seriously, so Green IT(Information Technology) using RFID/USN is concerned in order to solve environmental problems. In this paper, we implemented that public area like an underground parking lot could reduce unnecessary energy consumption used by lighting control system based on Ubiquitous Sensor Networks. The lighting control system for underground parking lot is comprised of intersection nodes for watching cars enter and leave, and light node for controlling light. It applies the routing protocol based on hierarchical cluster, hierarchical addressing method, and probability filtering method for the specific place like an underground parking lot. The result after experiments shows that the lighting control system could decrease electrical energy consumption to around 61.7%.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.5
• /
• pp.1-10
• /
• 1998

Satellite communication is getting more important in the coming 21st century because of its wide areas sevice capability, ease of access, and fast channel establishment. As such, satellite communication networks will be the basis of the global communication system in cooperation with the ground ATM networks. In this paper, we consider an efficient transmission methodology of ATM cells over the satellite communication channel. We first analyze possible bottlenecks and performance deterioration factors in the case, and then propose an enhanced cell trasmission mechanism. In order to use satellite channel for ATM cell transmission, the application of complicated channel coding is inevitable. However, the forwared error control such as convolutional encoding brings forth burst errors, which calls for the application of some kind of interleaving mechanism to randomize the burst errors at the receiver. Another aspect which should b econsidered in satellite communication system is the inherent transmission delay, which can be very considered in satellite communication system is te inherent transmission delay, which can be very critical to the delay-sensitive ATM traffic. Therefore, we propose that the processing delay at the block interleaving stage should be controlled propose a variable-size block interleaving mechanism which utilizes the predicted transmission delay for each traffic in the queues of the transmitter. According to the computer simulation, the proposed mechanism could improve the overall performance by drastically reducing the ATM cell drop rate owing to the excessive transmission delay.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.12 no.4
• /
• pp.619-627
• /
• 1999

다분야 통합 시스템의 설계문제는 다량의 설계변수와 구속조건으로 구성되며 다수의 공학적 현상으로 연관되어 있다. 다분야 통합 최적설계 문제를 효과적으로 다루기 위해서는 다양한 해석분야의 공학적 설계원리를 동시에 고려하여 균형 있고 유기적인 방법으로 최적의 설계를 결정하는 체계적인 설계자동화기술이 요구된다. 다분야 통합 설계문제를 위한 효율적인 설계방법론으로 분리기반 최적화 기법이 적용되는데 이 방법은 한 단위의 대규모 설계문제를 여러 개의 하부시스템으로 분리하여 독립적으로 최적화를 수행하고 각 하부 시스템으로부터의 설계해 사이의 중재 및 통합화를 거쳐 최종적으로 수렴된 최적설계를 찾는 방법이다. 본 논문에서는 분리기반 최적화기법을 다분야 통합최적 설계문제에 적용하는데 필요한 시스템분리기법을 유전알고리즘 및 다층 역전 파 신경회로망을 이용하여 정립하였다. 시스템분리기법을 검증하기 위해 최근 미국 Boeing사에서 개발중인 고속 민간항공기인 HSCT의 시뮬레이션기반 설계문제를 이용하였다. 대규모 설계시스템의 분리결과는 전체 설계문제의 특성을 파악하기 위한 자료로 활용되며 향후, 분리기반 최적화과정에서 최종적으로 통합된 최적설계를 탐색하는데 필요한 기반구조를 제공한다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.5
• /
• pp.1043-1050
• /
• 2018

In this paper, we propose a reinforce learning based on cooperative sensing scheme to select optimal secondary users(SUs) to enhance the detection performance of spectrum sensing in Cognitive radio(CR) networks. The SU with high accuracy is identified based on the similarity between the global sensing result obtained through cooperative sensing and the local sensing result of the SU. A fusion center(FC) uses similarity of SUs as reward value for Q-learning to determine SUs which participate in cooperative sensing with accurate sensing results. The experimental results show that the proposed method improves the detection performance compared to conventional cooperative sensing schemes.

• Journal of KIISE:Information Networking
• /
• v.30 no.1
• /
• pp.82-96
• /
• 2003

The Internet has evolved into the global computer network due to the openness of its protocol, but such evolution brings about new risks and threats. To protect computer networks safely, it is the best way that preventing an attacher from intruding beforehand. However, to provision against all attacks causes the degradation of network performance as well as to prevent unknown attacks is very hard. Secure Combination, the framework which establishes a mutual collaboration and cooperation between the trusted zones, could protect systems from the potential attacks. This frameworks can predict attacks by exchanging security information and cooperating with each zone. It is a dynamic and powerful security architecture that rapidly enables updating security policy and deploying response modules.

• Journal of Communications and Networks
• /
• v.11 no.3
• /
• pp.248-260
• /
• 2009

To connect a mobile ad hoc network (MANET) with an IP network and to carryout communication, ad hoc network node needs to be configured with unique IP adress. Dynamic host configuration protocol (DHCP) server autoconfigure nodes in wired networks. However, this cannot be applied to ad hoc network without introducing some changes in auto configuration mechanism, due to intrinsic properties (i.e., multi-hop, dynamic, and distributed nature) of the network. In this paper, we propose a scalable autoconfiguration scheme for MANETs with hierarchical topology consisting of leader and member nodes, by considering the global Internet connectivity with minimum overhead. In our proposed scheme, a joining node selects one of the pre-configured nodes for its duplicate address detection (DAD) operation. We reduce overhead and make our scheme scalable by eliminating the broadcast of DAD messages in the network. We also propose the group leader election algorithm, which takes into account the resources, density, and position information of a node to select a new leader. Our simulation results show that our proposed scheme is effective to reduce the overhead and is scalable. Also, it is shown that the proposed scheme provides an efficient method to heal the network after partitioning and merging by enhancing the role of bordering nodes in the group.

• The KIPS Transactions:PartC
• /
• v.18C no.6
• /
• pp.405-412
• /
• 2011

Our proposed mobility management scheme is based on Multicasting and HIP(Host Identity Protocol) in PMIPv6(Proxy Mobile IPv6) Networks, and allows users to handoff within and across different administrative domains. The main advantage of our scheme is to enable the inter-domain handoff of both types of nodes with a reduced signalling overhead and packet losses. Specifically, the scheme enables the interworking between host-based and network-based mobility support, by means of the interaction between PMIPv6 with Multicasting and HIP. Performance evaluations demonstrate that our scheme improves the handoff latency and packet losses compared to other global mobility management protocols.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.3880-3899
• /
• 2019

Distributive desynchronization algorithms based on pulse-coupled oscillator (PCO) models have been proposed for achieving collision-free wireless transmissions. These algorithms do not depend on a global clock or infrastructure overheads. Moreover, they gradually converge to fair time-division multiple access (TDMA) scheduling by broadcasting a periodic pulse signal (called a 'firing') and adjusting the next firing time based on firings from other nodes. The time required to achieve constant spacing between phase neighbors is estimated in a closed form or via stochastic modeling. However, because these algorithms cannot guarantee the completion of desynchronization in a short and bounded timeframe, they are not practical. Motivated by the limitations of these methods, we propose a practical solution called PD-DESYNC that provides a short and deterministic convergence time using a flag firing to indicate the beginning of a cycle. We demonstrate that the proposed method guarantees the completion of desynchronization within three cycles, regardless of the number of nodes. Through extensive simulations and experiments, we confirm that PD-DESYNC not only outperforms other algorithms in terms of convergence time but also is a practical solution.