• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권5호
• /
• pp.1604-1617
• /
• 2014

This paper proposes RIX-MAC (Receiver-Initiated X-MAC), a new energy-efficient MAC protocol based on an asynchronous duty cycling. RIX-MAC improves energy efficiency through utilizing short preambles and adopting the receiver-initiated approach, where RIX-MAC minimizes sender nodes' energy consumption by enabling transmitters to predict receiver nodes' wake-up times. It also reduces receiver nodes' energy consumption by decreasing the number of control frames. We use the network simulator to evaluate RIX-MAC's performance. Compared to the prior asynchronous duty cycling approaches of X-MAC and PW-MAC, the proposed protocol shows a remarkable improvement in energy-efficiency and end-to-end delay.

• Journal of Semiconductor Engineering
• /
• 제1권1호
• /
• pp.31-37
• /
• 2020

We present an approximate DRAM architecture for energy-efficient deep learning. Our key premise is that by bounding memory errors to non-critical information, we can significantly reduce DRAM refresh energy without compromising recognition accuracy of deep neural networks. To validate the key premise, we make extensive Monte-Carlo simulations for several well-known convolutional neural networks such as LeNet, ConvNet and AlexNet with the input of MINIST, CIFAR-10, and ImageNet, respectively. We assume that the highest-order 8-bits (in single precision) and 4-bits (in half precision) are protected from retention errors under the proposed architecture and then, randomly inject bit-errors to unprotected bits with various bit-error-rates. Here, recognition accuracies of the above convolutional neural networks are successfully maintained up to the 10^-5-order bit-error-rate. We simulate DRAM energy during inference of the above convolutional neural networks, where the proposed architecture shows the possibility of considerable energy saving up to 10 ~ 37.5% of total DRAM energy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3494-3510
• /
• 2019

Wireless sensor networks encounter energy saving as a major issue as the sensor nodes having no rechargeable batteries and also the resources are limited. Clustering of sensors play a pivotal role in energy saving of the deployed sensor nodes. However, in the cluster based wireless sensor network, the cluster heads tend to consume more energy for additional functions such as reception of data, aggregation and transmission of the received data to the base station. So, careful selection of cluster head and formation of cluster plays vital role in energy conservation and enhancement of lifetime of the wireless sensor networks. This study proposes a new mutation chemical reaction optimization (MCRO) which is an algorithm based energy efficient clustering protocol termed as MCRO-ECP, for wireless sensor networks. The proposed protocol is extensively developed with effective methods such as potential energy function and molecular structure encoding for cluster head selection and cluster formation. While developing potential functions for energy conservation, the following parameters are taken into account: neighbor node distance, base station distance, ratio of energy, intra-cluster distance, and CH node degree to make the MCRO-ECP protocol to be potential energy conserver. The proposed protocol is studied extensively and tested elaborately on NS2.35 Simulator under various senarios like varying the number of sensor nodes and CHs. A comparative study between the simulation results derived from the proposed MCRO-ECP protocol and the results of the already existing protocol, shows that MCRO-ECP protocol produces significantly better results in energy conservation, increase network life time, packets received by the BS and the convergence rate.

• 한국통신학회논문지
• /
• 제40권6호
• /
• pp.1058-1069
• /
• 2015

무선 센서 네트워크에서 위치 기반 라우팅은 위치 정보를 이용하여 데이터 메시지를 전달하기에 효과적인 방법으로 알려져 있다. 위치 기반 라우팅에서 대표적인 그리디 포워딩이 불가능할 경우, 페이스 라우팅이 복구 수단으로 쓰이고 있다. 페이스 라우팅은 교차된 연결들을 제거한 평면 그래프를 바탕으로 구성되며 이는 가장 가까운 이웃 노드들의 연결을 제외한 나머지 연결들은 모두 제거됨을 의미한다. 하지만 페이스 라우팅은 무선 반경을 고려하지 않아 한 홉 범위 안에서 멀리 갈 수 있는 링크들이 존재하더라도 평면 그래프의 규칙에 의해 짧은 링크들을 선택한다. 이것은 밀집된 지역에서 페이스 라우팅을 쓰일 경우 에너지 효율성을 저하시킨다. 본 논문에서는 평면 그래프에서 제거된 링크들을 찾고 이용할 수 있는 프로토콜을 제안한다. 본 프로토콜은 작은 크기의 컨트롤 메시지와 함께 무선 반경에서 제거된 링크를 찾는다. 시뮬레이션을 통해 제안 방안이 이전 페이스 라우팅 프로토콜보다 에너지 효율성이 높음을 보인다.

• 한국세라믹학회지
• /
• 제51권3호
• /
• pp.145-149
• /
• 2014

A hydroxyapatite microsphere was prepared using a spray-drying method. The change in the shape as a function of the slurry concentration and the change in the degree of shrinkage according to the heat-treatment temperatures were observed. To obtain biomaterials with improved bio-stability, $CaHPO_4{\cdot}2H_2O$ and $Ca(OH)_2$ were mixed at a ratio of 6 : 4 and then ball-milled to synthesize hydroxyapatite. The hydroxyapatite microsphere was prepared using 30 wt% ~ 80 wt% hydroxyapatite slurry by a spray-drying method. For concentrations lower than 50 wt% or higher than 80 wt%, doughnut-shaped microspheres were produced. However, perfect microspheres were produced when using slurry concentrations of 50 wt% ~ 70 wt%. A dense microstructure was observed after a heat treatment at temperatures higher than $1100^{\circ}C$ and the size was reduced by 24.3% at these temperatures.

• 한국컴퓨터정보학회논문지
• /
• 제15권8호
• /
• pp.149-156
• /
• 2010

센서네트워크에 사용되는 최신의 센서노드 기술은 많은 데이터 저장능력과 빠른 처리능력을 제공함으로써 다양한 응용분야에서 좀 더 효율적인 네트워크 환경을 구성할 수 있게 되었다. 그리고 네트워크 구축 환경도 트리 기반에서 클러스터 기반으로 변환 되었다. 그러나 기존 트리 기반의 센서네트워크에서 사용되던 질의처리 방법은 클러스터 기반의 네트워크에서는 효율성이 떨어진다. 따라서 클러스터 기반 네트워크에서 효율적으로 질의를 처리하는 새로운 방법이 필요하다. 그래서 본 논문에서는 클러스터 기반의 센서네트워크에서 데이터 속성 분류를 통한 필터링을 이용하여 에너지를 효율적으로 사용하는 질의 처리 기법을 제안한다. 제안 기법은 클러스터 기반의 네트워크 장점을 최대한 활용하여 질의 처리에 드는 에너지를 줄이고 좀 더 지능적으로 질의를 분배하도록 설계 하였다. 그리고 MATLab을 이용하여 제안된 기법이 에너지 효율성 측면에서 우수함을 입증하였다 본 논문에서는 고성능 컴퓨팅 시스템의 성능 향상을 위한 효율적인 동적 작업부하 균등화 정책을 제안한다.

• 한국전기전자재료학회논문지
• /
• 제32권6호
• /
• pp.522-527
• /
• 2019

In order to fabricate high-quality SiC substrates for power electronic devices, various single crystal growing methods were prepared. These include the physical vapor transport (PVT) and top seeded solution growth (TSSG) methods. All the suggested SiC growth methods generally use induction-heating furnaces. The temperature distribution in this system can be easily adjusted by changing the hot-zone design. Moreover, precise temperature control in the induction-heating furnace is favorably required to grow a high-quality crystal. Therefore, in this study, we analyzed the heat transfer in these furnaces to grow SiC crystals. As the growth temperature of SiC crystals is very high, we evaluated the effect of radiation heat transfer on the temperature distribution in induction-heating furnaces. Based on our simulation results, a heat transfer strategy that controls the radiation heat transfer was suggested to obtain the optimal temperature distribution in the PVT and TSSG methods.

• Nuclear Engineering and Technology
• /
• 제51권5호
• /
• pp.1181-1194
• /
• 2019

A new moving-mesh Finite Volume Method (FVM) for the efficient solution of the two-dimensional neutron diffusion equation is introduced. Many other moving-mesh methods developed to solve the neutron diffusion problems use a relatively large number of sophisticated mathematical equations, and so suffer from a significant complexity of mathematical calculations. In this study, the proposed method is formulated based on simple mathematical algebraic equations that enable an efficient mesh movement and CV deformation for using in practical nuclear reactor applications. Accordingly, a computational framework relying on a new moving-mesh FVM is introduced to efficiently distribute the meshes and deform the CVs in regions with high gradient variations of reactor power. These regions of interest are very important in the neutronic assessment of the nuclear reactors and accordingly, a higher accuracy of the power densities is required to be obtained. The accuracy, execution time and finally visual comparison of the proposed method comprehensively investigated and discussed for three different benchmark problems. The results all indicated a higher accuracy of the proposed method in comparison with the conventional fixed-mesh FVM.

• 한국주거학회:학술대회논문집
• /
• 한국주거학회 2005년도 추계학술대회 논문집
• /
• pp.319-322
• /
• 2005

An energy problem has become one of the many Problems awaiting solution to human society. Energy needs to support not only industry, but also building in city. The types of residential building, office building and public building account for much proportion among the totally amount used energy. A public institution shows a fine example for an energy conservation in the present situation. But, the amount used energy of electricity, air-conditioning and heating is different from a building use and condition. Because the building data of the energy consumption by a load is insufficient, it is unable to select a proper energy source. According to this study, it analyzed an energy load by each building that had investigated the actual conditions of energy consumption about public buildings in Daegu City. In order to plan the efficient energy use both existing buildings and new buildings, this study will present the efficient energy use plan, which has operated to new buildings of a public institution, at present.

• 대한설비공학회:학술대회논문집
• /
• 대한설비공학회 2006년도 하계학술발표대회 논문집
• /
• pp.959-964
• /
• 2006

Recently, an international oil price shows fluctuation in 70 dollars per barrel and it is said that this trend is able to continues for much longer. Because energy consumption in buildings accounted for about 25% of the nation's total demand for energy, Energy censervation and efficiency of buildings were very important issues. Main political measures relating to building energy use and saving were 'the system of accrediting green building', 'building energy certification system', 'energy consumption efficiency rating indication system', 'building energy code', 'high efficient energy materials certification system'. This paper presents approaches to improve building energy efficiency policy.