김영근 교수

Federated learning enables a cluster of decentralized mobile devices at the edge to collaboratively train a shared machine learning model, while keeping all the raw training samples on device. This decentralized training approach is demonstrated as a practical solution to mitigate the risk of privacy leakage. However, enabling efficient FL deployment at the edge is challenging because of non-IID training data distribution, wide system heterogeneity and stochastic-varying runtime effects in the f

Deep learning inference is increasingly run at the edge. As the programming and system stack support becomes mature, it enables acceleration opportunities in a mobile system, where the system performance envelope is scaled up with a plethora of programmable co-processors. Thus, intelligent services designed for mobile users can choose between running inference on the CPU or any of the co-processors in the mobile system, and exploiting connected systems such as the cloud or a nearby, locally conn

Off-the-shelf embedded systems have adopted heterogeneous multi-core processors which have high-performance big cores and low-power small cores. Though there are two different types of cores in heterogeneous multi-core processors, conventional DVFS (Dynamic Voltage and Frequency Scaling)-based DTM (Dynamic Thermal Management) techniques do not utilize the different types of cores to cool down hot cores. Rather, they primarily reduce the voltage and frequency of the hot cores, leading to performa

Recent smart devices have adopted heterogeneous multi-core processors which have high-performance big cores and low-power small cores. Unfortunately, the conventional task scheduler for heterogeneous multi-core processors does not provide appropriate amount of CPU resources for multimedia applications (whose QoS is important to users), resulting in energy waste; it often executes multimedia applications and non-multimedia applications on the same core. In this paper, we propose an advanced task

To improve mobile experience of users, recent mobile devices have adopted powerful processing units (CPUs and GPUs). Unfortunately, the processing units often consume a considerable amount of energy, which in turn shortens battery life of mobile devices. For energy reduction of the processing units, mobile devices adopt energy management techniques based on software, especially OS (Operating Systems), as well as hardware. In this survey paper, we summarize recent OS-level energy management techn

As GPU power consumption in smartphones increases with more advanced graphic performance, it becomes essential to estimate GPU power consumption accurately. The conventional GPU power model assumes, simply, that a GPU consumes constant power when turned on; however, this is no longer true for recent smartphone GPUs. In this paper, we propose an accurate GPU power model for smartphones, considering newly adopted dynamic voltage and frequency scaling. For the proposed GPU power model, our evaluati

To prolong battery life of mobile devices, image processing applications often exploit offloading techniques which run some or all of the computations on remote servers. Unfortunately, the existing offloading techniques do not consider the fact that data transmission time and energy consumption of wireless network interfaces exponentially increase when signal strength decreases. In this paper, we propose an adaptive offloading for image processing applications, which considers wireless signal st

본 논문의 연구목적은 ‘3.11 코로나19팬데믹’ 감염병 재해 이후 글로벌 정치사회의 변용에 대해 개관하고, 한국과 일본이 어떻게 대응하고 관리하고 있는지 즉 거버넌스 및 위기대응에 관해 비교 분석하는 데 있다. 특히, 국가 혹은 비정부 행위자들의 감염병 대비 및 대응체제의 차이점과 공통점을 고찰하고 있다. 감염병 재해가 초래하는 학문분과 및 그 범주는 정치, 경제, 인문사회, 문화, 사상, 스포츠, 지구환경, 과학기술, 공동체, 언어적 측면 등이 융복합적으로 상호작용하며 맞물려 있다. 한국과 일본의 감염병 재해 거버넌스를 비교해보면, 의료 및 과학기술만으로 해결할 수 없다는 점을 알 수 있다. 즉 과학기술과 인문‧사회‧예술의 융합을 통한 접근 및 대응책이 매우 중요하다고 할 수 있다. 이는 코로나19 감염병 재해에 대응하기 위해서는 사회적 재해 및 사상이나 이념의 변화를 아우르는 ‘인문적 재해’ 대응도 포함하는 학제적 접근이 절실했다는 것을 나타내는 것이다. 결론적으로, 한국은

Recently, mobile devices have employed heterogeneous multi-core processors which consist of highperformance big cores and low-power small cores. In heterogeneous mobile multi-core processors, the conventional DVFS (Dynamic Voltage and Frequency Scaling)-based DTM (Dynamic Thermal Management) is still adopted; it does not actively utilize the small cores to resolve thermal problem. In this paper, we propose M-DTM (Migration-based DTM) for heterogeneous mobile multi-core processors. In case of the

3.11東日本大震災後、日本経済(政策)の変化や、世界で最もダイナミックなアジア太平洋地域におけるTPP(環太平洋経済連携協定)交渉に関する関心が高まっている。特に、北東アジアにおけるTPPや自由貿易圏構想を通じて経済統合の新たな制度的基盤を整備することができるのかが、大きな関心を引いているのは間違いない。それにもかかわらず、先行研究では北東アジアにおける二国間交渉における阻害要因分析やFTA締結による経済的な要因分析など、断片的な研究が一般的である。しかも、TPP政策を中心にした3.11東日本大震災後の日本経済と北東アジア経済協力に関する先行研究は殆ど見当たらないのが現状である。そこで、本論文は日本の通商(経済)政策をはじめ、北東アジア経済協力の現状とTPP制度、特に日本のTPP構想についても視野に入れ、北東アジアにおける経済協力の進路を総合的に再検討することを目的とする。具体的な分析内容は次の通りである。第Ⅱ章では、日本の通商政策の変化とFTA構想を分析し、第Ⅲ章では、日本のTPP構想と国内外の政策を分析している。特に、北東アジア経済協力の現状と日本のTPP構想(日本がTPPを通し

본고의 목적은 일본 재해연구(진재학)에 관한 이해를 통해 한국에 있어서 재해연구에 대한 학문적 어프로치 기반을 구축하기 위하여 재해 후 일본 재해거버넌스의 변용을 본석하는 이론적인 틀을 제공하는 데 있다. 대지진(대재해) 발생 후 일본의 경우 자연재해 뿐 아니라 국민의 사상 (인문학적인 재해)와 사회구조(사회적인 재해), 나아가 경제구조와 경제정책에도 커다란 변화의 목소리가 높아진다. 지진 및 그 이후 사회 문화 역사적인 변화에 대해 연구하는 진재학도 피해를 입은 지역과 정부, NGO등 다각적인 시점에서 고려되어야 하며, 더구나 지진재해와 쓰나미재해, 원전 사고 등에 의한 방사능 재해로 등 복한적인 연쇄위험을 시야에 넣은 연구가 필요한 시대가 되었다. 진재학은 의제(아젠다)와 분석 레벨의 다양하가 진행되고 있다. 미증유(예상외)의 대재해를 초래한 동일본대지진은 복합연쇄위기라고도 불리워 지는 융합적인 리스크였다. 일본의 문제로만 치부하는 데 그치지 않고 근래 국제적인 관심을 모아 재해

We analyze the transmission performance of vestigial sideband (VSB) signals with electrical dispersion compensator (EDC) by numerical simulations. Optimizing the dispersion and bandwidth of EDC, and the extinction ratio and chirp of the modulator, we can improve the dispersion tolerance of VSB signals. VSB signals with negative chirp have better transmission performance while VSB signals with positive chirp are advantageous to the implementation of EDC due to low optimum dispersion in EDC. VSB t

To prolong battery life of mobile devices, applications often exploit offloading techniques which run computations on remote servers. Unfortunately, the existing offloading techniques do not consider the fact that data transmission time and energy consumption of wireless network interfaces exponentially increase when signal strength decreases. In this paper, we propose an adaptive offloading technique that considers signal strength. Our technique estimates gain (reduced computation time and ener