• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.615-619
• /
• 2021

This study focuses on memory systems that are optimized to increase performance and energy efficiency in many embedded systems such as IoT, cloud computing, and edge computing, and proposes a performance improvement technique. The proposed technique improves memory system performance based on machine learning algorithms that are widely used in many applications. The machine learning technique can be used for various applications through supervised learning, and can be applied to a data classification task used in improving memory system performance. Data classification based on highly accurate machine learning techniques enables data to be appropriately arranged according to data usage patterns, thereby improving overall system performance.

• Journal of Sensor Science and Technology
• /
• v.31 no.5
• /
• pp.301-306
• /
• 2022

With the recent widespread implementation of Internet of Things (IoT) technology driven by Industry 4.0, self-powered sensors for wearable and implantable systems are increasingly gaining attention. Piezoelectric nanogenerators (PENGs) and triboelectric nanogenerators (TENGs), which convert biomechanical energy into electrical energy, can be considered as efficient self-powered sensor platforms. These are energy harvesters that are used as low-power energy sources. However, they can also be used as sensors when an output signal is used to sense any mechanical stimuli. For sensors, collecting high-quality data is important. However, the accuracy of sensing for practical applications is equally important. This paper provides a brief review of the performance advanced by the materials and structures of the latest PENG/TENG-based wearable sensors and intelligent applications applied using artificial intelligence (AI)

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.165-172
• /
• 2023

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.339-341
• /
• 2023

This paper presents a multithread and efficient cryptographic hardware access (MECHA) for efficient and fast cryptographic operations that eliminates the need for context switching. Utilizing a UNIX domain socket, MECHA manages multiple requests from multiple applications simultaneously, resulting in faster processing and improved efficiency. We comprise several key components, including the Server thread, Client thread, Transceiver thread, and a pair of Sender and Receiver queues. MECHA design is portable and can be used with any communication protocol, with experimental results demonstrating a 83% increase in the speed of concurrent cryptographic requests compared to conventional interface design. MECHA architecture has significant potential in the field of secure communication applications ranging from cloud computing to the IoT, offering a faster and more efficient solution for managing multiple cryptographic operation requests concurrently.

• Electronics and Telecommunications Trends
• /
• v.29 no.4
• /
• pp.110-122
• /
• 2014

무선환경은 가짜 클론 디바이스가 진짜인 것처럼 위장한 해킹공격에 매우 취약한 것으로 잘 알려져 있다. 그것은 단말기와 기지국(AP: Access Point)이 위변조가 쉬운 디바이스 식별자(예로 MAC(Medium Access Control) 주소, SSID, BSSID(Basic Service Set Identification) 등)를 이용하여 상호 인증하기 때문이다. 무선핑거프린팅(Wireless Fingerprinting)은 통신과정에서 발생되는 무선신호 특성으로부터 디바이스를 고유하게 식별하는 핑거프린트를 추출하여 송신 디바이스가 가짜 클론 디바이스인지 아닌지 여부를 식별하는 기술이다. 본 기술은 무선물리계층 보안을 위한 인증 및 키 생성, 무선 침입탐지, 공격자의 위치/방향/거리 추적, 무선 포랜식 및 보안관제의 성능을 결정하는 핵심기술로 활용되고 있다. 향후 등장이 예상되는 M2M 무선랜, 무선인지네트워크, 무선센서, 무선차량통신, IoT 무선통신환경에서도 본 기술의 중요성은 더욱 증가하리라 본다. 본고에서는 무선 디바이스의 핑거프린팅 개념을 이해하고, 기술 분류에 따른 세부기법 연구 및 보안응용 동향을 분석함으로써 본 기술의 발전방향을 조망해보고자 한다.

• Journal of Korea Multimedia Society
• /
• v.23 no.3
• /
• pp.476-482
• /
• 2020

In this paper, we propose a block-based modularity architecture design method for distributed machine learning. The proposed architecture is a block-type module structure with various machine learning algorithms. It allows free expansion between block-type modules and allows multiple machine learning algorithms to be organically interlocked according to the situation. The architecture enables open data communication using the metadata query protocol. Also, the architecture makes it easy to implement an application service combining various edge computing devices by designing a communication method suitable for surrounding applications. To confirm the interlocking between the proposed block-type modules, we implemented a hardware-based modularity application system.

• Annual Conference of KIPS
• /
• 2017.11a
• /
• pp.462-463
• /
• 2017

Although the mirror has been used for a long time primarily for personal purposes, such as adornment by reflecting feature of mirror, it can also be used to expand from its legacy function to safety and lots of unpredictable applications. However, the mirror can be used as an information or investigation base, it is subject to human beings in the private region, objects and private purpose in the public and industrial fields. This paper suggests the wide usage of mirror equipped with smart technology, considering current Korea domestic regulation by law.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1393-1405
• /
• 2017

It is well known that duty-cycling control by dynamically adjusting the polling interval according to the traffic loads can effectively achieve power saving in wireless sensor networks. Thus, there has been a significant research effort in developing polling interval adaptation schemes. Especially, Dynamic Low Power Listening (DLPL) scheme is one of the most widely adopted open-looping polling interval adaptation techniques in wireless sensor networks. In DLPL scheme, if consecutive idle (busy) samplings reach a given fixed threshold, the polling interval is increased (decreased). However, due to the trial-and-error based approach, it may significantly deteriorate the system performance depending on given threshold parameters. In this paper, we propose a novel DLPL scheme, called SDL (Sequential hypothesis testing based Dynamic LPL), which employs sequential hypothesis testing to decide whether to change the polling interval conforming to various traffic conditions. Simulation results show that SDL achieves substantial power saving over state-of-the-art DLPL schemes.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.11 no.5
• /
• pp.287-298
• /
• 2016

This paper proposes an effective smart plug system capable of monitoring power, cutting off standby power and overload current. The key design concept is focused on measurement accuracy, self power consumption and controlling via smart phone application. The system is composed of several plugs and a hub, and adopts a star-topology-styled TDMA wireless protocol for communication between plug and hub. The test result shows that the implemented smart plug system meets K10026 regulation and is worth in electrical safety, energy saving, easy living.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.3
• /
• pp.1144-1158
• /
• 2016

In this paper, delay-constrained data transmission is considered over error-prone networks. Network coding is deployed for efficient information exchange, and an approximate decoding approach is deployed to overcome potential all-or-nothing problems. Our focus is on determining the cluster size and its impact on approximate decoding performance. Decoding performance is quantified, and we show that performance is determined only by the number of packets. Moreover, the fundamental tradeoff between approximate decoding performance and data transfer rate improvement is analyzed; as the cluster size increases, the data transfer rate improves and decoding performance is degraded. This tradeoff can lead to an optimal cluster size of network coding-based networks that achieves the target decoding performance of applications. A set of experiment results confirms the analysis.