Earticle

TinyML 기술은 소형화된 디바이스에서도 머신러닝 모델을 실행할 수 있도록 설계된 기술로, 소형 마이크로컨트롤 러에서 효율적인 데이터 처리를 가능하게 한다. 본 논문에서는 TinyML의 이론적 배경과 주요 기술을 살펴보고, 애플리케이션에서 활용 가능한 다양한 서비스 사례를 제시한다. 특히 스마트 서비스, 예측을 기반으로 한 유지보수, 데이터 분석 등 실질적인 응용 사례들을 통해 TinyML의 활용 가능성과 잠재력에 대하여 논의하며, 현재 제공 중인 서비스와 개발 중인 기술들의 활용도를 분석한다. 또한, TinyML 기술이 앞으로 우리의 삶에 미칠 긍정적인 영향을 고찰하고, 기술 발전을 통해 가능한 연구의 방향성을 제시한다. 결론적으로 TinyML 기술은 다양한 디바이스 에서 머신러닝의 가능성을 크게 확장시키며, IoT와 스마트 시스템의 핵심 기술로 자리 잡을 것으로 기대된다.

TinyML technology is designed to enable the execution of machine learning models even on miniaturized devices, facilitating efficient data processing on small microcontrollers. This paper explores the theoretical background and key technologies of TinyML, presenting a variety of services that can be utilized in applications. In particular, practical use cases such as smart services, maintenance based on prediction, and data analysis are discussed to highlight the potential and applicability of TinyML. The paper also examines the utilization of currently available services and technologies under development. Furthermore, it reflects on the positive impact that TinyML technology may have on our lives and suggests directions for future research enabled by advancements in the technology. In conclusion, TinyML technology significantly expands the potential of machine learning across diverse devices and is expected to become a cornerstone of IoT and smart systems.

본 논문에서는 수중 펌프에 대한 동작 상태 모니터링, 고장 진단 및 예측 알고리즘에 대하여 연구하였다. 모니터링 유닛은 수중펌프의 센서 값을 모니터링 제어유닛으로 485통신을 통해서 전송되며, 모니터링 제어 유닛은 센서 값을 변환하여 서버 컴퓨터와 임베디드시스템에 송신한다. 서버 및 임베디드시스템은 데이터베이스에 센서 값을 저장하고 화면에 디스플레이 하여 수중 펌프의 동작 상태를 확인 할 수 있다. 임베디드시스템에서는 데이터 베이스에 저장된 센서 값을 읽어 들여, 개별 및 복합진단을 수행하고, 서버 및 모니터링 제어 유닛에게 송신하여 결과를 표시할 수 있게 하였다. 수중펌프의 예측진단은 23개의 개별 센서의 분류별로 나눠서 다층페셉트론을 구현 하였고, 학습 및 테스트를 통해서 가중치를 설정하였으며, 실시간 입력되는 센서 값을 통하여 알고리즘의 성능을 확인 할 수 있었다. 제안된 알고리즘은 수중 펌프 제품 경쟁력에 도움이 될 것으로 판단된다.

In this paper, operation status monitoring, fault diagnosis, and prediction algorithms for submersible pumps were studied. The monitoring unit transmits the sensor value of the submersible pump to the monitoring control unit through 485 communication, and the monitoring control unit converts the sensor value and transmits it to the server computer and embedded system. The server and embedded system store the sensor value in the database and display it on the screen to check the operation status of the submersible pump. In the embedded system, it is possible to read the sensor values stored in the database, perform individual and complex diagnosis, and display the results by sending them to the server and monitoring control unit. . For predictive diagnosis, the multi-layered peceptron was implemented by dividing by classification of 23 individual sensors, and the performance of the algorithm was confirmed through learning and testing. The proposed algorithm is judged to be helpful in the competitiveness of submersible pump products.

최근 모니터링 및 예측 시스템을 이용하여 사전에 결함을 발견하고 이를 경고하는 시스템이 활발히 연구되고 있다. 차량 안전 관리에 있어서도 예측 결함 분석 기술을 적용하여 자동차 휠 베어링의 고장 유무 및 고장 유형을 조기 에 경고하는 시스템이 필요하다. 본 논문에서는 휠 베어링과 결합 된 센서 모듈과 각 센서 모듈에서 차량 가속 정보 및 진동 정보를 수집, 저장 및 분석하는 진단 시스템을 제시하였다. 제안된 센서 모듈은 저비용으로 차량의 휠 베어링 상태를 모니터링하며, 이렇게 수집된 데이터를 활용하여 진단 및 고장 예측 기능을 수행하는 방안을 연구하였다. 개발된 센서 모듈과 예측 분석 시스템은 가진 테스트 장비 및 실제 차량을 이용하여 테스트하고 그 유효성을 평가하였다.

There is a need for a system that provides early warning of presence and type of failure of automobile wheel bearings through the application of predictive fault analysis technologies. In this paper, we presented a sensor module mounted on a wheel bearing and a diagnostic system that collects, stores and analyzes vehicle acceleration information and vibration information from the sensor module. The developed sensor module and predictive analysis system was tested and evaluated thorough excitation test equipment and real automotive vehicle to prove the effectiveness.

스마트키는 다양한 임베디드 환경에서 활용하고 있지만 사용자의 위치에서 신호의 증폭을 통한 원격지 공격이 발생하고 있음을 알 수 있다. 방어 기법에 대한 기존 연구는 다수의 센서를 사용하거나 인증 속도의 개선을 위한 해시 함수를 사용한 경우가 있는데 이는 전력 소모를 증가시키거나 1종 오류가 발생할 가능성을 가지고 있다. 이에 본 논문에 서는 컨트롤러와 호스트 장치간의 인증 방식을 개선하여 사용자의 이동 정보와 클라이언트 인증 기반의 임베디드 보안 컨트롤러 모델을 제안하고자 한다. 제안하는 모델에 대하여 아두이노 보드와 GPS 및 블루투스를 이용한 통신을 위하여 암호화 알고리즘을 적용하였으며 인증을 위하여 사용자의 이동 정보를 사용하여 경로 분석을 통해 인증을 수행하였다. 그리고 제안하는 모델을 사용하여 동작 수행을 하더라도 작동 편이성에 큰 영향을 미치지 않았음을 암호화 및 복호화 시간 측정을 통하여 확인하였다. 제안하는 모델의 임베디드 보안 컨트롤러는 이륜차와 같은 리모트 컨트롤러와 이동 또는 고정형 호스트 장치를 가지고 있는 시스템 구조에서 적용할 수 있으며 연구 과정에 암호화 및 복호화 시간이 각각 100ms 이내에 처리를 수행할 수 있음을 확인하였으며 향후 경로 데이터 관리 방법에 대한 추가연구 및 암호화 및 복호 화 소요 시간과 데이터 통신 시간을 줄일수 있는 프로토콜에 대한 연구가 더 필요할 것으로 판단된다.

A smart key has been used in a variety of embedded environments and there also have been attacks from a remote place by amplifying signals at a location of a user. Existing studies on defence techniques suggest multiple sensors and hash functions to improve authentication speed; these, however, increase the electricity usage and the probability of type 1 error. For these reasons, I suggest an embedded security controller based on client authentication and user movement information improving the authentication method between a controller and a host device. I applied encryption algorithm to the suggested model for communication using an Arduino board, GPS, and Bluetooth and performed authentication through path analysis utilizing user movement information for the authentication. I found that the change in usability was nonsignificant when performing actions using the suggested model by evaluating the time to encode and decode. The embedded security controller in the model can be applied to the system of a remote controller for a two-wheeled vehicle or a mobile and stationary host device; in the process of studying, I found that encryption and decryption could take less then 100ms. The later study may deal with protocols to speed up the data communication including encryption and decryption and the path data management.

Most systems used in industrial sites, actually have non-linearity and uncertainty. Therefore there are a lot of difficulties in evaluating conditions of these systems. Generally, the quantitative analysis and expression are found hard because the general public cannot easily make an accurate interpretation on the systems. Thus development of a system that utilizes an expertise from skilled analysts is required. In this research, a real-time sensor signal conditioning system and Fuzzy-expert system have been separately set up into an inference algorithm. So that it ensures a fast, accurate, objective and quantitative operational condition value provided to the manager. Therefore, FE_AFCDM is suggested in this literature, as an effective system for diagnosing the problems related to the air compressor. It can quantify the uncertain and absurd condition to operate the air compressor facilities safely and financially.

The moisture content of crude oil notably affects various aspects of oil production, storage, transportation, and exploration. However, accurately measuring this moisture content is challenging because of numerous influencing factors, leading to a lack of precision in existing detection methods. This inadequacy hinders the progress of China's petroleum industry. To overcome these challenges, this paper proposes a conductivity-based method for measuring crude oil moisture content. By employing an embedded system, we designed a sensor comprising five electrodes. Additionally, we developed signal excitation and signal processing circuits. Moreover, a software program was designed to analyze and compute the output signal using fast Fourier transform operations. This facilitated the identification of flow patterns, computation of relevant flow rates, and establishment of correlation rates based on frequency spectral characteristics. Based on experimental results, we established a functional relationship between measurement parameters and crude oil moisture content. This study enhanced the precision of moisture content measurement, thereby addressing existing limitations and fostering the advancement of China's petroleum industry.

이동 차량에서도 멀티미디어 응용 서비스에 대한 요구가 급증함에 따라, MOST(Media Oriented Systems Transport) 프로토콜이 주목을 받게 되었다. MOST 프로토콜은 짧은 제어 메시지, 비동기 패킷, 그리고 예약형 동기 스트림 데이터와 같은 세가지 형태의 통신 모드를 지원한다. MOST는 다양한 전송모드를 지원하여 차량 내부의 네트워크에 응용하는 프로토콜로서 적합하다. 본 논문에서는 MOST 프로토콜을 장착한 AMS(Automotive Media Server) 플랫폼 기반의 임베디드 시스템을 구현하고 제어 포트와 소스포트의 동기 채널을 통한 네트워크 통신테스트를 동해 시스템을 검증한다. MOST POF(Plastic Optical Fiber) 네트워크 상에서 서로 멀티미디어 통신을 할 수 있는 프로토타입 임베디드 플랫폼을 구현하여 시스템의 동작 테스트를 위해 DivX 디코더를 연결하고 비디오 스트림과 제어 메시지를 MOST 네트워크를 통해 전송하여 동작을 확인하였다.

For growing need for the multimedia application in the vehicles, the MOST protocol has been focused on. The MOST protocol supports three kinds of communication modes; short control message, asynchronous packets, and reserved synchronous stream data. Because of a variety of transportation, the MOST is suitable for various applications in vehicle environment. In this paper, we implemented embedded system which is MOST-enabled AMS platform and tested the network communication operation through the control port and the synchronous channel of the source port. We implemented the prototype platforms which communicate each other on the MOST's POF network. Moreover we implemented the DivX decoder attached AMS platform and verified the operation by transferring the video stream and the control messages through the MOST network.

This paper proposes a frame work for embedded system management. The management system can automatically monitor and maintain the Internet connected embedded-systems such as sensor devices, set-top box, web-pad, information appliances, PDA, etc. Users can easily diagnose system state, resolve system problems, maintain and update applications, and back up user information using the management system. We implement prototype for the management system on embedded Linux system and High -available Linux system.

A large portion of the embedded system development process involves the integration of hardware and software. Unfortunately, communication across the hardware/software boundary is tedious and error-prone to create. This paper presents an automatic hardware/software interface generation system. As the front-end of hardware/software co-design frameworks, a system designer defines XML specifications for hardware functions. Our system generates hardware/software interfaces including Device Driver, Driver API, and Device Controller from these specifications. Embedded software designers can easily use hardware just like system libraries. Our system reduces the mistakes and errors that can be occurred when a software programmer directly connects software to hardware, and supports balancing labors between hardware developers and software programmers. Moreover, this system can be used as the back-end for a hardware/software co-design framework.

We have designed an embedded controller for the control of a diesel generator using an embedded system. The generator is monitored and controlled remotely via the internet in real-time. The proposed digital controller is designed to handle precisely the distortions and noises of the signals emanating from the diesel generator, and enables abnormal operation of the diesel generator to be notified to the remote manager using the Short Message Service (SMS) of the Internet, which enables the appropriate personnel to take action by remote control according to the incoming messages.

Vehicle detector system based on image processing technology is a significant domain of ITS (Intelligent Transportation System) applications due to its advantages such as low installation cost and it does not obstruct traffic during the installation of vehicle detection systems on the road[1]. In this paper, we propose architecture for vehicle detection by using image processing. The architecture consists of two main parts such as an image processing part, using high speed FPGA, decision and calculation part using CPU. The CPU part takes care of total system control and synthetic decision of vehicle detection. The FPGA part assumes charge of input and output image using video encoder and decoder, image classification and image memory control.

The paper presents a novel embedded multifunctional media sever (EMMS) for mobile devices to receive various media programs. Being different from other contemporary system research, the paper mainly studies how to design a reception solution for terrestrial digital television (DTV) on mobile devices and how to enable mobile devices can receive DTV program, enjoy video-on-demand (VOD), achieve video surveillance and relay Internet video program via local Wi-Fi simultaneously. In the system design, we integrate broadcasting-terrestrial DTV tuner, streaming media re-transmission system, VOD disk, video camera and access interface to the Internet into EMMS, which can either receive terrestrial DTV radio signals and demodulate out digital transport stream (TS), or can read streaming media bit-stream from VOD disk, surveillance camera or access interface to the Internet. The experimental results show the proposed system is stable and quality-efficient. Comparing with the other systems, the proposed system has the least packet loss rate and response time.

IT security companies have been releasing file system filter driver security solutions based on the whitelist, which are being used by several enterprises in the relevant industries. However, in February 2019, a whitelist vulnerability was discovered in Microsoft Edge browser, which allows malicious code to be executed unknown to users. If a hacker had inserted a program that executed malicious code into the whitelist, it would have resulted in considerable damage. File system filter driver security solutions based on the whitelist are discretionary access control (DAC) models. Hence, the whitelist is vulnerable because it only considers the target subject to be accessed, without taking into account the access rights of the file target object. In this study, we propose an industrial device security system for Windows to address this vulnerability, which improves the security of the security policy by determining not only the access rights of the subject but also those of the object through the application of the mandatory access control (MAC) policy in the Windows industrial operating system. The access control method does not base the security policy on the whitelist; instead, by investigating the setting of the security policy not only for the subject but also the object, we propose a method that provides improved stability, compared to the conventional whitelist method.

본 논문에서는 실제 값과 같은 데이터의 불확실성과 유사성을 측정 할 수 있는 퍼지 엔트로피와 유사성 측정이 소개되고 있다. 퍼지 엔트로피와 유사성 측정의 디자인이 설명하고 입증했다. 획득 수단은 연산 프로세스에 적용되고 논의되었다. 이러한 의사 결정과 퍼지 게임 이론과 같은 데이터 정량화 결과의 연장도 논의되었다.

In this paper, fuzzy entropy and similarity measure to measure the uncertainty and similarity of data as real value were introduced. Design of fuzzy entropy and similarity measure were illustrated and proved. Obtained measures were applied to the calculating process and discussed. Extension of data quantification results such as decision making and fuzzy game theory were also discussed.

We present a device model for organic light emitting diodes(OLEDs) which includes charge injection, transport, recombination, and space charge effects in the organic materials. The model can describe both injection limited and space charge limited current flow and the transition between them. Calculated device current, light output, and quantum and power efficiency are presented for different cases of material and device parameters and demonstrate the improvements in device performance in bilayer devices. These results are interpreted using the calculated spatial variation of the electric field, charge density and recombination rate density in the device. We find that efficient OLEDs are possible for a proper choice of organic materials and contact parameters.

We propose a novel programmable miniature vision module based on a custom designed analog VLSI (aVLSI) chip. The vision module consists of the optical flow vision sensor embedded with commercial off-the-shelves digital hardware; in our case is the Intel XScale PXA270 processor enforced with a programmable gate array device. The aVLSI sensor provides gray-scale imager data as well as smooth optical flow estimates, thus each pixel gives a triplet of information that can be continuously read out as three independent images. The particular computational architecture of the custom designed sensor, which is fully parallel and also analog, allows for efficient real-time estimations of the smooth optical flow. The Intel XScale PXA270 controls the sensor read-out and furthermore allows, together with the programmable gate array, for additional higher level processing of the intensity image and optical flow data. It also provides the necessary standard interface such that the module can be easily programmed and integrated into different vision systems, or even form a complete stand-alone vision system itself. The low power consumption, small size and flexible interface of the proposed vision module suggests that it could be particularly well suited as a vision system in an autonomous robotics platform and especially well suited for educational projects in the robotic sciences.