• 대한임베디드공학회논문지
• /
• 제4권1호
• /
• pp.35-41
• /
• 2009

The H.264/SVC enables network-adaptive video transmission to smart device which uses wireless network. But, quality scalability of H.264/SVC does not consider personal subjective image quality. In addition, its network efficiency also does not optimized because it uses MGS(Medium Grained Scalability) and CGS(Coarse Grained Scalability). Thus, this paper proposed a new scalable ROI algorithm for not only subjective image quality improvement but also network adaptation. To experiment our proposed a scheme, we added designed algorithm to JSVM(Joint Scalable Video Model) open source video codec of H.264/SVC. Experiment was performed according to the pre-defined scenario for simulating various network conditions. Finally, experimental result showed our proposed scalable ROI scheme. It is better than traditional non-selective scheme in subjective video quality.

• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• 대한임베디드공학회논문지
• /
• 제16권3호
• /
• pp.113-117
• /
• 2021

The data gathering delay and the network lifetime are important indicators to measure the service quality of wireless sensor and actuator networks (WSANs). This study proposes a dynamically cluster head (CH) selection strategy and automatic scheduling scheme of collectors for prolonging the network lifetime and shorting data gathering delay in WSAN. First the monitoring region is equally divided into several subregions and each subregion dynamically selects a sensor node as CH. These can balance the energy consumption of sensor node thereby prolonging the network lifetime. Then a task allocation method based on genetic algorithm is proposed to uniformly assign tasks to actuators. Finally the trajectory of each actuator is optimized by ant colony optimization algorithm. Simulations are conducted to evaluate the effectiveness of the proposed method and the results show that the method performs better to extend network lifetime while also reducing data delay.

• 로봇학회논문지
• /
• 제18권1호
• /
• pp.88-92
• /
• 2023

Lithium-ion batteries are actively used in various industrial sites such as field robots, drones, and electric vehicles due to their high energy efficiency, light weight, long life span, and low self-discharge rate. When using a lithium-ion battery in a field, it is important to accurately estimate the SoC (State of Charge) of batteries to prevent damage. In recent years, SoC estimation using data-based artificial neural networks has been in the spotlight, but it has been difficult to deploy in the embedded board environment at the actual site because the computation is heavy and complex. To solve this problem, neural network lightening technologies such as network pruning have recently attracted attention. When pruning a neural network, the performance varies depending on which layer and how much pruning is performed. In this paper, we introduce an optimized pruning technique by improving the existing pruning method, and perform a comparative experiment to analyze the results.

• Smart Structures and Systems
• /
• 제1권1호
• /
• pp.63-82
• /
• 2005

This paper introduces a technology for robust and low maintenance cost sensor network capable to detect accelerations below a micro-g in a wide frequency bandwidth (above 1,000 Hz). Sensor networks with such performance are critical for navigation, seismology, acoustic sensing, and for the health monitoring of civil structures. The approach is based on the fabrication of an array of high sensitivity accelerometers, each utilizing Fabry-Perot cavity with wavelength-dependent reflectivity to allow embedded optical detection and serialization. The unique feature of the approach is that no local power source is required for each individual sensor. Instead one global light source is used, providing an input optical signal which propagates through an optical fiber network from sensor-to-sensor. The information from each sensor is embedded onto the transmitted light as an intrinsic wavelength division multiplexed signal. This optical "rainbow" of data is then assessed providing real-time sensing information from each sensor node in the network. This paper introduces the Fabry-Perot based accelerometer and examines its critical features, including the effects of imperfections and resolution estimates. It then presents serialization techniques for the creation of systems of arrayed sensors and examines the effects of serialization on sensor response. Finally, a fabrication process is proposed to create test structures for the critical components of the device, which are dynamically characterized.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2008년도 춘계 종합학술대회 논문집
• /
• pp.555-559
• /
• 2008

임베디드 시스템이 발전함에 따라 다양한 플랫폼을 가진 임베디드 디바이스에서 플랫폼 독립성을 위한 자바 기술이 급속도로 발전하고 있다. SUN 사의 CDC(Connected Device Configuration)에 정의된 CVM(Classic Virtual Machine)은 이러한 플랫폼 독립적인 자바 환경을 제공한다. 특히 셋톱박스나 스마트폰과 같은 임베디드 시스템에서는 네트워크 기능을 위해 CDC의 기본 프로파일인 FP(Foundation Profile)를 사용한다. 본 논문에서는 실시간 운영체제 iRTOS에서 네트워크 기능을 구현하기 위한 네트워크 API인 FP의 네이티브 메소드에 대해 설계 및 구현한 내용을 기술한다.

• Journal of Information Processing Systems
• /
• 제18권3호
• /
• pp.428-442
• /
• 2022

A reliable automatic passenger counting (APC) system is a key point in transportation related to the efficient scheduling and management of transport routes. In this study, we introduce a lightweight head detection network using deep learning applicable to an embedded system. Currently, object detection algorithms using deep learning have been found to be successful. However, these algorithms essentially need a graphics processing unit (GPU) to make them performable in real-time. So, we modify a Tiny-YOLOv3 network using certain techniques to speed up the proposed network and to make it more accurate in a non-GPU environment. Finally, we introduce an APC system, which is performable in real-time on embedded systems, using the proposed head detection algorithm. We implement and test the proposed APC system on a Samsung ARTIK 710 board. The experimental results on three public head datasets reflect the detection accuracy and efficiency of the proposed head detection network against Tiny-YOLOv3. Moreover, to test the proposed APC system, we measured the accuracy and recognition speed by repeating 50 instances of entering and 50 instances of exiting. These experimental results showed 99% accuracy and a 0.041-second recognition speed despite the fact that only the CPU was used.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2005년도 추계종합학술대회
• /
• pp.795-798
• /
• 2005

In this paper, we analyses of additive crypto-module architecture for a sensor network. Recent research in sensor networks has raised security issues for small embedded devices. Security concerns are motivated by the development of a large number of sensor devices in the field. Limitations in processing power, battery life, communication bandwidth and memoryconstrain devices. A mismatch between wide arithmetic for security and embedded data buscombined with lack of certain operations. Then, we compared the architecture of crypto-module in this paper.

• 대한임베디드공학회논문지
• /
• 제1권2호
• /
• pp.73-81
• /
• 2006

The home network is generally separated from the Internet, as it is made up of a private network due to security issues and the lack of IPv4 addresses space. Also, a user may want to move from a terminal to another terminal connected in the home network during communicating with people outside the home. In this case, people connected in the Internet, or another home network could not communicate the user at the home. These limitations prevent a SIP-capable device connected in the home network from communicating with another SIP-capable device connected in the Internet or the outside of the home network. To overcome the limitations, This paper proposes the Adaptive SIP Application Server System as a software architecture that a user inside of the home can communicate with people outside of the home when the home is composed of a private IP-based network. Moreover, the proposed architecture provides the session mobility that allows the user to maintain a media session even if changing the terminal inside of the home during the session established. The proposed system was implemented over a home server device which acts functionality as a connection point for transmitting IP packets between a home network and the Internet.

• 컴퓨터교육학회논문지
• /
• 제17권2호
• /
• pp.115-124
• /
• 2014

하이퍼큐브와 스타 그래프는 상호연결망으로 널리 알려져 있다. 상호연결망의 임베딩은 임의의 연결망 G를 다른 연결망 H에 사상하는 것이다. 상호연결망 G가 H에 적은 비용으로 임베딩 가능하다는 것은 연결망 G에서 개발된 알고리즘들을 연결망 H에서 효율적으로 이용할 수 있는 장점이 있다. 본 논문에서는 HCN과 HON 사이의 임베딩과 스타(star)그래프와 하프팬케익그래프 사이의 임베딩을 분석한다. 연구 결과로 HCN(n,n)은 HON($C_{n+1},C_{n+1}$)에 연장율 3에 임베딩 가능하고, HON($C_d,C_d$)를 HCN(2d-1,2d-1)에 임베딩 비용은 O(d)임을 보인다. 또한 스타그래프는 하프팬케익그래프에 연장율 11, 확장율 1에 임베딩 가능하고, 평균 연장율은 8이다. 본 연구 결과는 HCN 연결망과 스타그래프에서 이미 개발된 여러 가지 알고리즘을 HON 연결망과 하프팬케익그래프에서 효율적으로 이용할 수 있음을 의미한다.