• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 추계학술대회
• /
• pp.345-346
• /
• 2018

Internet of Things (IoT) data is collected in real time and is treated as highly reliable data because of its high precision. However, IoT data is not always highly reliable data. Because, data be often incomplete values for reasons such as sensor aging and failure, poor operating environment, and communication problems. So, we propose the methodology for solve this problem. Our methodology implements multiple LSTM networks to individually process the data collected from the sensors and a single LSTM network that batches the input data into an array. And, we propose an efficient method for constructing LSTM in IoT environment.

• 한국음향학회지
• /
• 제30권2호
• /
• pp.86-91
• /
• 2011

In the underwater communications, the acoustic channel is in poor communication conditions, such as long propagation delay, narrow bandwidth, and high bit-error rate. For these bad acoustic channels, we propose an efficient automatic repeat request (ARQ) for multi-hop underwater network by using the concepts of concurrent bi-directional transmission, multiple sub-packets, and overhearing data packet instead of the acknowledgement signal. Our results show that the proposed ARQ significantly reduces the transmission latency especially in high BER compared with the existing Stop and Wait ARQ.

• Journal of information and communication convergence engineering
• /
• 제7권2호
• /
• pp.176-181
• /
• 2009

This study is part of a project to develop and improve mooring systems for oceanographic use that include an electro-optical sensor, 1MHz Nortek Aquadopp Doppler Profiler and AIRMAR multipurpose Sensor. The adaption of Doppler current profilers to measure directional wave spectra has provided a new instrumentation approach to coastal and nearshore oceanographic studies. The HEIOB is developed are light weight and of a compact design, and can be easily installed in marine environment. Since there are no base station and gateways in marine environments, we selected CDMA and Orbcomm to send the data information. Therefore, the data can be sent by either e-mail service or Short Message Service (SMS). This paper will present some of scientific sensor results regarding real-time oceanographic and meteorological parameters such as wind spend, wind direction, wave direction, and etc. The modeling and test results highlight the engineering challenges associated with designing these systems for long lifetimes. It can also be used in future application to build wave observation buoy network in real-time using multiple ubiquitous buoys that share wave data and allow analysis of multipoint, multi-layer wave profiler.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2015년도 춘계학술대회
• /
• pp.116-119
• /
• 2015

다수의 이동센서노드를 이용한 환경감시 알고리즘을 제안한다. 제안하는 알고리즘은 사전 정보가 없는 개활 지역에서 이동센서노드들을 배치하여 센싱 커버리지를 최대화 하는 것을 목적으로 한다. 이동센서노드는 보다 높은 센싱 커버리지를 획득하기 위하여 최대의 센싱 커버리지를 획득할 때 까지 반복적으로 재배치된다. 커버리지 최대화를 위하여 ILP(Integer Linear Programming) 기반의 최적화 문제를 구성한다. 시뮬레이션 결과에 의하면 제안된 알고리즘은 이동센서노드들을 보다 높은 관심지역으로 이동시킬 수 있으며 최대의 센싱 커버리지 획득이 가능함을 시뮬레이션을 통하여 검증한다.

• 한국통신학회논문지
• /
• 제34권3B호
• /
• pp.213-224
• /
• 2009

Most of existing routing methods in wireless sensor networks to counter the local eavesdropping-based packet-tracing deal with a single asset and suffer from the packet-delivery latency as they prefer to take a separate path of many hops for each packet being sent. Recently, the author proposed a routing method, GSLP-w(GPSR-based Source-Location Privacy with crew size w), that enhances location privacy of the packet-originating node(i.e., active source) in the presence of multiple assets, yet taking a path of not too long. In this paper, we present a refined routing(i.e., next-hop selection) procedure of it and empirically study privacy strength and delivery latency with varying the crew size w(i.e., the number of packets being sent per path). It turns out that GSLP-w offers the best privacy strength when the number of packets being sent per path is randomly chosen from the range [$1,h_{s-b}/4$] and that further improvements on the privacy are achieved by increasing the random walk length TTLrw or the probability prw that goes into random walk(where, $h_{s-b}$ is the number of hops of the shortest path between packet-originating node s and sink b).

• Smart Structures and Systems
• /
• 제3권4호
• /
• pp.455-474
• /
• 2007

A flexibility-based distributed computing strategy (DCS) for structural health monitoring (SHM) has recently been proposed which is suitable for implementation on a network of densely distributed smart sensors. This approach uses a hierarchical strategy in which adjacent smart sensors are grouped together to form sensor communities. A flexibility-based damage detection method is employed to evaluate the condition of the local elements within the communities by utilizing only locally measured information. The damage detection results in these communities are then communicated with the surrounding communities and sent back to a central station. Structural health monitoring can be done without relying on central data acquisition and processing. The main purpose of this paper is to experimentally verify this flexibility-based DCS approach using wired sensors; such verification is essential prior to implementation on a smart sensor platform. The damage locating vector method that forms foundation of the DCS approach is briefly reviewed, followed by an overview of the DCS approach. This flexibility-based approach is then experimentally verified employing a 5.6 m long three-dimensional truss structure. To simulate damage in the structure, the original truss members are replaced by ones with a reduced cross section. Both single and multiple damage scenarios are studied. Experimental results show that the DCS approach can successfully detect the damage at local elements using only locally measured information.

• 한국컴퓨터정보학회논문지
• /
• 제13권7호
• /
• pp.157-164
• /
• 2008

본 논문에서는 센서 노드들이 트리(tree) 구조를 가지며 주기적인 트래픽을 전송하는 무선센서 네트워크에서 싱크 노드로 데이터를 효율적으로 전송하기 위한 스케줄링 알고리즘을 치안한다. 관련 연구[1]에서는 주기적인 트래픽을 효율적으로 전송하는 알고리즘을 제안하였으나, 트리구조가 깊어지고, 자식 노드들이 많아짐에 따라 에너지 소모가 증가하고, 데이터 처리율이 떨어지는 문제점을 보였다. 제안하는 알고리즘은 유휴 타임 슬롯(Time Slot)을 활용하고, 데이터를 중복되지 않게 전송함으로 효율적인 모니터링을 가능케 한다. 또한 유사 데이터에 대한 효율적인 전송 방안 및 긴급 데이터가 발생할 경우 EDP(Emergency Data Period)를 추가하여 전송대기시간을 줄이게 하는 방법도 제안하고자 한다. 성능 분석을 통하여 제안하는 알고리즘이 기존연구보다 더 높은 데이터 처리율을 보이고, 에너지 소모를 적게 함을 보였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 추계학술발표대회
• /
• pp.830-832
• /
• 2019

IoT 기술의 도래로 인하여 실생활에 사용되는 사물들에 Sensor가 부착되어 시간마다 Sensor data가 발생하는 세상이 열리게 되었다. 이러한 IoT Device들에 부착되어 있는 sensor를 통하여 수집이 된 data는 방대한 양을 가지기 때문에 Deep Learning에 적용하는데 충분하며 아주 중요한 역할을 한다. 이러한 IoT Device들은 우리의 실제 생활에 아주 가까이 다양한 환경으로 접할 수 있다. 예를 들어 스마트시티, 스마트팩토리, 스마트홈 등이 있다. 이러한 것들은 우리의 일상생활에 편리함과 직결되어 있다. 본 논문에서는 Smart home 환경에서의 Multi Resident Activity Recognition이다. Smart home의 가구에 부착되어 있는 센서에서 발생된 센서데이터를 활용하여 1) Training Similarity Network, 2) Embedding, 3) Clustering, 4) Recognizing 네 단계 프로세스를 거쳐 문제를 해결한다. 그 결과, 우리가 제안한 프로세스를 통하여 차원 축소 효과와 Un-seen data를 효과적으로 처리할수 있게 된다.

• ETRI Journal
• /
• 제45권6호
• /
• pp.1079-1089
• /
• 2023

Stroke is one of the leading causes of long-term disability worldwide, placing huge burdens on individuals and society. Further, automatic human activity recognition is a challenging task that is vital to the future of healthcare and physical therapy. Using a baseline long short-term memory recurrent neural network, this study provides a novel dataset of stretching, upward stretching, flinging motions, hand-to-mouth movements, swiping gestures, and pouring motions for improved model training and testing of stroke-affected patients. A MATLAB application is used to output textual and audible prediction results. A wearable sensor with a triaxial accelerometer is used to collect preprocessed real-time data. The model is trained with features extracted from the actual patient to recognize new actions, and the recognition accuracy provided by multiple datasets is compared based on the same baseline model. When training and testing using the new dataset, the baseline model shows recognition accuracy that is 11% higher than the Activity Daily Living dataset, 22% higher than the Activity Recognition Single Chest-Mounted Accelerometer dataset, and 10% higher than another real-world dataset.

• 제어로봇시스템학회논문지
• /
• 제19권2호
• /
• pp.125-130
• /
• 2013

Compared to OWR (One-Way Ranging) method that requires precise network time synchronization, TWR (Two-Way Ranging) method has advantages in building an indoor WPAN (Wireless Personal Area Network) location system with lower cost. However, clock offsets of nodes in WPAN system should be eliminated or compensated to improve location accuracy of the TWR method. Because conventional clock offset elimination methods requires multiple TWR transactions to reduce clock offset, they produce network traffic burden instead. This paper presents a clock offset estimation method that can reduce clock offset error with a single TWR transaction. After relative clock offsets of sensor nodes are estimated, clock offsets of mobile tags are estimated using a single TWR communication. Simulation results show that location accuracy of the proposed method is almost similar to the conventional clock offset elimination method, while its network traffic is about a half of the conventional method.