• Journal of Advanced Navigation Technology
• /
• v.25 no.4
• /
• pp.299-304
• /
• 2021

Existing human activity recognition systems detect activities through devices such as wearable sensors and cameras. However, these methods require additional devices and costs, especially for cameras, which cause privacy issue. Using WiFi signals that are already installed can solve this problem. In this paper, we propose a CNN-based human activity recognition system using channel state information of WiFi signals, and present results of designing and implementing accelerated hardware structures. The system defined four possible behaviors during studying in indoor environments, and classified the channel state information of WiFi using convolutional neural network (CNN), showing and average accuracy of 91.86%. In addition, for acceleration, we present the results of an accelerated hardware structure design for fully connected layer with the highest computation volume on CNN classifiers. As a result of performance evaluation on FPGA device, it showed 4.28 times faster calculation time than software-based system.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.120-126
• /
• 2020

We present a user recognition method using human body impulse response signals. The body compositions vary from person to person depending on the portion of water, muscle, and fat. In the body communication study, the body has been interpreted circuit models using capacitance and resistances, and its characteristics are determined by the body compositions. Therefore, the individual body channel is unique and can be used for user recognition. In this paper, we applied pseudo impulse signals to the left hand and recorded received signals from the right hand. The empirical mode decomposition (EMD) method removed noise from the received signals and 10 peak values are extracted. We set the differences between peak amplitudes as a key feature to identify individuals. We collected data from 6 subjects and achieved accuracy of 97.71% for the user recognition application.

• Structural Engineering and Mechanics
• /
• v.63 no.6
• /
• pp.771-777
• /
• 2017

The applications of smartphones or other portable smart devices have dramatically changed people's lifestyle. Researchers have been investigating useage of smartphones for structural health monitoring, earthquake monitoring, vibration measurement and human posture recognition. Their results indicate a great potential of smartphones for measuring pedestrian-induced loads like walking, jumping and bouncing. Smartphone can catch the device's motion trail, which provides with a new method for pedestrain load measurement. Therefore, this study carried out a series of experiments to verify the application of the smartphone for measuring human-induced load. Shaking table tests were first conducted in order to compare the smartphones' measurements with the real input signals in both time and frequency domains. It is found that selected smartphones have a satisfied accuracy when measuring harmonic signals of low frequencies. Then, motion capture technology in conjunction with force plates were adopted in the second-stage experiment. The smartphone is used to record the acceleration of center-of-mass of a person. The human-induced loads are then reconstructed by a biomechanical model. Experimental results demonstrate that the loads measured by smartphone are good for bouncing and jumping, and reasonable for walking.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.3
• /
• pp.262-267
• /
• 2013

The recognition of human emotional state is one of the most important components for efficient human-human and human- computer interaction. In this paper, four emotions such as fear, disgust, joy, and neutral was a main problem of classifying emotion recognition and an approach of visual-stimuli for eliciting emotion based on physiological signals of skin conductance (SC), skin temperature (SKT), and blood volume pulse (BVP) was used to design the experiment. In order to reach the goal of solving this problem, half-against-half (HAH) multi-class support vector machine (SVM) with Gaussian radial basis function (RBF) kernel was proposed showing the effective techniques to improve the accuracy rate of emotion classification. The experimental results proved that the proposed was an efficient method for solving the emotion recognition problems with the accuracy rate of 90% of neutral, 86.67% of joy, 85% of disgust, and 80% of fear.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.1
• /
• pp.41-49
• /
• 2005

Electromyogram (EMG) signal generated by voluntary contraction of muscles is often used in rehabilitation devices because of its distinct output characteristics compared to other bio-signals. This paper proposes a novel EMG-based human-computer interface for electric-powered wheelchair users with motor disabilities by C4 or C5 spine cord injury. User's commands to control the electric-powered wheelchair are represented by shoulder elevation motions, which are recognized by comparing EMG signals acquired from the levator scapulae muscles with a preset double threshold value. The interface commands for controlling the electric-powered wheelchair consist of combinations of left-, right- and both-shoulders elevation motions. To achieve a real-time interface, we implement an EMG processing hardware composed of analog amplifiers, filters, a mean absolute value circuit and a high-speed microprocessor. The experimental results using an implemented real-time hardware and an electric-powered wheelchair showed that the EMG-based human-computer interface is feasible for the users with severe motor disabilities.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.3
• /
• pp.286-292
• /
• 2009

This paper proposes a knee-wearable robot system for assisting the muscle power of human knee by processing EMG (Electromyogram) signals. Although there are many muscles affecting the knee joint motion, the rectus femoris and biceps femoris among them play a core role in the extension and flexion motion, respectively, of the knee joint. The proposed knee-wearable robot system consists of three parts; the sensor for measuring and processing EMG signals, controller for estimating and applying the required knee torque, and actuator for driving the knee-wearable mechanism. Ultimately, we suggest the motion control method for knee-wearable robot system by processing the EMG signals of corresponding two muscles in this paper. Also, we show the effectiveness of the proposed knee-wearable robot system through the experimental results.

• Proceedings of the KIEE Conference
• /
• 1993.11a
• /
• pp.346-349
• /
• 1993

A method for image compression based on decomposition is presented. We design the efficient coding technique for upper band image signals. This coding technique with directive 1-D DPCM is based on the statistical properties of upper bands. Lower band image signals is encoded using 2-D DPCM. The directive 1-D DPCM is performed, scanning upper bands according to edge direction. And then the predicted error signals of upper band sis coded using B1 and Huffman code, and the predicted error signals of lower band is coded using Huffman code. The proposed system shows improved performance when compared with other existing methods with respect to peak signal to noise ratio(PSMR) and human visual system(HVS) properties.

• Journal of Biomedical Engineering Research
• /
• v.15 no.4
• /
• pp.455-464
• /
• 1994

The intelligent trajectory control method that controls moving direction and average velocity for a prosthetic arm is proposed by pattern recognition and force estimations using EMG signals. Also, we propose the real time trajectory planning method which generates continuous accelleration paths using 3 stage linear filters to minimize the impact to human body induced by arm motions and to reduce the muscle fatigue. We use combination of MLP and fuzzy filter for pattern recognition to estimate the direction of a muscle and Hogan's method for the force estimation. EMG signals are acquired by using a amputation simulator and 2 dimensional joystick motion. The simulation results of proposed prosthetic arm control system using the EMG signals show that the arm is effectively followed the desired trajectory depended on estimated force and direction of muscle movements.

• Nuclear Engineering and Technology
• /
• v.25 no.3
• /
• pp.394-402
• /
• 1993

A novel framework for quantitative estimates of the benefits of signals on nuclear power plant availability and economics has been developed in this work. The models developed in this work quantify how the perfect signals affect the human operator's success in restoring the power plant to the desired state when it enters undesirable transients. Also, the models quantify the economic benefits of these perfect signals. The models have been applied to the condensate feedwater system of the nuclear power plant for demonstration.