• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1469-1476
• /
• 2022

Recently, many researches have been conducted to estimate blood pressure using ECG(Electrocardiogram) and PPG(Photoplentysmography) signals. In this paper, we designed and implemented a mobile system to monitor blood pressure in real time by using 1-D convolutional neural networks. The proposed model consists of deep 11 layers which can learn to extract various features of ECG and PPG signals. The simulation results show that the more the number of convolutional kernels the learned neural network has, the more detailed characteristics of ECG and PPG signals resulted in better performance with reduced mean square error compared to linear regression model. With receiving measurement signals from wearable ECG and PPG sensor devices attached to the body, the developed system receives measurement data transmitted through Bluetooth communication from the devices, estimates systolic and diastolic blood pressure values using a learned model and displays its graph in real time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.2
• /
• pp.406-423
• /
• 2014

The recent advances in wireless communication systems and semiconductor technologies are paving the way for new applications over wireless sensor networks. Health-monitoring application (HMA) is one such emerging technology that is focused on sensing and reporting human vital signs through the communication network comprising sensor devices in the vicinity of the human body. The sensed vital signs can be divided into two categories based on the importance and the frequency of occurrence: occasional emergency signs and regular normal signs. The occasional emergency signs are critical, so they have to be delivered by the specified deadlines, whereas the regular normal signs are non-critical and are only required to be delivered with best effort. Handling the occasional emergency sign is one of the most important attributes in HMA because a human life may depend on correct handling of the situation. That is why the underlying network protocol suite for HMA should ensure that the emergency signs will be reported in a timely manner. However, HMA based on IEEE 802.15.4 might not be able to do so owing to the lack of an appropriate emergency-handling mechanism. Hence, in this paper, we propose a new emergency-handling mechanism to reduce the emergency reporting delay in IEEE 802.15.4 through the modified superframe structure. A fraction of an inactive period is modified into three new periods called the emergency reporting period, emergency beacon period, and emergency transmission period, which are used opportunistically only for immediate emergency reporting and reliable data transmission. Extensive simulation is performed to evaluate the performance of the proposed scheme. The results reveal that the proposed scheme achieves improved latency and higher emergency packets delivery ratio compared with the conventional IEEE 802.15.4 MAC.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.10A
• /
• pp.823-829
• /
• 2011

Recently, wireless body area network (WBAN) has received much attention as an application for the ubiquitous healthcare system. In WBAN, each sensor nodes and a personal base station such as PDA have an energy constraint and computation overhead should be minimized due to node's limited computing power and memory constraint. The reliable data transmission also must be guaranteed because it handles vital signals. In this paper, we propose a systematic network coding scheme for WBAN to reduce the network coding overhead as well as total energy consumption for completion the transmission. We model the proposed scheme using Markov chain. To minimize the total energy consumption for completing the data transmission, we made the problem as a minimization problem and find an optimal solution. Our simulation result shows that large amount of energy reduction is achieved by proposed systematic network coding. Also, the proposed scheme reduces the computational overhead of network coding imposed on each node by simplify the decoding process.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.