This study explores the implementation of intrusion detection solutions, specifically Snort, in industrial control environments that use network segregation as a security measure. Unlike traditional deployments at gateway ports, our research integrates Snort directly within isolated networks, customized with specific operational technology (OT) rules. The effectiveness of this system was tested using the TWISC@NCKU Critical Infrastructure cybersecurity testbed in Taiwan. This findings indicate that this strategic approach can successfully detect anomalies in network traffic, effectively addressing the common challenge of limited monitoring in segregated network environments.

View full abstract

Creating a duty roster that meets all the various requirements of nurse rostering is extremely challenging. Consequently, many researchers have studied nurse rostering. Despite these efforts, the shift schedules generated by these studies are often not practical in their initial form, as they require adjustments to accommodate various constraints and evaluation criteria. Thus, we have proposed a method for revising duty roster using Q-learning in a constructive nurse rostering. This paper explores the potential for developing a practical duty roster that accommodates nurses with varying duty plan valuations. This involves considering each nurse's lifestyle. Additionally, we visualize the duty plan valuations of the revised rosters we obtain.

View full abstract

Since Russia invaded Ukraine in 2022, information related to the invasion has increased. This information is spread through social networking services and TV programs. This study proposes a framework for a disinformation game system. In the game, a player spreads disinformation. In particular, we discuss topics related to the Russo–Ukrainian War in February 2022. The game system provides players with mental immunity to disinformation by simulating its spread. Three experiments were conducted in this game. In experiments on the proposed game, people gradually lost their resistance to disinformation by repeatedly receiving disinformation.

View full abstract

With the recent emergence of Amazon warehouses and service robots at airports and hotels, it is believed that we will enter an era in which humans and robots will struggle to coexist. Elementary and junior high school students need robotics education, which sparks their interest at an early stage. In response, higher education institutions are actively working on robot education, but most of them focus on online tracing using microcomputers. In this study, students from a technical college organized a workshop for third-year junior high school students to teach robotics about maze exploration. The lecture was evaluated by means of a questionnaire.

View full abstract

The proposed system utilizes facial recognition technology to detect and track the user, eliminating the need for physical effort in pushing a shopping cart. Additionally, the robotic trolley is equipped with an ultrasonic sensor that estimates the distance walked by the user, providing them with valuable information about their physical activity levels during the shopping trip. The implementation of this system leverages the OpenCV computer vision library within the Python programming framework, enabling the integration of the facial recognition and distance estimation capabilities. Evaluation of the developed prototype has demonstrated significant improvements in the quality of life and independence of elderly individuals, as they can now navigate the shopping environment with greater comfort and ease. This innovative robotic shopping companion represents a promising solution to enhance the shopping experience for the aging population, addressing their unique needs and empowering them to maintain their autonomy, well-being, and public health during their daily activities.

View full abstract

Miyazaki Prefecture faces several challenges in its tourism industry and has adopted various strategies to address them. In 2016, technologies like VR and AR gained attention. In response, our company developed a new smartphone app leveraging virtual technology, offering unique features to attract more tourists and encourage repeat visits.

View full abstract

Integrative data analysis, which involves the simultaneous recording of EEG, motion, and eye gazing in humans, is increasingly recognized for its potential to yield new scientific insights. To establish these simultaneous measurements, signal synchronization of multiple time courses is necessary, time data from different measurement instruments must be managed finely, and time offsets must be corrected. However, such accurate time management from the triggers requires a fine design of circuit boards with the integrated transistor-transistor logic signal voltage. This study presents a design concept and mechanism for a trigger circuit, demonstrating that the proposed experimental system effectively addresses these challenges. A custom trigger circuit board, combined with a microcontroller, was developed to handle voltage signals and ensure seamless synchronization between different devices. The trigger signal output from the circuit board can be controlled and demonstrated through system integration across multiple measurement instruments. It should be noted that the accuracy of trigger signal output may be influenced in part by the system clock of the operating system, but the integrated trigger circuit board and the measurement system enable precise and reliable synchronous measurements, with the potential for further synchronous analysis.

View full abstract

This research addresses the challenges posed by the "curse of dimensionality" in machine learning, particularly in the context of non-image data. As the number of features increases, the complexity and computational costs escalate, making accurate model construction difficult. Two primary strategies are employed to mitigate this issue: dimensionality reduction techniques and methods that restrict feature combinations. This study aims to develop a generic DeepInsight framework that systematically transforms variable configuration, feature extraction, and model building processes. By applying this framework to stock price prediction, we seek to enhance the predictive accuracy and validate the effectiveness of Deep Learning approaches in non-image data analysis.

View full abstract

Cryptocurrency price fluctuations, though widely studied, remain unpredictable, posing risks for investors. While many aspire to profit, the volatility makes accurate prediction challenging. Deep learning has recently gained traction as a promising approach for cryptocurrency price forecasting. This study focuses on long-term price prediction for Bitcoin, Ethereum, Litecoin, and Cardano using LSTM, GRU, and Bi-LSTM models. The performance of these models is evaluated and compared with findings from previous studies.

View full abstract

This paper proposes a high-speed parameters estimation method for compartment model where output function is described by the convolution between input function and impulse response, which is like a time-invariant linear system. The proposed method uses linear regression analysis based on the equivalently transformed equation that can be obtained using the processing of Differentiation of Convolution with Exponential function (DCE). In this paper, taking the issue of parameters estimation in PET (Positron Emission Tomography) inspection system and RLC series electrical circuit for examples, we reveal that the method can estimate parameters of those impulse responses in high speed, by showing the experimental results. Moreover, we explain the advantage of our method comparing to the conventional standard method such as conjugate gradient method and the extended Newton one.

View full abstract

In high-frequency electromagnetic field analysis, it is necessary to divide the waveform into elements whose maximum side length is 1/10 to 1/20 of the wavelength so that the waveform can be expressed with low error when a space is divided. Therefore, in many cases, it becomes a largescale problem. In this study, due to the need for electromagnetic compatibility (EMC) evaluation of the surrounding electromagnetic field generated when a microwave scalpel is used in surgery, a large-scale electromagnetic analysis method for the microwave region inside the operating room has been developed. In this paper, the introduced parallel mesh subdivision function enables highspeed mesh generation with hundreds of millions of elements. In addition, this method provided a successful analysis.

View full abstract

Ensuring the data integrity and safety of dam infrastructure is a major challenge. In recent years, numerous cyberattacks have targeted critical infrastructure, underscoring the importance of security systems. To address this challenge, we propose DamChain, a warning system designed to improve the data integrity and security of dam infrastructure based on blockchain technology. The decentralized ledger of blockchain records data in a transparent and immutable manner, which is important for ensuring the integrity of data and preventing unauthorized modification. This system aims to provide a solution that maintains the data integrity and is designed to provide a warning mechanism about water levels, improving the security and data integrity of dams.

View full abstract

Testbeds play a crucial role in cybersecurity research for critical infrastructure by simulating realistic environments. This paper presents an in-depth examination of the MiniDAM toolkit and our developed testbed, which is modeled on actual dam operation standards. We provide a detailed overview of its functionalities. Additionally, a comparative study is conducted between MiniDAM, our testbed, MiniCPS, and the Secure Water Treatment (SWaT) testbed. This work further discusses the process of generating datasets and incorporating additional features. The capabilities of MiniDAM are demonstrated as a robust platform, significantly contributing to the advancement of research in dam cybersecurity.

View full abstract

This paper describes an innovative design by using blockchain to improve the verification and security of data in industrial control systems. Combining the decentralized-blockchain, the Programmable Logic Controller, and Human Machine Interface. The design strengthens the overall data security. With the immutable recording function of blockchain, PLC can manage data interactions and conduct real-time monitoring. After testing and simulation, the practicability and effect of this innovative design were proved, and shows the potential of this work.

View full abstract

The music notation of Guqin (“古琴”, Chinese seven-string zither) named “Jianzi Pu (減字譜, simplified music notation of Guqin)” was invented at the Middle of A.D. 700, and Guqin music remained more than 600, however, only about 100 of them are played in nowadays. The reason is that the handwritten “Jianzi Pu” is hard to be understood even for experts or professional Guqin players. In this study, we applied deep learning methods such as VGG16 and YOLOv5 to the recognition of a Guqin notation “Sen-O-So” (仙翁操, Melody of the Immortal Elder). Firstly, we created a dataset including 55 kinds of single characters of Sen-O-So in 4,951 images from 23 versions found on the Internet and obtained by data augmentation, i.e., image processing such as rotation, enlargement (zoom-in), reduce (zoom-out), various filtering, etc. Secondly, we compared the recognition rates of VGG16 and YOLOv5 in the experiment. The average accuracies of 55 classes images by VGG16 and YOLOv5 were 87.50% and 88.47% respectively for the test data. Additionally, we created a dataset of Sen-O-So video clips to match the recognition results of single characters by YOLOv5 and realized an online ancient music restoration system development.

View full abstract