In 2021, Masaoka, Ichikawa, and Fujieda proposed generating unpredictable random numbers by sampling a hardware LFSR (Linear Feedback Shift Register) at uneven intervals, where the source of entropy was the fluctuation of sampling intervals. Then, in 2023, Chiba and Ichikawa generated random numbers that passed the NIST test by utilizing weather data downloaded from the Internet as the source of entropy. This study proposes the use of digital images as the entropy source for random number generation. Our evaluation results indicate that a 64-bit LFSR should be used, and that fluctuations of the sampling intervals should be generated with the least significant bit of each pixel value. The random numbers generated by the proposed method passed both the Diehard and NIST SP 800-22 tests.

Chiba and Ichikawa (2023) proposed an unpredictable random number generator (URNG) that samples a linear feedback shift register (LFSR) at varying intervals fluctuated by weather data. Although their URNG meets the criteria of the Diehard and NIST tests, it requires a substantial amount of weather data. This study examines various design options to reduce wind direction data. Our results show that the resulting random sequence passes the Diehard and NIST tests under the following conditions: (1) use of a 64-bit or longer LFSR, (2) incorporation of a robust hash function (e.g., MD5) for data selection, and (3) adoption of 64 or greater number of observations with sufficient entropy.

Currently, the “Small Boat Lecture Class” at fisheries high schools faces challenges with varying individual student progress within the curriculum. To address these issues, we designed a personalized learning support system (Referred to as “the system”) utilizing video lectures based on instructional design theory. The effectiveness of the system was evaluated by comparing it with traditional lecture-based teaching. The system integrated Microsoft Teams and video lectures and was implemented through a learning management system. The educational effectiveness of the proposed system was verified throughout the annual curriculum by conducting knowledge assessment tests confirmed. It was shown that self-paced learning improved student motivation and outcomes. Additionally, challenges such as sample size, the development workload of instructional materials, and the need for tailored approaches to students with slower comprehension were identified.

This study investigates the relations between three types of different fuselage geometries and the lift coefficient of spaceplanes using fluid dynamic simulation. The differences are clarified by the local forces, such as pressure and local-lift coefficients. The results suggest no significant differences in the lift coefficient between the vertical-long circle and the true one. Concurrently, the horizontal fuselage had a significantly lower lift coefficient than other fuselages. Similarly, the lift gradient of the horizontal fuselage showed lower values than other fuselages. These lower values were caused by the differences in the pressure distribution at the wing root near the fuselage, which produced two major differences in the local-lift coefficient. First, the increase in lift at the wing leading edge does not occur even if the angle of attack is higher. Second, negative lift occurs at the fuselage behind the wing trailing edge, and the value becomes larger for the horizontal-long circle fuselages.

This study proposes a Long-Short Term Memory (LSTM) model to assess the risk of falling among the elderly considering daily gait conditions. In constructing such an LSTM model, it is extremely difficult to collect gait data (time-series data of triaxial acceleration and triaxial angular velocity) from elderly persons at high risk of falling. This study addresses this problem by generating gait data with high risk of falling by healthy adults wearing orthoses developed in a previous study to compensate for knee extensor strength. Using the generated gait data, we developed an LSTM model to assess the gait state with high risk of falling, as quantified by the Timed Up and Go test score. The results show that the gait condition with high risk of falling can be estimated with 100% accuracy when the input time-series length exceeds 3 steps of gait data. Finally, this paper describes the use of the LSTM model developed in this study for transfer learning and fine-tuning to establish a model for fall risk assessment based on the actual gait condition of elderly people.

This study develops a generative AI literacy education program for students at the Faculty of Education, University of the Ryukyus, and assesses its effectiveness using the ARCS Motivation Model. With the rapid advancement of generative AI technology, AI literacy is increasingly important in the educational field. The program, part of the 7th-grade course “Utilization of ICT in Education”, includes lectures and practical exercises on the mechanisms and application of generative AI. To evaluate the program's effectiveness, a questionnaire based on Keller's ARCS model was administered, assessing four factors: Attention, Relevance, Confidence, and Satisfaction. The results indicated that students rated the program particularly highly on the factors of “Attention” and “Satisfaction”, demonstrating increased interest and a sense of achievement toward generative AI. These findings confirm the effectiveness of the program and validate its design.

This paper describes the development of an anomaly detection system for players in a rugby match to provide appropriate aid and avoiding severe injury. In the developed system, deep neural networks with fine tuning and its combination of different types of construction are key techniques for anomaly detection in rugby match footage. Certain appearance-based characteristics are captured from movie streams, and the system uses deep neural networks to automatically classify concerned scenes, especially, lying down scenes. The proposed system performs very well by recognizing many types of appearances of players within occlusion from a free viewpoint. Experimental results show that the system can effectively detect concerned players with high accuracy.

This study examined the decline in fishery production and trends in marine resources along the Okinawa coast. Using catch data from 1996 to 2019 provided by the Okinawa Prefectural Fisheries and Ocean Technology Center, specifically, from set nets in Chinen and Yomitan, we compared the effectiveness of PCA and t-SNE, two visualization techniques for dimensionality reduction. Our results showed that t-SNE provided a more accurate analysis. Following this, we conducted cluster analysis, which revealed that Chinen's clusters from the 1990s were distinct, suggesting that the reclamation of the Awase tidal flat, which started in 2002, may have had an impact. By contrast, no clear decade-based clusters were observed in Yomitan, though annual variations in sea conditions were evident. Additionally, an analysis of the annual catch volume trajectories after dimensionality reduction indicated that recent trajectory patterns in both Chinen and Yomitan diverge from historical trends. This suggests that recent changes in sea conditions may be affecting the fish species' composition differently than that in previous years. The findings of this study offer a more precise understanding of the trends and status of marine resources along the Okinawa coast, providing a foundation for future resource management measures and fisheries policies.

In electric railroads, inverter-controlled vehicles have been used for a long time. In recent years, efficiency of each subsystem of energy-saving traction drive systems has been increased. However, the energy efficiency during operation has not been studied, especially during cruising control. This study developed a method to analyze and reduce powering energy consumption in cruising control based on operation data obtained by an IoT remote monitoring system. The study also showed the achievement of energy saving by performing simulation.

The efficiency of a wireless power transfer system changes with the load. To solve this problem, an impedance matching method has been developed. Conventional methods control the equivalent load to the optimal value using circuit parameters and the detected load voltage and current. However, they require estimating the mutual inductance M in case of misalignment or an airgap change. M-estimation has some issues such as a limited operating range. In this study, we developed a high-efficiency operation method that can respond to a change in M without its estimation by controlling the currents in the coils to equalize them. We confirmed that the developed method enables highly efficient operation without M-estimation despite misalignment.

This study proposes a maximum thrust control method for a linear synchronous motor with half-wave-rectified self-excitation. First, we formulate the currents to realize the maximum thrust per ampere control for the linear motor to make it simple and practical. Subsequently, a new field weakening control and maximum thrust per volt control schemes are proposed to determine the current components such that the thrust is truly maximum within the current and voltage limits. The proposed control methods can be used to obtain a maximum thrust control method over the full velocity range for the linear motor that produces maximum thrust within the current and voltage limits. The operation characteristics of the proposed control method are clarified through simulations. The results show that the proposed control method improves the thrust in the high-velocity region.

Air conditioners with an active filter have excess distortion compensation capacity during periods of light loads such as intermediate seasons. It was assumed that this excess capacity can improve the leading power factor at the power receiving point. This study investigates the possibility of using the excess distortion compensation capacity in an automatic power factor regulator of an air conditioner with an active filter to enhance the leading power factor at the power receiving point of the electric system during periods of light loads. Based on the three-year results of a demonstration test, the desired control level was achieved for power factor improvement at the power receiving point. Thus it is expected that the static capacitor can be discarded by power factor control using air conditioners with active filters.

This study describes the experimental verification of series-connected power semiconductor devices and quantifies the voltage imbalance which occurs during the switching operation when the gate magnetic coupling and the addition of capacitor between drain and gate terminals methods are combined. Experiments using 3.3kV/750A SiC-MOSFET/SiC-SBD power modules showed the voltage imbalance evaluation originating from the difference in gate signal transmission and gate drive voltage for the conventional, magnetic coupling, capacitor addition methods and the two methods in combination. The combination of the two methods minimizes the voltage imbalance under these conditions. Moreover, the relationship between capacitance and switching losses is clarified.

This study presents the design, construction, and experimental validation of a high-force-density tubular permanent-magnet (PM) linear motor for aerospace applications. Initially, a strut structure was developed to enhance the contamination resistance and structural rigidity of the motor, which was validated through experimental testing. Subsequently, a motor topology was designed using two-dimensional finite element analysis to achieve a target thrust density of 400N/kg (specifically 500N/kg for the electromagnetic component) while maintaining a low detent force. The motor design was optimized for major pole number/slot number combinations of 8/24, 16/24, 20/24, 28/24, 32/24, and 40/24. The optimal design, featuring a three-phase 16/24 combination with a single-layer winding, demonstrated the highest force-to-mass density. The proposed design and its anticipated drive characteristics were experimentally validated using a prototype, confirming the effectiveness of the overall design.

Our research focuses on power electronics. In particular, we research motor drive systems and semiconductor power converters. In the motor drive system, switched reluctance motor and synchronous reluctance motor are studied. In the semiconductor power converter, high-frequency isolated DC-DC converter and the Maximum Power Point Tracking control for photovoltaic and wind power generations are researched.