34254 research outputs found
Sort by
A Hybrid Cluster-based Approach for Emergency Message Transmission with Navigation Integration
No full textIn the realm of modern transportation systems, prioritizing road safety stands as an utmost necessity. Vehicular Ad Hoc Networks (VANETs) emerge as a promising solution, offering a dynamic platform for the seamless dissemination of Emergency Messages (EM). By issuing proactive alerts and warnings, VANETs empower drivers to swiftly make informed decisions, mitigating potential collisions and lowering road hazards. Nevertheless, challenges like the swift movement of vehicles, uneven distribution of network nodes, and signal attenuation present significant hurdles that can lead to a decline in VANET's transmission reliability and responsiveness. Addressing these challenges, this paper delves into an innovative approach known as HCBEN. This approach integrates Road Side Units (RSUs) and clusters to tackle the aforementioned problems. By grouping vehicles into clusters, redundancy can be minimized, and the network's scalability can be improved. Moreover, the incorporation of RSUs extends the communication range, facilitating seamless transmission over long distances while ensuring timely and dependable delivery of EMs. Furthermore, we introduce data from vehicle navigation systems to forecast potential future paths, thereby optimizing EM propagation efficiency. Through this collaborative framework, EM transmission achieves high delivery raters, mitigating broadcast storm issues and extending the transmission ranges
3D Hybrid Silicon Photonics III-V Semiconductor Optical Amplifiers integrated Electro-absorption Modulator
No full textSilicon photonics technology continues to advance and mature on silicon materials, and it is applicable to existing CMOS processes. However, due to the non-direct bandgap characteristics of silicon, an external light source system is required for operation, making the current external light source systems complex and expensive. In order to effectively reduce manufacturing costs and achieve mass production, the integration of the light source onto silicon photonic chips becomes crucial. Therefore, integrating active and passive components is one of the key technologies in the development of silicon photonic integrated circuits.
This study adopts mature wafer bonding techniques in the laboratory to vertically integrate III-V (three-five) group chips onto SOI (Silicon On Insulator) substrates. After completing the III-V active components, alignment is achieved using III-V waveguides, and selective bottom-cut etching of silicon waveguides is realized through dry etching and self-alignment methods. By employing CF4 and O2 for high-selectivity etching of silicon, we successfully etch only the silicon without affecting the III-V active waveguides.
This experiment integrates III-V epitaxial chips into SOI substrates, achieving vertical integration of both active and passive components. Among them, the electro-absorption modulator (EAM) of the active component exhibits an extinction ratio (ER) exceeding 15dB, while the integrated semiconductor optical amplifier (SOA) achieves gain saturation at 65mA, with a gain exceeding 7dB. This confirms the resonance effect generated between the horizontal gradient waveguide and silicon, achieving an optical mode conversion efficiency of over 80%. The III-V p-i-n heterogeneous structure designed in the laboratory exhibits a strong Quantum-confined Stark effect (QCSE), especially regarding the measurement of the impact of reverse bias on the quantum well
Influence of crack number and location on the strength of PCB (Printed Circuit Board)
No full textDuring assembly process of electronic packages, printed circuit boards (PBC), and pins, microscale cracks not visible by naked eyes could exist. The existence of these microcracks could result in damages of electronic components. The location of the crack can be anywhere. In fact, the cracks can exist in any location in the electronic devices. In the literature, most of the papers about crack propagation growth use Virtual Crack Closure Technique (VCCT) and Cohesive Zone Element (CZE). However, a predefined crack propagation path has to be defined in advance when VCCT method and CZE method are used. The crack propagation paths are not predictable in priori. In tis thesis, the influence of the crack locations on the strength of printed circuit boards is investigated in a hope to provide advices for extend the life of printed circuit boards.
The models for simulations in this thesis are built from electronic packages, printed circuit boards, and pins. EXTENDED FINITE ELEMENT METHOD (XFEM) are used to conduct numerical simulations. Firstly, in the first group simulations, we consider the location of a single crack on the strength of PBC. In the second group of simulations, the cracks in the first group are rotated with respect to x,y , and z axis. In the third group simulations, we investigate the influence of location of two cracks on the strength of PBC. The results show the location can affect the strength of PCB and the existence of another crack reduces the strength of PCB
Stock identification and distribution pattern of the silky shark (Carcharhinus falciformis) in Indo-Pacific Ocean
No full textThe Silky shark (Carcharhinus falciformis), a member of the family Carcharhinidae, is found in the diverse marine ecosystems across the oceans worldwide. There has been a substantial decline in their population, primarily attributed to excessive fishing activities. These sharks are frequently caught as bycatch and are in high demand for their fins in international trade. There is an urgent need to operate an international and regional strategies for their conservation and management. In which understanding of their stock structure and relative fishery indexes is fundamental for effective management planning.
In Chapter 2, insights into the genetic stock identification of the silky shark within the Indo-Pacific region are presented. A prominent genetic distinction is observed between populations in the Aceh region compared to other areas, with the effective population size (Ne) as low as 2.3, suggesting a distinctive and relatively small population. These insights are pivotal for targeted conservation and management measures. The population of the silky shark in Aceh is recommended to be recognized as a distinct stock for further management.
Chapter 3 Three distinct genetic groups are defined based on previous findings. This chapter employs element analysis to test whether it could be a tool to verify the stock structure revealed by the genetic study. Principal Component Analysis (PCA) and individual elemental assays were used to determine if there are discernible differences attributable to region, stock, or gender in the element composition. The findings conclude that, based on the PCA, there are no significant differences across genders, regional group and genetic stocks. However, significant variations are observed in the concentration of potassium (K), lithium (Li) and magnesium (Mg) among stocks, as well as in the regional distributions of these and calcium (Ca). The absence of comprehensive species length, biological and environmental data limits the ability to use element analysis in stock identification.
Chapter 4 examines the distribution and relative abundance of the silky shark in the Indian Ocean, using Logbook and observer data from Taiwanese large-scale tuna longline fisheries. A zero-inflated negative binomial (ZINB) model is used to normalize the catch effort data. The analysis detects significant size differences between male and female sharks, with juveniles predominantly found between latitudes 10\uc2\ub0 S and 10\uc2\ub0 N. This comprehensive approach enhances the understanding of spatial and temporal patterns of the silky shark populations in the Indian Ocean and contributes to the development of effective stock assessments and management strategies.
With the collective findings from this thesis, it is imperative to classify the silky shark population into three separate management units in the Indo-Pacific region. An in-depth evaluation of the status of the silky shark in the Oceans remains allowing for fishing is urgently needed to understand its present conditions and develop improved fishery management approaches
Application of One Dimensional Convolutional Neural Networks to Optical Fiber Oil Slick Sensing
No full textOffshore oil spills, which have garnered widespread attention worldwide, damage marine ecosystems. In port areas, because of intensive maritime traffic and operations, real-time oil spill detection and response is extremely critical. Failure to do so causes oil slicks to spread, dramatically increasing the costs of emergency response and cleanup. Furthermore, ocean currents and sea breezes further exacerbate the range of oil pollution, expanding the scope of ecological and environmental damage. It is imperative to develop real-time oil slick sensors. In this study, a high-sensitivity optical oil slick sensor was improved, and a one-dimensional convolutional neural network was applied for real-time automatic oil slick detection.
The fiber optic oil slick sensor, developed by Hsieh et al. (2023), was improved by modifying the core layer of the conventional optical fiber as the sensing area. An oil slick sensing experiment was performed according to the difference in refractive index between seawater and slick oil, revealing that the sensor was susceptible to light intensity. Therefore, a beam splitter was integrated into the optical system to enhance the identification of oil slick signals. The improved sensor is capable of accurately identifying the refractive index difference between seawater and oil slick under adverse weather conditions such as rain, fog, smog, or insufficient and unstable light sources. Furthermore, this study employed ablation experiments to evaluate the impact of training parameters on the performance of the one-dimensional convolutional neural network model. The trained model was then applied to a human-machine interface, enabling real-time automatic recognition of whether the sensor has detected oil, thus providing an effective tool for assessing maritime oil spill situations.
The sensor is particularly suitable to use for active, real-time, and automatic oil slick detection at locations prone to oil spills. Additionally, this study discusses the impact of data preprocessing on model performance. The oil slick detection by trend analysis is possible for both the model trained with min-max normalization and the model trained without preprocessing. Moreover, the model demonstrates greater stability and generalizability across different ranges of input signals when using Min-Max normalization
The effect of Cultural Intelligence and Relational Support on Perceived Employability of international student in Taiwan: the mediation role of Self-efficacy
No full textFrom a perception of an international student, combined with the attraction in retaining foreign talents of Taiwan, through social cognitive career theory, the goal of this research is to look into what affects international students\ue2 perceived employabiliy, which occurs when students are planning for their career path, on their cultural intelligence and relational support. Moreover, this research explores the potential mediating effect of a self-assessment factor (i.e. self-efficacy) in the correlation between perceived employability and cultural intelligence. The proposed research model was tested with data collected from 198 international students studying in both long-term (bachelor, master, doctoral) and short-term (exchange,language) in Taiwan joining a structured survey. Hierarchical regression is utilized to assess the collected data. The results indicate that cultural intelligence influences international students\ue2 perception of employability, whereas the impact of relational support to perceived employability is found to be insignificant. Furthermore, the association between cultural intelligence on perceived employability is fully mediated by self-efficacy. This research is expected to add to the current literature on perceived employability while also providing pratical implications for retention strategies
Synthesis of 2D Transitional Metal Oxide/Sulfide Nanosheets by Probe ultrasonication and their Photothermal and Biosensing Applications
No full textIn the first study, a rapid, disposable, and cost-free solar steam generator device made of recyclable plastic bottles for the purification of seawater/wastewater by a solar evaporation method is demonstrated. This device utilizes the photothermal conversion of solar energy to heat using the mixture of 2D TiO2 and Ti2O3 nanosheets synthesized via the probe ultrasound followed by co-precipitation. The water purification device showed a competitive maximum evaporation rate of 1.17 kg m-2 h-1 under one sun illumination (1.0 kW m-2) and showed a percentage efficiency of 79.9 %. The real-world water samples of seawater and wastewater were demonstrated for fast purification within one minute and the resulting clean/drinking water from the device showed a significant improvement in their quality up to WHO-recommended drinking water limits.
In the second study, tungsten disulfide (WS2) nanosheets have been intensively investigated in many fields due to their high crystallinity, outstanding electrochemical properties, and structural integrity. A cutting-edge bottom-up approach with the assistance of probe ultrasonication and annealing was used to synthesize 2D WS2 nanosheets. These nanosheets were modified by probe ultrasonication-assisted doping of Eu2O3 to obtain the reddish-orange color fluorescence with a 68.32% relative quantum yield. This as-synthesized material was used to develop a quenching and shifting associated turn-off turn-on combined type bio-sensor for the detection of human catecholamines. It showed a perfect and wide linear range of 50-800 \uc2\ub5M for norepinephrine, epinephrine, and dopamine while the reported detection limits were 0.36 \uc2\ub5M, 0.91 \uc2\ub5M, and 0.15 \uc2\ub5M, respectively. It showed a great recovery of 95% for the human urine sample analysis with spiked analytes. In addition, this could be used as a drug quality assurance option for neurotransmitter drugs
The research on Job Characteristic, Psychological Empowerment and Psychological Contract Breach- A case study of Taiwanese expatriates
No full textExpatriates are the key for today's companies to expand international markets and maintain their strongholds in the face of the trend of globalization. However, there are still many companies that do not provide sufficient resources and support for their overseas personnel, or even do not Allow foreign personnel to gain a full understanding of work assignments and authorization of rights and responsibilities
; in this state, foreign personnel may be affected by factors such as the special nature of work in the country where they are assigned, the national customs of the host country, the political economy, and the company culture of the local organization. Under the catalysis, it affects the inner adjustment of personnel stationed abroad. The question this study intends to explore is the different nature of work that expatriates face in overseas units. Is the complexity the reason for the Psychological Empowerment and Psychological Contract Breach of expatriates?
And can possessing specific Personality Traits or increasing and fulfilling Organizational Commitment weaken the degree of Psychological Empowerment and Psychological Contract Breach of expatriates, thereby improving their work performance and willingness?
The main subjects of this study are Taiwanese full-time workers who have been posted or are currently working in overseas units. Questionnaires were distributed to respondents in different industries through convenience sampling. A total of 109 questionnaires were eventually collected, with a questionnaire response rate of 68.1%; and Narrative statistics, reliability analysis, correlation analysis and regression analysis were used to conduct data analysis. The research results found that:
01. There is a positive relationship between Psychological Empowerment and Psychological Contract Breach.
02. There is a positive relationship between Job Characteristic and Psychological Empowerment.
03. Psychological Empowerment partially mediates the relationship between Job Characteristic and Psychological Contract Breach.
Finally, academic and practical implications are proposed based on the results of this study, hoping to provide future suggestions and directions for subsequent research
Investigation on Electrical Analysis and Reliability Issues in Next-Generation High-Power Devices
No full textIn recent years, the demand for automotive chips has been increasing with the rise of electric vehicles and intelligent vehicles. The cost of automotive chips in the overall vehicle cost has also been increasing year by year, from 1% in 1950 to 50% in 2030. Among them, power management ICs (PMICs) play an important role in automotive electronics. They can realize DC power protection, low static leakage current, and reduce electromagnetic interference, which can effectively improve the battery life of electric vehicles and the fuel efficiency of traditional vehicles. Applications cover advanced driver assistance systems (ADAS), digital cockpits, and battery management. Therefore, the performance and reliability of PMIC-related components are very important.
Advanced PMICs mainly use bipolar-CMOS-DMOS (BCD) technology to integrate three different types of components on a single chip, and extract their own advantages and functions, including: (1) Bipolar has high operating current, high linear transfer, and low 1/f noise characteristics, which are used for signal amplification; (2) CMOS has small size, anti-interference, and energy saving, which are used for digital computing; (3) The high voltage tolerance of DMOS is used for high-power driving. In BCD technology, the ideal DMOS has low on-state resistance (Ron) and high breakdown voltage (Vbd). To reduce Ron, DMOS uses a contact field plate (CFP) structure, which connects the field plate to the gate, covers part of the drift region, and attracts minority carriers in the drift region, which can reduce the resistance of the drift region and lower the overall Ron. To improve Vbd, the reduced-surface-field (RESURF) is used to improve the withstand voltage of the drift region. The emergence of the CFP structure has improved the overall performance of LDMOS, but there are still serious problems in terms of reliability.
This work studies the reliability of LDMOS under non-conducting stress (NCS) and hot carrier instability (HCI) tests. In the NCS test, it is found that LDMOS devices will have a serious problem of linear region on-state current (Ion) decrease after a period of time. The analysis shows that the reason is in the resist protective oxide layer (RPO) of the device. In the NCS process, the electrons are injected into RPO due to the influence of the electric field, which causes the resistance of the drift region of the device to rise, and finally affects Ron, which in turn increases the power consumption of the device. Therefore, if the quality of RPO can be effectively improved, it can greatly improve the application value of LDMOS.
In the HCI test, it is found that LDMOS devices will have Ion decrease and threshold voltage (Vth) shift problems after a period of time. The analysis shows that the degradation mechanism can be divided into two stages. The first stage is the collision ionization that occurs under CFP, which leads to electron injection into RPO, resulting in Ion degradation. The second stage of degradation is the positive bias temperature instability (PBTI) effect caused by Joule heat, which causes defects at the SiO2/Si interface in the device, and injects the electrons from the channel into SiO2, thereby increasing the subthreshold swing (S.S.) of the device and causing Vth to change. Through temperature variation experiments and TCAD simulation, it is proved that HCI will indeed produce serious PBTI effects and cause device degradation in the long-term HCI process.
In the third part of this work, we study the performance improvement of fast recovery diodes (FRD) by using supercritical fluid technology. Generally, MOSFETs will generate an internal body diode due to their own structural characteristics during the manufacturing process. This diode can protect MOSFETs from collapsing when an inverse voltage is applied. However, in switch power applications, because the diodes need to be switched quickly, the dynamic characteristics of the diodes have a great impact on the overall performance of MOSFETs. In order to improve the efficiency of the device, FRD are usually connected in parallel to improve the switching efficiency. In order to achieve the condition of fast switching, FRD are doped with metal as the recombination center of the diode, so as to increase the recombination speed of minority carriers, so as to improve the switching speed. However, the addition of metal will increase the energy level in Si, which will lead to an increase in leakage current. In this section, using supercritical fluid technology to passivate the defects in the devices, so that FRD can reduce leakage current without affecting the switching characteristics. It will also use the capacitance-voltage (C-V) characteristic curve to find the location of the defects and further propose the corresponding physical model
Exploring Inter-brain Electroencephalogram Patterns through Hyperscanning for Assessing Social Cognition during Jigsaw Puzzle Solving
No full textSocial interaction is an indispensable aspect of our daily lives, consuming a significant portion of our time. Today, the emphasis on individual privacy underscores fundamental human rights, indirectly highlighting the inseparable relationship between social interaction and people. Social interaction influences our progress in various ways, either through collaborative support or competitive challenges that sharpen us. In recent years, the support of hyperscanning technology has provided a powerful new means to study the neural mechanisms of social behavior, opening a new field for neuroscience research. Hyperscanning studies the activity of multiple brains during social cooperation. In team-based learning environments, cooperation and competition are crucial aspects of social interaction. To closely observe team interactions in real-life scenarios, this study used a newly developed turn-based puzzle game, dividing tasks into three different modes: solo, cooperative, and competitive, conducted on a computer. The study aims to identify biomarkers of people's social interactions through a task paradigm closer to real-life scenarios, allowing players to engage in tasks in a relaxed and lifelike manner through a game-like approach.
By collecting signals through hyperscanning electroencephalography, the signals were preprocessed and analyzed in the time domain, time-frequency domain, and synchronous connectivity analysis. Different aspects of the EEG signals were observed in different modes. The results from the single-player mode validated the effectiveness of the study paradigm, including the observation of P300 component waves and feedback negativity in event-related potentials and event-related spectral perturbations. Participants exhibited stronger responses in the occipital lobe under the correct condition and feedback negativity in the central frontal lobe under the error condition.
In the two-player mode, we observed differences between competitive and cooperative social activities, finding that in the cooperative mode, the P300 component wave decayed more slowly compared to the competitive mode in event-related potentials. In event-related spectral perturbations, cooperation showed significantly greater \uce\ub4 and \uce\ub8 synchronization responses in the frontal and occipital lobes compared to competition, indicating a tendency toward cooperation. Furthermore, in the \uce\ub4, \uce\ub8, and \uce\ub1 bands, the inter-brain synchronicity increased in the competitive mode compared to the cooperative mode, indicating a higher utilization of attention and cognitive resources.
Surprisingly, in the synchronous connectivity analysis, the improvement in focus under competitive conditions led to denser intra-brain synchronization compared to cooperative conditions. Equally unexpectedly, the inter-brain synchronization between the two participants was also denser under competitive conditions compared to cooperative conditions. In turn-based gaming tasks, competitive situations were more likely to cause brain synchronization, as each correct move by the opponent created competitive pressure on the player. The competitive mode required more attention and cognitive resources, focusing on analyzing the situation on the screen to gain an advantage.
After various analyses and feature extraction, attempts were made to distinguish high autism tendencies in generally healthy subjects, yielding balanced accuracy rates of 72.8% and a root mean square deviation of 0.26, respectively, from support vector machines and support vector regression