04. Trường Đại học Khoa học tự nhiên (University of Science)
Tài nguyên nội sinh
Các bộ sưu tập số Luận văn Thạc sĩ, Luận án Tiến sĩ, Đề tài Báo cáo khoa học… của Thư viện Trường ĐH. Khoa học tự nhiên – ĐHQG-HCM được xây dựng nhằm hỗ trợ cho hoạt động giảng dạy và nghiên cứu của Nhà trường.
Duyệt 04. Trường Đại học Khoa học tự nhiên (University of Science) theo Kiểu tài liệu "Bài trích"
Ấn phẩm
A closed-form solution for a queueing model of energy efficient Ethernet links
To save energy consumption of Ethernet switches, IEEE has standardized a new energyeffient operation for Ethernet links with a low-power state and transition mechanisms between the high-power state for transporting traff and the low-power state. In this paper, we propose a queueing model with the Markov Modulated Compound Poisson Process that is able to characterize backbone packet traffi We derive a closed-form solution for the stationary distribution of the proposed queueing model. We show that our model can capture an entire system where the transition times are constant.
Ấn phẩm
A comparison of benzene-core PCF optical characteristics with hexagonal and circular lattices
This work compares the optical characteristics of benzene photonic crystal fiber with hexagonal lattice (H-PCF) and circular lattice (C-PCF). The difference in the radius between the rings is used to optimize effective area, nonlinearity, attenuation, and chromatic dispersion simultaneously. A series of numerical studies show that the effective mode area and attenuation are small for C-PCF and large for H-PCF geometries. That is the reason why the nonlinear coefficient of H-PCF is larger. Meanwhile, flat dispersion over a wide wavelength range is achieved mostly with H-PCFs in both dispersion modes. Our optimized fibers with near-zero flat dispersion, small loss, and large nonlinearity can enhance supercontinuum generation efficiency to generate a broad spectrum with high coherence.
Ấn phẩm
A comprehensive review of rock dust for soil remineralization in sustainable agriculture and reliminary assessment of nutrient values in micronized porous basalt rock from Nghe An province, Viet Nam
This review examines the use of rock powder as a mineral-derived fertilizer to promote soil remineralization in sustainable agriculture. Soil remineralization has emerged as a key strategy for ensuring long-term soil fertility and reducing reliance on synthetic fertilizers. This review critically assesses the methodological approaches employed in various studies, considering factors such as local rock mineralogy, soil conditions, crop types, and plant nutrient uptake. The review encompasses methods such as nutrient value analysis, nutrient release rate assessment, and crop cultivation experiments conducted in both laboratory and field settings, as well as implications for soil remineralization. Additionally, we present findings from a preliminary study on the nutrient content of basalt rock from Nghe An Province, Vietnam. This investigation focuses on assessing the nutrient values in a specific porous basalt material processed into micro- and nanometer-sized particles using a high-energy ball mill, thereby enhancing the efficiency of nutrient release. The study employs various analytical techniques, including XRD, XRF, SEM/EDS, and ICP-MS, to quantify macro- and micro-nutrient compositions. While basalt rock samples may exhibit limitations in providing macronutrients (like K) compared to commercial azomite rock, they offer advantages in supplying secondary macronutrients (such as Ca). This analysis provides insights into the potential use of micronized basalt rock for soil remineralization.
Ấn phẩm
A critical review on factors affecting the two-stage anaerobic digestion of biodegradable solid waste
Anaerobic digestion (AD) technology has attracted considerable attention from the scientific community and has become a crucial component of sustainable solid waste management systems. It is the only biotechnological process capable of converting highmoisture biomass waste into energy through a series of biochemical steps, including hydrolysis, acidogenesis, acetate production, and methane generation. Notably, the two-stage anaerobic digestion (TAD) process, which separates methanogenesis and hydrolysis into two distinct reactors, offers significant advantages over conventional methods. The performance of AD systems is influenced by various factors, including growth conditions (such as carbon-tonitrogen ratio, pH, and temperature), operational parameters (such as retention time and organic loading rate), feedstock pretreatment, and potential inhibitors. While these aspects have been widely studied in single-stage anaerobic digestion systems, research on their impact in TAD systems remains limited. This study aims to provide a comprehensive review of the factors affecting TAD systems. It synthesizes the latest research findings from recent years and discusses optimal operating conditions to enhance TAD performance.
Ấn phẩm
A high effective CuInS2 photocatalyst for oxidative desulfurization of dibenzothiophene in fuel oil
In this study, CuInS2 materials were synthesized by hydrothermal method. The effect of different reaction times was investigated. The structure and morphology of the materials were characterized by XRD, SEM, TEM, UV-Vis and EDX measurement methods. The obtained results showed that with a reation time of 24 h, the CuInS2 sample (CIS-24) had an average particle size of about 50 nm and a band gap energy value of Eg = 1.48 eV, with the ability to strongly absorb visible light. The photocatalytic activity of CIS-24 was investigated by the degradation of dibenzothiophene under visible light irradiation. The photocatalytic efficiency reached a maximum of 97.9 % at an initial concentration of 500 ppm after 5 h of reaction at 70 oC with a catalyst dosage of 50 mg and 1.0 mL of H2O2 as an oxidant.
Ấn phẩm
A high-efficiency positive buck–boost converter with three-mode selection circuit and soft start circuit
Positive Buck-Boost converters are considered the optimal choice for battery-powered applications that require high power conversion efficiency and extend battery life. The converter offers versatility in processing input voltages over a wide range thanks to its three operating modes: Boost, Buck, and Buck-Boost (Bck-Bst) while maximizing battery capacity. The efficiency of the converter can be severely affected by a severe ripple in the output voltage and current when transitioning between modes. To reduce the output instability and output ripple during operation, a three-mode selection circuit that uses a very small delay between mode transitions is proposed to minimize the above effects. The proposed chip was designed and implemented on CMOS 0.18 µm process. In addition, a high peak efficiency of 95.6 % can be achieved under the conditions of a wide input range of 2.5 V - 5 V.
Ấn phẩm
A hybrid model using the pretrained bert and deep neural networks with rich feature for extractive text summarization
Deep neural networks have been applied successfully to extractive text summarization tasks with the accompany of large training datasets. However, when the training dataset is not large enough, these models reveal certain limitations that affct the quality of the system’s summary. In this paper, we propose an extractive summarization system based on a Convolutional Neural Network, an Encoder-Decoder, and a Fully Connected network for sentence selection. The pretrained BERT multilingual model is used to generate embedding vectors from the input text. These vectors are combined with TF-IDF values to produce the input of the text summarization system. Redundant sentences from the output summary are eliminated by the Maximal Marginal Relevance method. Our system is evaluated with both English and Vietnamese languages using CNN and Baomoi datasets, respectively. Experimental results show that our system achieves better results compared to existing works using the same dataset. It confims that our approach can be effctively applied to summarize both English and Vietnamese languages.
Ấn phẩm
A hybrid multi-constraint lagrangian relaxation based aggregated cost based segment routing in qos aware software defined networks
The Industrial Internet is a cornerstone of Industry 4.0, emphasizing the need for high reliability and low latency in network control. It enables the integration of diverse IoT sensing and actuation devices for efficient monitoring and management of industrial systems. SoftwareDefined Networking (SDN), a modern network architecture, offers centralized control by logically separating the control and data planes. This global view facilitates flexible network administration, optimized performance, and enhanced adaptability. However, meeting the increasing demands for efficient and dependable network services in SDNs has made Quality of Service (QoS) assurance a critical challenge. This study introduces a novel approach Hybrid Multi-constraint Lagrangian Relaxation-based Aggregated Cost (HMLR-AC) Segment Routing for QoS-aware routing in SDNs. HMLR-AC combines the strengths of Lagrangian Relaxation (LR) and Aggregated Cost (AC) to enhance routing decisions. Additionally, it incorporates an H-permissible Paths Routing Scheme (HPRS) that effectively routes traffic within cardinality constraints. Utilizing real-time traffic insights and a global network perspective, HMLR-AC dynamically adapts routing strategies. Simulation results demonstrate notable improvements in QoS parameters, proving HMLR-AC as a promising solution for future QoS-driven SDN infrastructures.
Ấn phẩm
A new forestry bacterial isolate Brevundimonas sp. negatively affecting on Caenorhabditis elegans
Bacteria are food sources for the Caenorhabditis nematodes. This interaction becomes a model to understand the effects of bacteria on the nematodes and other host organisms. In this research, we report the identification of the environmental Brevundimonas sp. CFBb114 in Cat Tien National Park with the 16S rDNA. CFBb114 inhabits within the microhabitats of the nematode genus Caenorhabditis. We fed Caenorhabditis elegans with CFBb114 to investigate the effects of CFBb114 on C. elegans, and found that the bacteria reduced longevity and reproduction and changed the behavior of C. elegans. This research will facilitate the study of how host organisms evolve mechanisms against the impacts of bacteria.
Ấn phẩm
A novel weighted ensemble approach for enhancing image retrieval effectiveness with deep learning models
Content-based image retrieval (CBIR) is becoming increasingly important amid the rapid growth of image data. Traditional CBIR approaches, which rely on features such as color, shape, and texture, often face limitations in accuracy. Even when using features extracted from deep learning models, these approaches still fall short of fully meeting user expectations. To enhance retrieval effctiveness, this study introduces an ensemble approach that utilizes feature sets from multiple deep learning models. In our method, retrieved images are determined through an aggregation of recommendations from deep learning models, with each model’s vote assigned a specifi weight. This weight is comprehensively evaluated based on the similarity between the recommended image and the query, the model’s reliability, and the distribution of images recommended by each model. To validate the effctiveness of this approach, we conducted experiments using VGG16, ResNet50, EffientNetB0, DenseNet201, Swin, and Clip, pre-trained on ImageNet for feature extraction. Three model combinations, (1) VGG16, ResNet50, and EffientNetB0, (2) ResNet50, EffientNetB0, and DenseNet201, and (3) Swin and Clip, were explored within the proposed ensemble framework on the Oxford-17-Flowers, Caltech-101, CIFAR-10, and ISIC-2018 datasets. The suggested approach routinely outperforms individual models, according to experimental results, providing better retrieval accuracy on most datasets.
Ấn phẩm
A proto-brown dwarf candidate in ρ ophiuchus
Brown dwarfs with masses below 0.075 solar masses are thought to form like lowmass stars (e.g., the Sun). However, it is still unclear how the physical formation processes occur in brown dwarfs at the ealiest stages (i.e., proto-brown dwarfs) of their formation. Up to date, only a few proto-brown dwarfs have been detected. The detection of proto-brown dwarfs offers us excellent benchmarks to study the formation process of brown dwarfs, and thus understand their formation mechanism. In this paper, we present our identification of a proto-brown dwarf candidate in the star-forming region ρ Ophiuchus. The candidate shows a small-scale bipolar molecular outflow that is similar to the outflows observed in other young brown dwarfs. The detection of this candidate provides us with additional important implications for the formation mechanism of brown dwarfs.
Ấn phẩm
A qualitative investigation of electrical conductivity in three-layer graphene structures
We study the interlayer distance d effect on the electrical conductivity σ1, σ2, σ3 of a three-layer monolayer graphene system (3MLG) made of parallel-placed monolayer graphene layers at T=0 K by two methods: the fist based on the relation between the impurity - electron effective interactions W11(d), W22(d), W33(d) and σ1(d), σ2(d), σ3(d); the second being traditional calculations. We pay attention to the fist method consisting of three steps. In the fist step, we calculate W11, W22, W33 by using the multi-component random phase approximation (M-RPA). In the second step, we defie the dependence of W11, W22, W33 on interlayer distance d, dielectric constant ε1, ε2, ε3, ε4, and temperature T in the case of a 3MLG system. In the third step, we identify the proportional relations between W11(d), W22(d), W33(d) and σ1(d), σ2(d), σ3(d) which were done in the framework of the semiclassical Boltzmann theory and the relaxation time approximation. Based on the obtained results, we deduce the rules of change of σ1, σ2, σ3 when d varies which after that, are checked by the second method.
Ấn phẩm
A rectangle-Quartet metamaterial for dual-band perfect absorption in the visible region
Based on rectangle-shaped structures, we create a dual-band metamaterial perfect absorber (DMPA) in the optical region. The independent-polarization absorption is a significant advantage as well as the simple integrated progress for constituent materials. In particular, absorption can be obtained to be over 90% in a bandwidth of 140 THz (from 608 THz to 748 THz), which is still remained well in the oblique incident angles for the TE-polarization. Our results can be regarded as the groundwork for the near future, applications such as photodetectors, energy converters, sensitive inspection technologies, and more.
Ấn phẩm
A study on the spawning season of 3 Acropora species in Nha Trang bay, Southern waters of Vietnam
Specimens for the study on coral spawning of three species were collected at two sites in Nha Trang bay, South Vietnam, in 2014, 2015, 2016, 2018 and 2019. The determination of spawning seasons was based on observations of gonadogenesis development of A. florida and A. robusta and variation of egg average sizes of these two species and A. hyacinthus. Data analysis and comparative discussions allow us to assume that the 3 Acropora species exhibited single cycle spawning annually and their spawning period occurred in March/April between the full moon and crescent moon. However, their maturity periods were not the same, starting before and ending after the full moon for A. florida and A. robusta but starting after the full moon and lasting until the crescent moon for A. hyainthus. Further studies are needed to improve the understanding of coral spawning and support coral restoration using sexual reproduction.
Ấn phẩm
A time series forecasting model based on linguistic forecasting rules
The fuzzy time series (FTS) forecasting models have been studied intensively over the past few years. The existing FTS forecasting models partition the historical data into subintervals and assign the fuzzy sets to them by the human expert’s experience. Hieu et al. proposed a linguistic time series by utilizing the hedge algebras quantifiation to converse the numerical time series data to the linguistic time series. Similar to the FTS forecasting models, the obtained linguistic time series can defie the linguistic, logical relationships which are used to establish the linguistic, logical relationship groups and form a linguistic forecasting model. In this paper, we propose a linguistic time series forecasting model based on the linguistic forecasting rules induced from the linguistic, logical relationships instead of the linguistic, logical relationship groups proposed by Hieu. The experimental studies using the historical data of the enrollments of University of Alabama and the daily average temperature data in Taipei show the outperformance of the proposed forecasting models over the counterpart ones. Then, to realize the proposed models in Viet Nam, they are also applied to the forecasting problem of the historical data of the average rice production of Viet Nam from 1990 to 2010.
Ấn phẩm
A unified approach to exponential stability analysis for a general class of switched time-delay linear systems
This paper proposes a unifid approach to study global exponential stability for a class of switched time-delay linear systems described by general linear functional diffrential equations. Several new delay-dependent criteria of global exponential stability are established for these systems over the sets of switchings satisfying the assumption on the minimum dwell time or having the average dwell time. As particular cases, the obtained results are shown to include and improve many previously known results. An example is given to illustrate the proposed method.
Ấn phẩm
Adversarial attack and defense in AI-powered intrusion detection
Content-based image retrieval (CBIR) is becoming increasingly important amid the rapid growth of image data. Traditional CBIR approaches, which rely on features such as color, shape, and texture, often face limitations in accuracy. Even when using features extracted from deep learning models, these approaches still fall short of fully meeting user expectations. To enhance retrieval effctiveness, this study introduces an ensemble approach that utilizes feature sets from multiple deep learning models. In our method, retrieved images are determined through an aggregation of recommendations from deep learning models, with each model’s vote assigned a specifi weight. This weight is comprehensively evaluated based on the similarity between the recommended image and the query, the model’s reliability, and the distribution of images recommended by each model. To validate the effctiveness of this approach, we conducted experiments using VGG16, ResNet50, EffientNetB0, DenseNet201, Swin, and Clip, pre-trained on ImageNet for feature extraction. Three model combinations, (1) VGG16, ResNet50, and EffientNetB0, (2) ResNet50, EffientNetB0, and DenseNet201, and (3) Swin and Clip, were explored within the proposed ensemble framework on the Oxford-17-Flowers, Caltech-101, CIFAR-10, and ISIC-2018 datasets. The suggested approach routinely outperforms individual models, according to experimental results, providing better retrieval accuracy on most datasets.
Ấn phẩm
Aggregation of Zoospores on Sharklet Microtopographic Surfaces
Surfaces with engineered microtopographies are potential candidate against biofouling to replace the use of biocides in the marine environment. Understanding the antifouling mechanism of microtopographic surfaces against marine microorganisms, however, has been limited. In this work, we theoretically studied the aggregation of Ulva linza zoospores on the Sharklet topographic surfaces by employing the extended Surface Energetic Attachment (SEA) model proposed in a previous work. The energy parameters of the model were obtained by matching theoretical results with experimental data for one type of Sharklet surface. Monte Carlo simulations were then carried out for a series of Sharklet surfaces with various numbers of distinct features. In agreement with prior experimental results, our simulations indicate that engineered topographies promote smaller aggregates than those on a smooth surface. Furthermore, we show that the maximum effect of the Sharklet topography on the aggregate size of U. linza can be obtained with just 3 distinct features.
Ấn phẩm
Ambiguities from nuclear interactions in the 12C (p , 2p )11B reaction
We investigate the impact of ambiguities coming from the choice of optical potentials and nucleon-nucleon scattering cross sections on the spectroscopic factors extracted from the 12C(p , 2p)11B reaction. These ambiguities are evaluated by analyzing the cross sections of the 12C(p , 2p)11B reaction at 100 and 200 MeV within the framework of the distorted-wave impulse approximation with realistic choices of nuclear inputs. The results show that the studied ambiguities are considerably large in this energy region and careful choices of nuclear inputs used in the reaction calculations are required to extract reliable structure information.
Ấn phẩm
An initial study on ocean acidification in Southern waters of Vietnam
Ocean acidification (OA) refers to the increase of dissolved CO2 and the reduction in the pH of seawater as a consequence of the absorption of large amounts of carbon dioxide (CO2) by the oceans. This process is the result of large quantities of CO2, produced by vehicles and industrial and agricultural activities. Over the past decades there have been many worldwide studies focusing on potential impacts of OA. However, researches regarding this issue remain scarce in Vietnam. In this paper, data of pH, total alkalinity (TA), dissolved inorganic carbon (HCO3-, CO32-, CO2), partial pressure of CO2 (pCO2) and the state of aragonite saturation (Ωar) measured in Southern waters of Vietnam in 2018 were used to: (1) Provide the initial data of OA parameters in Southern waters of Vietnam; (2) Compare the current situation of OA in Southern waters of Vietnam with the situation of world oceans. The results showed that mean values of pH, TA and CO32- concentrations were 8.04 (7.92–8.11), 2300.28 µmol/kgSW (2,144.10–2,523.15), 218.83 µmol/kgSW (151.32–262.83), respectively. These values were higher in offshore areas than in coastal areas, especially at the estuaries. The average value of pCO2 was 414.47 µatm (327.93–568.59), higher when compared with that of other areas (370 µatm). On the other hand, the state of aragonite saturation of the studied area had the similar patterns of TA and CO32- concentrations. Most of values were always greater than 3, with this saturation state, the marine calcifiers are more likely to survive and reproduce.
Ấn phẩm
An integrated avoiding and approaching human systems for mobile service robots in dynamic social environments
In this study, we present a socially aware navigation framework for mobile service robots by integrating the systems proposed in our previous studies into a completed mobile robot navigation system, including the human detection and tracking system that is used to detect and track humans; the encoder and laser-based localization and mapping system that is used to determine the position of the robot in the environment; the approaching pose estimation system that is used to estimate the approaching pose of the robot to the human; and the social timed elastic band-based motion planning system that is used to socially navigate the mobile robots to approach the human. In addition, in the paper we describe in detail the motor control system and the design of our mobile robot platform, which is then utilized to conduct experiments in the real-world environments. We verify the feasibility and usefulness of the proposed socially aware robot navigation framework through a series of experiments in a corridor-like environment. The experimental results show that our proposed framework is able to drive the mobile robots to both avoid and approach the humans, providing the safety and comfort for the humans and socially acceptable behaviors for the mobile service robots in the dynamic social environments.
Ấn phẩm
An optimal segmentation method for processing medical image to detect the brain tumor
In the field of medical physics, detection of brain tumors from computed tomography (CT) or magnetic resonance (MRI) scans is a difficult task due to complexity of the brain hence it is one of the top priority goals of many recent researches. In this article, we describe a new method that combines four different steps including smoothing, Sobel edge detection, connected component, and finally region growing algorithms for locating and extracting the various lesions in the brain. The computational algorithm of the proposed method was implemented using Insight Toolkit (ITK). The analysis results indicate that the proposed method automatically and efficiently detected the tumor region from the CT or MRI image of the brain with an accuracy above 90%. This result is comparable to physician’s manual practice in clinical and computational software in medicine. It is very clear for physicians to separate the abnormal from the normal surrounding tissue to get a real identification of related areas; improving quality and accuracy of diagnosis, which would help to increase success possibility by early detection of tumor as well as reducing surgical planning time. This is an important step in correctly calculating the dose in radiation therapy later.
Ấn phẩm
Analysis of Temperature-dependent Extended X-ray Absorption Fine Structure Oscillation of Distorted Crystalline Cadmium
In this paper, the temperature-dependent extended X-ray absorption fine structure (EXAFS) of distorted crystalline cadmium has been analyzed using an efficient calculation-model. The analysis procedure is based on evaluating the influence of temperature on the phase shift and amplitude reduction of EXAFS oscillation that is expressed in terms of the EXAFS Debye-Waller factor. The anharmonic EXAFS cumulants are calculated by expanding the anharmonic correlated Debye model based on the anharmonic effective potential that depends on the structural characteristics of distorted crystalline cadmium. The numerical results satisfy well with those obtained using the experimental data and other models at various temperatures. The obtained results indicate that this theoretical model is useful for calculating and analyzing the experimental EXAFS data of distorted crystalline metals.
Ấn phẩm
Analysis of wave spectrum by Blackman-Tukey method and fast fourier method
Energy of surface waves is almost dissipated as propagating from shallow water to muddy flat and in mangrove forest. The study aims to analyze energy spectral density by using Blackman-Tukey (BT) and Fast Fourier Transform (FFT) methods in order to analyze the wave energy in mangrove areas in Cu Lao Dung (Soc Trang province). BT method is easy to use especially in short time series. Selection of lag number m is very important to determine the energy amount and number of peaks. Whereas, FFT method helps us to analyze the shift of spectral energy as waves propagate into shallower water. The results show that wave energy is dissipated from shallow water to muddy flat and into mangrove forests. The spectral energy shifts from low frequency to higher frequency as propagating into mangrove forests. This can prove the non-linear characteristics of waves in mangrove forests and the complicated hydrodynamic processes in mangrove forests.