Journal of Military Science and Technology

Non-invasive systems for sleep monitoring: Respiratory rate and sleep posture classification

Tran Duc Nghia, Vi Manh Tuyen, Pham Quang Huy, Tran Duc Tan — Wed, 25 Jun 2025 00:00:00 +0000

Sleep plays an essential role in human health. Monitoring sleep has increasingly become an important tool for gaining deeper insights into sleep behavior and detecting related health issues. Polysomnography (PSG) in clinical settings is the gold standard for sleep analysis; however, it is expensive and challenging to implement over long periods. As a result, home-based sleep monitoring methods, particularly non-invasive sensor-based systems, are gaining significant attention. This paper focuses on reviewing recent studies related to non-invasive sleep monitoring systems, including both wearable and non-wearable methods. These systems are designed to continuously measure and monitor users' breathing patterns while also detecting and classifying their sleep postures. Additionally, the paper explores future directions for developing respiratory monitoring and sleep posture classification systems that can operate flexibly across different environments, including settings outside professional medical facilities.

Determination of the kinetic parameters of the projectile when penetrating the wooden target by finite element methods

Phan Ba Binh, Phan Van Vu — Wed, 25 Jun 2025 00:00:00 +0000

The inertial force that occurs when a projectile impacts or penetrates depends on various factors, such as the diversity and heterogeneity of the physical and mechanical characteristics of the obstacle, as well as the structural features of the projectile. These factors are not calculated accurately but need to be combined with actual testing to determine. Many studies have been conducted on the process of projectiles penetrating into soil, rock, concrete, as well as research on projectiles penetrating steel plates. However, up to now, studies on bullets penetrating wooden targets are still limited. This is because, in reality, wooden targets are not common battlefield targets. In fuze tests, wooden targets are widely used to ensure safety and standardisation during testing, but the acceleration values obtained to activate the fuze have not been quantified. Based on this reality, this article investigates the applicability of formulas for the process of projectiles penetrating steel targets in order to apply them when calculating the penetration of projectiles into wooden targets. The simulation results identified the critical coefficient when projectiles penetrate into oak wood targets, with projectile sizes ranging from 57 to 100 mm, velocities ranging from 500 ÷ 1000 m/s, and wood thicknesses ranging from 10 to 50 mm, with values ranging around 1320÷1825. The test firing results for a specific case (76mm projectile, 30mm thick target) do not deviate much from the simulation.

Modeling and dynamic analysis of a battlefield water filtration station considering road surface profiles

To Viet Thanh — Wed, 25 Jun 2025 00:00:00 +0000

During battlefield operations, mobile water filtration stations are used to supply clean water to military units, one of which is the VFS-2.5 filtration station. When in motion, it tows a trailer carrying a generator and chemical tanks for water filtration. The dynamic model, constructed as a multi-body system, is developed as a planar model with a two-axle truck as the towing vehicle. The study considers the elasticity of the suspension system, tires, and trailer hitch while neglecting the influence of road slope. Based on the dynamic model, the Lagrange equation of the second kind is used to establish the system of differential equations describing the motion of the system. The research model can be utilised to evaluate the stability of the water filtration station, and the aim is to improve the suspension system on the generator trailer to minimise the vibrations of the assembly. This is a highly significant issue in the field of national defence and security.

Effect of sintering temperature and time on surface hardness of cold-pressed SrFe₁₂O₁₉ pellets: An orthogonal experimental approach

Mai Van Huy, To Thanh Loan, Tran Thi Viet Nga, Nguyen Manh Tien, Tran Duc Hoan — Wed, 25 Jun 2025 00:00:00 +0000

This study explores the influence of sintering temperature and time on the surface hardness and microstructure of cold-pressed SrFe₁₂O₁₉ permanent magnet compacts. A two-factor orthogonal experimental design was employed to construct a regression model with high predictive accuracy (R² = 97.23%, AICc = 55.14). Rietveld refinement of X-ray diffraction (XRD) data confirmed the formation of a single-phase of M-type hexagonal structure, with crystallite sizes ranging from 39.5 to 61.8 nm and microstrain values dependent on sintering conditions. The optimal hardness of 93.63 HV was obtained at 1000 °C for 120 minutes, closely matching the model prediction (95.287 HV) with a deviation of only 1.74%. The results demonstrate that controlling sintering parameters is crucial for minimizing microstrain and maximizing hardness, offering practical insights for the processing of high-performance ferrite-based magnetic ceramics.

Research on the influence of cutting parameters and optimization of spherical surface form error in ultra-precision turning using Box-Behnken and Genetic Algorithm

Wed, 25 Jun 2025 00:00:00 +0000

Ultra-precision diamond turning is a crucial method in manufacturing optical components and precision mechanical parts, especially for machining spherical surfaces. However, the form error of the machined surface is influenced by various factors such as spindle speed, depth of cut and feed rate. This study analyzes the effects of cutting parameters on form error when machining spherical surfaces on an ultra-precision lathe. Fifteen experiments were conducted with machining parameters, including spindle speed, feed rate, and depth of cut, within the recommended range of the lathe. A form error model was developed based on the Box-Behnken model to simulate and evaluate the accuracy of the machined surface. Experimental analysis shows that all machining parameters significantly affect form error. Increasing or decreasing both spindle speed and depth of cut reduces form error, while increasing feed rate generally leads to higher form error. The Genetic Algorithm (GA) was applied to optimize cutting parameters, achieving a minimum form error of 0.846 µm with an optimal spindle speed of 2000 rpm, feed rate of 5 µm/min, and depth of cut of 8 µm. This study develops a predictive model for form error in diamond turning of spherical surfaces and optimizes cutting parameters. It also improves understanding of how machining parameters affect form error, aiding process prediction and optimization for better machining quality in single-point diamond turning (SPDT).

Method of commanding with amplitude pulses for a single-channel missile controller

Pham Cong Tu, Nguyen Quang Vinh, Phan The Son — Wed, 25 Jun 2025 00:00:00 +0000

Single-channel missile controlling is a method of using a rudder to control the spinning missile to change both pitch and yaw. In essence, the single-channel missile (SCM) control system still follows the dynamic equations of the general aircraft control system, except that the rudder angle signal will simultaneously create control force in both pitch and yaw. Therefore, the method of commanding the single-channel missile rudder has its own characteristics and plays a role in distinguishing the single-channel missile control system from other aircraft control systems. The commanding methods that have been proposed are ON-OFF commanding and continuous sinusoidal signal commanding. This paper presents a method for establishing a single-channel missile control command using the amplitude pulse method. By approaching the desired total control force according to the error, the rudder angle changes according to the amplitude pulse law. This commanding method can be applied to similar lines of rudders, making it more convenient in modern control.

Adaptive sliding mode control of a three-mass elastic electromechanical system under parameter uncertainty and external disturbances

Nguyen Viet Phuong, Nguyen Duy Khanh — Wed, 25 Jun 2025 00:00:00 +0000

This paper enhances the control quality of a three-mass elastic electromechanical system under parameter uncertainties and external disturbances. For the control system design, the paper considers the control object as a three-mass elastic electromechanical system with three interdependent control loops. A complete mathematical model of the object is then established. A robust adaptive sliding controller is proposed for the three-mass elastic system to counteract the influence of unknown external disturbances and model uncertainties caused by parameter uncertainty. The highlight of the adaptive sliding controller is its mechanism for adaptively adjusting the controller gain by estimating the upper bound of the combined external disturbances and system uncertainties. Thus, the controller ensures compensation for the influence of the above-mentioned negative factors. The stability of the proposed adaptive control system is investigated using Lyapunov stability theory. Simulation results on MATLAB/Simulink comparing the performance of the proposed adaptive sliding mode controller with the conventional PID controller demonstrate the effectiveness of the proposed approach.

A crypto-coding method based on punctured turbo codes for wireless communication systems

Hoang Thi Phuong Thao, Dinh Van Linh — Wed, 25 Jun 2025 00:00:00 +0000

Encryption and error correction play important roles in secure and reliable wireless communication systems. However, these two functions are performed independently in two different layers which consume a lot of resources. In this work, a crypto-coding method is proposed for these two functions in a single step to reduce the computational complexity and hardware structure. The proposed crypto-coding method is implemented by using a secret key to control the punctured block of the Turbo codes. The secret key is generated from the wireless channel characteristics of single-input single-output (SISO) systems, which ensures the randomness according to the National Institute of Standards and Technology (NIST) version 800-22REV1A. The simulations are executed through the Additive White Gaussian Noise (AWGN) and Rayleigh channels. The simulated results show that the generated keys satisfy the randomness according to the NIST standard, which is evaluated by p -values below 0.01. In addition, the proposed crypto-coding method effectively improves the error correction ability and provides the same computational complexity as the conventional Turbo codes. Meanwhile, this method does not require any extra hardware to manage and distribute secret keys.

X-band bandpass filter structure for passive radar using SIW technology

Khuong Dinh Dung, Nguyen Dinh Tinh, Nguyen Tuan Hung, Vo Thanh Hung — Wed, 25 Jun 2025 00:00:00 +0000

This paper proposes an X-band bandpass filter structure for the Kolchuga passive radar system as an alternative to the Interdigital structure. The proposed structure uses substrate-integrated waveguide (SIW) technology with series-connected resonant frames arranged in a folded form. With the improved structure, the designed filter ensures the insertion loss, reflection coefficient in the pass band, and attenuation slope in the stop band requirements in the X-band for the Kolchuga radar while optimizing the size. Compared with the SIW bandpass filter using the conventional straight IRIS structure, the proposed filter structure reduces the size but achieves equivalent scattering parameters. The simulation results demonstrate the effectiveness of the proposed filter structure in meeting the required criteria and optimizing the size.

Multi-period power flow analysis in distribution systems with distributed generation using the nonlinear programming model

Pham Nang Van, Nguyen Thi Hoai Thu — Wed, 25 Jun 2025 00:00:00 +0000

The traditional power flow analysis of power systems is usually conducted over a single time period. However, many problems in the power systems, such as determining energy loss and optimizing the location and capacity of shunt capacitors, require performing power flow analysis over multiple time periods. This paper presents a method based on nonlinear programming (NLP) to calculate multi-period power flow (MPF) for a distribution grid with distributed generation (DG). This nonlinear optimization model features a constant objective function, with constraints defined by a system of power balance equations and predetermined values for the voltage magnitude and phase angle at the slack bus. The NLP model is programmed using the GAMS language, and the solution is computed using the KNITRO optimization solver. The proposed multi-time period power flow analysis method was evaluated on the distribution grid in Luc Ngan district, Bac Giang province, with 54 nodes and 96 time periods. The calculation results show that the solution of the proposed approach has a very small error compared to the traditional Newton-Raphson technique. At the same time, distributed generation units have a significant impact on energy loss and voltage profile in the power distribution network.

Improving the accuracy of head rotation angle estimation using inertial micro-electro-mechanical sensors and adaptive digital filters

Ha Ngoc Khoan, Tran Van Nghia, Le Ky Bien — Wed, 25 Jun 2025 00:00:00 +0000

The paper proposes a method to enhance the accuracy of head rotation angle estimation by integrating micro-inertial measurement sensors and digital filtering. The method is implemented under a stable gaze condition. Experimental results show that, compared to Mahony and Kalman filters, the proposed approach significantly improves the accuracy of head rotation angle estimation, with an RMS error of less than 0.1 degrees in horizontal plane rotation while maintaining real-time performance. These results are well-suited for applications in vestibulo-ocular reflex (VOR) signal recording for vestibular disorder diagnosis.

Adsorption and desorption of berberine chloride on kaolinite material

Vu Minh Trang, To Phuong Linh — Wed, 25 Jun 2025 00:00:00 +0000

Berberine is a very popular drug in Vietnam for intestinal diseases. Research on improving berberine purification is of interest due to the high demand for berberine in current drug consumption. Instead of using extraction processes with the presence of sulfuric acid, berberine was extracted by berberine adsorption from diluted NaOH solution with pH = 11 on the surface of kaolinite material and desorption with dilute HCl solution containing 5% NaCl with pH = 5 at 50 ^oC. The adsorption and desorption efficiencies of berberine under optimal conditions reached 91% and 92%, respectively, when using kaolinite absorbent content at 2000 ppm for 1 hour. This purification method was successfully applied to the medicinal plant sample of Coscinium fenestratum (Vàng đắng in Vietnamese) with a berberine total recovery efficiency of 81% and berberine content of 94%. The NMR spectrum confirmed the molecular structure of berberine.

Synthesis and structural characterization of CoMn₂O₄ binary metal oxide from metal-organic frameworks

Nguyen Van Bang, Nguyen Thi Phuong — Wed, 25 Jun 2025 00:00:00 +0000

The CoMn₂O₄ binary metal oxide with a spinel structure was synthesized from metal-organic framework (MOF) prepared using both microwave-assisted and hydrothermal methods. The time to synthesize MOF by hydrothermal method is much longer than that of the microwave-assisted method. Synthesized materials were characterized by X-ray diffraction (XRD), fourier-transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM), and Brunauer-Emmett-Teller (BET) surface area. The results showed that the CoMn₂O₄ oxide had high purity and a spherical nanocrystalline morphology with particle sizes ranging from 30 nm to 50 nm. Notably, the specific surface area of the material derived from MOF synthesized using the microwave-assisted method reaches 92,40 m²/g, significantly higher than that of the material synthesized by the hydrothermal method, which is 61,84 m²/g This demonstrates that the microwave-assisted method not only reduces synthesis time but also enhances the specific surface area of the material, which is a crucial factor for applications in catalysis, energy storage, and environmental treatment.

Synthesis of phosphate-based binders from aluminum slag for heat-resistant inorganic coatings

Le Manh Cuong, Lai The Vinh — Wed, 25 Jun 2025 00:00:00 +0000

This study presents the synthesis and optimization of a polyaluminum phosphate (PAP) binder using aluminum slag as the primary source of Al³⁺. The binder was synthesized via acid–base reactions between aluminum slag and phosphoric acid (H₃PO₄), with varying Al/P molar ratios, temperatures, and stirring speeds. Optimal conditions were identified at 30 wt% slag and 70 wt% H₃PO₄, 60 °C, and 800 rpm, producing a homogeneous gel with a viscosity of 101 mPa·s, specific gravity of 1.46 g/cm³, and excellent adhesion. FTIR analysis confirmed the formation of Al–O–P and pyrophosphate bonds, while thermal and mechanical stability were evaluated under controlled conditions. The results indicate that aluminum slag can be effectively valorized as a raw material for high-performance inorganic binders.

The process of phase formation in the synthesis of Al2O3 - Y2O3 - SiO2 frit by sol-gel and coprecipitation method

Wed, 25 Jun 2025 00:00:00 +0000

Frits are utilized in various high-temperature and structural applications due to their exceptional properties. However, most frits are fabricated via solid-state reactions, which require high sintering temperatures and extended time to achieve homogeneity. Therefore, this study investigates the phase transition in the Al₂O₃ -Y₂O₃ -SiO₂ frit using sol-gel method, which enables lower synthesis temperature and achieves higher purity. The energy dispersive X-ray spectroscopy (EDX) results confirm the successful fabrication of Al₂O₃ -Y₂O₃ -SiO₂ frits. The effect of the sintering conditions on the phase transition was studied. The X-ray diffraction patterns revealed that the frit remained completely amorphous up to 1000 °C. Although the crystalline phases Y₂O₃·2SiO₂, Y₂O₃·SiO₂, and α-Al₂O₃ begin to appear around 1100 °C, the overall degree of crystallinity increases with temperature, indicating a gradual transformation from an amorphous to a more crystalline structure. A fully crystallized structure of the Al₂O₃ -Y₂O₃ -SiO₂ frit is observed at 1400 ^oC. The differential thermal analysis (TGA) results confirm the glass transition, crystallization and eutectic temperature of the Al₂O₃ -Y₂O₃ -SiO₂ frit. The optimized sintering condition was determined to be 1400 ^oC for 5 hours.

The extraction procedure of a pluthiophenol-rich extract from the leaves of Pluchea indica L. collected in Hanoi

Vu Minh Trang, Nguyen Thi Hong Anh — Wed, 25 Jun 2025 00:00:00 +0000

In this study, we evaluated some conditions affecting the extraction of pluthiophenol rich in pluthiophenol from the leaves of Pluchea indica collected in Hanoi. The investigated factors included extraction method, extraction solvent, medicinal plant size, material mass/solvent volume ratio, extraction time, and number of extraction times. Additionally, several methods of refining and enriching pluthiophenol in the Pluchea indica extract were also investigated. The criteria used to evaluate and select extraction conditions was the pluthiophenol content by the HPLC method. The best results included using hot extraction at 80 ^oC, extraction solvent of 60% ethanol, material size of 1 mm, extraction three times and 2 hours each time, with material/solvent ratios of 1/8, 1/6, and 1/6 (g/mL). Refining and enriching the pluthiophenol component was done using a liquid-liquid distribution and column chromatography (Amberlite XAD4 polymer adsorbent). The obtained extract after refining had a pluthiophenol content of 6.12%, and the pluthiophenol recovery efficiency of the whole process was 79.74% on a laboratory scale (50 g). In the 1 kg scale extraction procedure, the obtained extract had a pluthiophenol content of 6.29%.

Black-box model functionality stealing for Vietnamese sentiment analysis

Pham Xuan Cong, Do Viet Binh, Hoang Trung Nguyen, Tran Cao Truong — Wed, 25 Jun 2025 00:00:00 +0000

Black-box deep learning models often keep critical components such as model architecture, hyperparameters, and training data confidential, allowing users to observe only the inputs and outputs without understanding their internal workings. Consequently, there is growing interested in developing "knockoff" models that replicate the behavior of these black-box models without direct access to internal details. We have conducted extensive studies on function extraction attacks targeting English text sentiment analysis models. By employing random or adaptive sampling methods, we have successfully reconstructed knockoff models that achieve functionality equivalent to the original models with high similarity. In this study, we extend our investigation to sentiment analysis datasets in Vietnamese. Experimental results demonstrate that for black-box models in Vietnamese text sentiment analysis, our method remains effective, successfully constructing models with equivalent functionality.

FGSM Attack on CNN-based image classifiers: Vulnerability analysis and an effective defense strategy

Doan Huong Giang, Pham Thi Thanh Thuy — Wed, 25 Jun 2025 00:00:00 +0000

Convolutional Neural Networks (CNNs) have demonstrated significant advantages and have, therefore, been widely applied across various domains. However, adversarial attacks have exposed critical vulnerabilities in these models, posing threats to the security and reliability of deep learning systems. Although numerous studies have investigated adversarial attacks on deep learning models, the specific impact of such attacks on CNN-based image classifiers remains an open issue, especially considering that many widely-used CNN models form the foundation of essential real-world applications. This study analyzes the vulnerabilities of CNN image classifiers under the Fast Gradient Sign Method (FGSM) adversarial attack and proposes an effective defense strategy named WR_FGSM. Experimental results on standard benchmark datasets show that several CNN models suffer significantly from FGSM attacks. The adversarial images generated by this attack not only deceive CNN-based image classifiers but also appear visually indistinguishable from the original images to the human vision. Our proposed WR_FGSM defense incorporates adversarial training—one of the most effective existing defense strategies—along with a regularization technique during the training process. This approach effectively safeguards CNN models against FGSM attacks while maintaining a balance between adversarial robustness and the generalization capability of the models.

Effect of nanosilica and nano zirconium oxide mixture modified by polydimethyl siloxane on the heat resistance of silicone paint

Nguyen Ba Ngoc — Wed, 25 Jun 2025 00:00:00 +0000

Silicone-based paints are known for their high-temperature resistance and durability in chemical environments. This paper presents research findings on a silicone paint formulation supplemented with a mixture of nano-silica and nano-zirconium oxide, which has been surface-modified with Polydimethylsiloxane (PDMS). The incorporated ratio was 0.45% modified nano-silica and 1.05% modified nano-zirconium oxide. Following thermal endurance evaluation (a thermal shock test conducted at 1,050 °C for 25 seconds), no micro-cracks were observed on the surface. This is attributed to the increased flexibility of the silicone coating at high temperatures and the more uniform dispersion of the modified nano-additive particles within the silicone resin matrix, thereby enhancing the thermal resistance of the paint film. The author also conducted a comparative heat resistance test using an acetylene torch, with the sample plate's exposure temperature reaching approximately 1,100 °C for a duration of 25 to 30 seconds. The performance of the modified nano-particle-containing silicone paint was compared against the BKV inorganic heat-resistant paint and alkyd paint on steel test panels from a CT-18 engine, yielding favorable results.

Research on a sealing solution using O-rings in the design and manufacture of underwater observation devices

Tran Xuan Tien, Le Van Hoang — Wed, 25 Jun 2025 00:00:00 +0000

This paper presents a comprehensive study on the design of pressure-resistant seals for underwater observation devices operating in seawater environments. The unique challenges posed by seawater, including its high salinity and corrosive elements, necessitate innovative material selection and precise engineering techniques to ensure the durability and efficiency of the seals. This research focuses on identifying corrosion-resistant elastomers and coatings suitable for withstanding high pressures and temperature fluctuations in the harsh marine environment. Additionally, the study delves into the impact of meticulous assembly practices on the long-term performance and reliability of the sealing components. The research team calculated and designed a prototype of the underwater observation device's shell and successfully conducted a water pressure test at a depth of up to 80 meters. The sealing limit primarily depends on the mechanical properties of the gasket material. When using metal gaskets, the sealing limit can reach up to 5000 psi, equivalent to a depth of 3400 meters, which is also the current testing equipment limit.

Image fusion using FPGA in multi-sensor optoelectronic systems

Nguyen Quang Hiep, Tran Dinh Hoa, Dang Van Tang — Wed, 25 Jun 2025 00:00:00 +0000

This study proposes a foundation for developing a real-time image fusion system for multi-sensor optoelectronic systems applied in defense and security, aiming to enhance target detection by combining images from thermal imaging and low-light channels. The VIP Board Big V3.8 with Cyclone IV FPGA, a custom FPGA development board, which integrates key peripherals such as SDRAM, VGA, HDMI, and image sensors, is utilized. The main hardware modules have been designed, including global clock management (PLL), UART communication, FIFO buffers, LCD display control, and image fusion. The image fusion module employs an algorithm that decomposes images into two layers, the base layer and the detail layer, enabling efficient data fusion from two input images to generate a high-quality output image. Experimental results demonstrate that the system operates stably, meeting speed and low-latency requirements, paving the way for future research on more powerful FPGA platforms.

Configuration of nonlinear couplers for creation of random optical multipulses

Bui Xuan Kien — Wed, 25 Jun 2025 00:00:00 +0000

In this paper, a configuration of nonlinear couplers with different coupling lengths is presented. Based on the influence of the transmission coefficient on the nonlinear index coefficient, the interval of the input intensity to generate overlap-transmission intensity at two output ports is determined. This behavior is simulated to confirm the connection of multiple nonlinear couplers as a configuration to generate random optical multi-pulses. The simulation results give us an opportunity to modulate laser pulses for information security of wireless or free-space laser communication.