An all-MOS memristor integrated circuit design for Vilnius chaotic circuits

Abstract:This paper proposes a low-power memristor integrated circuit based on CMOS technology.Comprising only five transistors,the circuit utilizes parasitic capacitance effects to achieve memristive characteristics,with no need for external biasing.Compared with existing designs,this circuit demonstrates significant advantages in structural complexity,operating frequency,area efficiency,and power consumption,making it suitable for highly integrated neuromorphic hardware systems.The circuit is designed using SMIC 0.18 μm CMOS technology with a 1.8 V power supply,achieving a maximum operating frequency of 100 MHz while consuming only 14.51 μW and occupying a layout area of merely 410 μm2.

Structural crack intelligence based on lightweight DeepLabv3+identification research
[Journal Article]CHEN Zhou, WEN Jiahao, LU Hanwen et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:In order to solve the problems of weak generalization ability and large storage resource demand of existing bridge crack semantic segmentation models,an improved DeepLabv3+model was proposed,which replaced the feature extraction network with the lightweight network MobileNetv2,and combined with the Swish activation function and transfer learning strategy.In order to verify the effectiveness of the improved model,a bridge crack dataset was constructed by using different types of bridge crack images under complex background interference,and the cracks of the improved DeepLabv3+model,DeepLabv3+model,Segnet model and Unet model were trained for crack recognition,and the recognition effects of the four models were compared and analyzed from the aspects of segmentation accuracy,average interaction ratio and model size.The analysis results show that the segmentation accuracy of the improved DeepLabv3+model reaches 93.41%,the average interaction ratio reaches 78.51%,and the F1 score reaches 83.60%.The size of the improved model is only 6.64 MB,which is at the same order of magnitude as the size of the Segnet model and significantly smaller than that of the DeepLabv3+and Unet models.

Study on the antibacterial activity and stability of the antimicrobial peptide Protegrin-1 against multidrug-resistant avian Escherichia coli
[Journal Article]HE Yinlin, XU Jing, TANG Biao et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:To explore the potential of the antimicrobial peptide Protegrin-1(PG-1)as an alternative to antibiotics,this study investigated its bacteriostatic activity,biofilm formation-inhibiting ability,and bacteriostatic stability against two clinically isolated avian Escherichia coli(E.coli)strains,namely HNY006 and HNJ006.The minimum inhibitory concentration(MIC)and minimum bactericidal concentration(MBC)of PG-1 against the two strains were determined by broth microdilution.The effect of PG-1 on biofilm formation was assessed by crystal violet staining.The inhibitory rate of PG-1 under different temperatures,pH levels,and UV treatment conditions was evaluated by spectrophotometry.The results showed that both the MIC and MBC of PG-1 against the two strains were 64 μg/mL,and PG-1 at the MIC concentration significantly inhibited biofilm formation of the two strains(P<0.05).Additionally,PG-1 exhibited good thermal stability,acid-base stability,and UV stability.The above results indicate that PG-1 inhibits avian E.coli,suppresses its biofilm formation,and exhibits good stability.

Analysis of the influence of cantilever length on the bearing characteristics of h-type double-row piles
[Journal Article]WANG Xiangqiu, LUO Xiaodong, LIU Jian-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:Taking the h-type double-row pile support structure of a deep foundation pit project in Shenzhen as the research object,a numerical analysis model of h-type double-row piles was constructed using MIDAS-GTS finite element software.Based on the verification of the model's effectiveness,the influence of different cantilever lengths(0 m,2 m,4 m,6 m)of the rear-row piles on the deep horizontal displacement,bending moment,and shear force of the h-type double-row pile support structure were explored.The research results indicate that,under the same excavation depth and design parameters of the foundation pit,h-type double-row piles can reduce the horizontal displacement of the support structure and optimize the distribution of bending moment and shear force of the front-row piles compared to traditional gantry double row piles.Based on a comprehensive analysis of the deep horizontal displacement,bending moment,and shear force of piles under different cantilever length conditions,it is reasonable to take a cantilever length of 6m for the rear-row piles in this engineering case.

Study on AGP-injective rings satisfying certain finite conditions
[Journal Article]LU Qi, CHEN Huaxi, WEI Wenlian-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:This study generalizes some conclusions on GP-injective rings and investigates AGP-injective rings satisfying finiteness conditions on the ascending chain of annihilators.It is proved that left nonsingular left AGP-injective rings that do not contain an infinite direct sum of non-zero principal left ideals are semisimple rings.By replacing ideals with GW-ideals,it is further proved that if every simple singular left-modules are AGP-injective,and any maximal essential left ideal of are GW-ideal,then R is left nonsingular.These results further enrich the theory of AGP-injective rings.

Measurement of field of view utilizing monocular camera calibration technology
[Journal Article]CHENG Xiaoqi, LIU Zhimin, ZENG Zhihao et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:Field of View(FOV)is a key parameter for evaluating the effective imaging range of a camera.To address the issues of low accuracy and complex operation in traditional FOV measurement methods,this paper proposes an FOV measurement approach based on single-target calibration technology.First,the precise calibration of the target camera is achieved using Zhang's calibration method to obtain the camera's intrinsic matrix and distortion coefficients.Then,undistortion is performed on the captured images using the camera's perspective projection model,and the maximum visible range in the undistorted image,represented by the largest inscribed orthogonal rectangle,is determined.Finally,the back-projection planes of the rectangle's edges are calculated using the intrinsic matrix,and the angles between the opposite planes of the top-bottom and left-right edges are determined,thus enabling the rapid calculation of the camera's vertical and horizontal FOVs.Experimental results show that the proposed FOV measurement method offers advantages such as operational flexibility and higher measurement accuracy compared to traditional methods.

Crop planting strategy based on linear programming model
[Journal Article]XU Tingting, FENG Zaiyong, XIE Xiaowei-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:Based on the relevant planting data in Question C"Crop Planting Strategies"of the 2024"Higher Education Society Cup"National College Student Mathematical Modelling Competition,this study constructs a linear programming model.The model takes maximising economic benefits as its objective function,while considering the constraints of planting area,crop growth law and soil fertility.The innovation of the research lies in the further introduction of new constraints for unsold and surplus parts caused by higher-than-expected yields,which are sold at 50%of the original price,as decision variables,thus improving the model.In this way,this study not only optimised the planting area and plot distribution,but also minimised the waste caused by unsold product,thereby maximising the economic benefits.

Design of an ultra-high-frequency memristor circuit based on 180 nm CMOS process

Abstract:This paper presents an ultra-high-frequency integrated memristor circuit designed based on the SMIC 180 nm CMOS process.The circuit utilizes a four-transistor configuration,exploiting the nonlinear switching characteristics of MOSFET and gate voltage control to achieve memristive behavior.The design was simulated and validated using Cadence simulation software.The results indicate that:the proposed memristor circuit features a simple topology with a compact layout area of merely 45 μm2 and a power consumption of 14.04 μW at 1 GHz;operating under a±0.9 V DC supply,the circuit achieves a maximum operating frequency of 1 GHz and exhibits well-defined hysteresis loops across the frequency range from 500 MHz to 1 GHz;the implemented Chua chaotic circuit demonstrates robust chaotic performance,suggesting its potential applications in chaotic integrated circuits,information security,and related domains.

Metagenomic identification and recombination analysis of mixed infection with diverse lineages of PRRSV-2 on a large-scale pig farm in Guangdong
[Journal Article]CHEN Xiuqiao, GUO Yajing, YAN Guangzhi et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:To investigate the genomic characterization of porcine reproductive and respiratory syndrome virus(PRRSV)in large-scale pig farms,viral metagenomics technology was applied to perform high-throughput sequencing of a lung tissue sample that was positive for PRRSV by RT-qPCR,and the sequencing results were subjected to virome analysis and genomic characterization.Two complete genome sequences of PRRSV were obtained from a lung tissue sample designated as GD202301 and GD202302,respectively.Genomic characterization showed that the complete genome sequence and ORF5 gene of GD202301 belonged to PRRSV-2 lineage 8 and lineage 3 in the genetic evolution analysis,respectively,and was produced by triple recombination of three strains,NADC30-like,JXA1-like and QYYZ-like.The complete genome sequence and the ORF5 gene of GD202302,both of which belonged to PRRSV-2 lineage 1 in the genetic evolutionary analysis,was triple recombined from three strains,NADC30-like,JXA1-like,and QYYZ-like.The presence of two types of PRRSV in a sample was found by viral metagenomics technology,which indicates that there is a mixed infection of two different spectrums of PRRSV in this farm,and provides data support for the understanding of PRRSV infection in this farm.

Circuit simulation and synchronization realization of fourth-order hyperchaotic system

Abstract:To study the physical realizability of the newly constructed fourth-order hyperchaotic system,the system equations were transformed,and a circuit was built based on the Multisim software.The parameters of the components in the circuit were set,and the simulation results of the circuit were compared with the phase trajectory of the chaotic system.The comparison showed consistency between the two,proving the realizability of the hyperchaotic system.The synchronization problem of the proposed hyperchaotic system under known and unknown parameters was studied.The function projective synchronization and adaptive improved function projective synchronization methods were adopted respectively,and the corresponding synchronization controllers were designed.The effectiveness of the designed controllers was proved based on the Lyapunov stability theorem.At the same time,the correctness and effectiveness of the designed controllers were further verified through numerical simulation in Matlab.

A note on the solution of third order matrix differential equations
[Journal Article]LIU Zhihui, LIAO Meiting, LIANG Hongxue-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:This paper discusses the third-order differential equations without first order items.According to the theory of higher order equations and the theory of Matrix,this paper first reduces the lower order of homogeneous equations by the method of degradation.Then,using Cardan formula and Euler method,the general solution of the homogeneous expression of a class of two dimensional,third order constant coefficient differential equation is obtained.Furthermore,one special solution formula of trigonometric function multiplying exponential function is studied by using the method of undetermined Matrix and comparing by column,the accuracy of the general solution and the special solution of the equation is verified by an example.

Analysis of the composition of distiller's grains wastewater of Chi-flavor baijiu
[Journal Article]HONG Jiazhi, MA Yucheng, ZHANG Shuning et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:To elucidate the chemical composition of distiller's grains wastewater from Chi-flavor baijiu,volatile components were qualitatively and quantitatively analyzed by gas chromatography,organic acids were determined by high-performance liquid chromatography,and non-volatile components were characterized by ultra-performance liquid chromatography-tandem mass spectrometry.The results indicated that the volatile components included esters(predominantly ethyl lactate and ethyl acetate),alcohols(mainly glycerol and 2,3-butanediol A),and acids(primarily lactic acid and acetic acid).Among the non-volatile components,amino acids and their derivatives showed the highest diversity(accounting for 29.35%of the total components),while fatty acids exhibited the greatest relative abundance(28.70%).These findings demonstrate that the main chemical constituents of the wastewater are carbohydrates,organic acids,alcohols,esters,amino acids and their derivatives,and fatty acids,providing a scientific basis for the treatment and resource utilization of distiller's grains wastewater.

Anti-malicious user attack algorithm based on trust

Abstract:Aiming at the problem that traditional collaboration algorithms cannot solve the attack of malicious users,a malicious user(MSU)attack prevention algorithm based on trust degree is proposed.This algorithm establishes a user behavior trust degree model to dynamically evaluate the perception reliability of collaborative users,effectively identifying and filtering malicious data.Simulation results show that in the presence of MSU attacks,the proposed algorithm can effectively reduce the system's missed detection rate and significantly improve the system's perception effect in different channel environments.

Study on the effect of annual sowing on the stage development of Jin Yin 131 sweet corn
[Journal Article]LIAO Xiaoxiao, CHEN Haoshen, WANG Yunbo et al.-Journal of Foshan University(Natural Science Edition)2025, No.06

Abstract:This study used the new sweet corn variety Jin Yin 131 as the test material,and conducted the annual cultivation experiment with 24 solar terms as the sowing time to explore the effect of annual sowing on its stage development.The results showed that sowing in the spring and autumn,which were the most suitable growing seasons,takes 5 days to seedling emergence,about 55 days to tassel and silk,and about 75 days to harvest.With the increase of average daily accumulated temperature,the development of each growth stage was advanced,and the harvest period could be advanced to 70 days;with the decrease of average daily accumulated temperature,each growth stage showed a delay,and the harvest period could be prolonged to about 120 days.There was a significant negative correlation between average daily accumulated temperature and the number of days for each stage of development.The study indicated that Jin Yin 131 sweet corn can be sown year-round,but its development at each stage was significantly affected by the average daily accumulated temperature.In production,it was recommended to sow in early spring and late autumn in the southeast fresh corn area to to fully leverage its high-quality and high-yield characteristics.

A community detection method integrating clustering coefficient and node similarity optimization

Abstract:Non-negative matrix factorization is widely used in the field of community detection due to its effectiveness and easy interpretation.Most existing methods only consider network topology information and ignore node similarity information.To address this problem,a community detection model based on symmetric non-negative matrix factorization that integrates clustering information and similarity is proposed.The algorithm uses a variety of indicators to describe the node similarity of the network,and further integrates the clustering coefficient to optimize the node similarity,so that it can play a better role and improve the accuracy of community discovery.This paper conducts experiments on four real-world network datasets,and compares the node similarity before and after optimization based on the evaluation indicators.The experimental results show that the optimization model in this paper can effectively improve the accuracy of network prediction.

Study on the influence of bearing positioning preload on the critical speed of rotor system
[Journal Article]LI Kui, JIANG Lingli, RAO Miaohong et al.-Journal of Foshan University(Natural Science Edition)2025, No.05

Abstract:Improper preload of the bearing may lead to increased vibration of the rotor system,thereby reducing the stability and service life of the rotor system.In this paper,the influence of axial clearance on the dynamic characteristics of the rotor system of tapered roller bearings under the condition of positioning preload is studied.By combining ROMAX software and ANSYS finite element software,a dynamic co-simulation model of bearing-rotor system is established.The model can analyze the influence of bearing stiffness on the critical speed of the rotor system under different preload conditions.The results show that increasing the axial clearance of the bearing properly can improve the stiffness of the bearing,and then improve the critical speed of the rotor system.

Effects of Epimedium brevicornum Maxim on growth performance and immune function in BALB/c mice
[Journal Article]WU Junjie, LIU Hanwen, WANG Yichen et al.-Journal of Foshan University(Natural Science Edition)2025, No.05

Abstract:This study aimed to investigate the regulatory effects of unfermented Epimedium brevicornum Maxim(EbM)and EbM fermented by lactococcus lactis on the growth performance and immune function of BALB/c mice.Twenty-eight 8-week-old BALB/c mice were divided into seven groups:a control group(N)and six experimental groups.The experimental groups included unfermented EbM at concentrations of 25 μg/mL(L1),45 μg/mL(M1),and 65 μg/mL(H1),as well as lactococcus lactis-fermented EbM at concentrations of 25 μg/mL(L2),45 μg/mL(M2),and 65 μg/mL(H2).Each group comprised four replicates(n=4).Mice in the experimental groups received intragastric administration of the respective treatments once every 24 hours for 7 days,while the control group received saline.The body weight of each mouse was measured before and after the experiment.Blood samples were collected to detect serum immunoglobulins(IgA,IgG,IgM)and the immune cytokine tumor necrosis factor-alpha(TNF-α).Compared with the control group,all experimental groups exhibited a significant reduction in body weight;serum levels of IgA,IgG,and IgM were significantly elevated,while TNF-α levels were significantly decreased in treated mice.These results indicate that both unfermented EbM and lactococcus lactis-fermented EbM enhance immune performance in mice to varying degrees,but result in a significant reduction in body weight.

Improved temporal network-based method for motor bearing fault diagnosis
[Journal Article]HUANG Xingyuan, YANG Tongguang, JIANG Lingli et al.-Journal of Foshan University(Natural Science Edition)2025, No.05

Abstract:Fatigue wear of the bearings in high-power variable-frequency motors causes the vibration signals to be mixed.In order to improve the accuracy of fault diagnosis,a multi-channel insulated bearing time information fusion diagnosis model was designed,and the coarse-grained features with temporal patterns were extracted from the measured fault data.The self-attention mechanism was introduced into the designed time information fusion diagnosis model for optimization,and the correlation weights between the data were calculated by constantly updating the weight coefficients.The proposed diagnostic framework demonstrated superior overall fault recognition rate of 99.1%,significantly outperforming conventional recurrent neural network architectures including GRU(94.7%),LSTM(91.2%),and RNN(88.4%).

Effects of β-blocker carvedilol on the proliferation of PRRSV-CHR6 strain in vitro
[Journal Article]LI Shun, YU Qiaoping, DENG Jingfei et al.-Journal of Foshan University(Natural Science Edition)2025, No.05

Abstract:This study aimed to investigate the impact of the β-blocker Carvedilol(CAR)on the in vitro inhibition of Porcine reproductive and respiratory syndrome virus(PRRSV)proliferation.A series of experiments such as CCK8,Western blotting,RT-qPCR,and indirect immunofluorescence were conducted to explore the efficacy and mechanism of CAR on PRRSV.The findings revealed that CAR exhibited a CC50 of 30.33 μM and an IC50 of 0.15 μM,demonstrating significant inhibitory effects on PRRSV proliferation(P<0.05).Furthermore,at concentrations of 5 μM,10 μM,and 20 μM,CAR displayed effective anti-PRRSV activity(P<0.05).Both pre-treatment and post-treatment with CAR at a concentration of 5 μM demonstrated substantial anti-PRRSV effects(P<0.05).Additionally,treatment with CAR at a concentration of 5 μM significantly upregulated the mRNA transcription levels of IFN-β,ISG15,and ISG56,as well as the expression levels of phosphorylated IRF3(P<0.05),suggesting that CAR may activate the type I interferon signaling pathway to inhibit PRRSV proliferation.This study represents the first evidence for significant in vitro anti-PRRSV activity by the β-blocker CAR while also elucidating its mechanism against PRRSV through activation of the type I interferon signaling pathway.

Correlation and principal component analysis of semen quality in different stall types of boars
[Journal Article]YANG Xiyan, NIE Jingkun, HE Jian et al.-Journal of Foshan University(Natural Science Edition)2025, No.05

Abstract:The study aimed to explore the effects of limited and large pens on the semen quality of adult boars,as well as the relationship between various semen indicators.A total of 50 Duroc boars in limited pens and 220 Duroc boars in large pens were selected as experimental subjects.Semen record data(including 22 semen trait indicators)were collected from a large-scale boar station under a group enterprise in Guangxi from December 2022 to August 2023.By means of analysis of variance,correlation analysis,and principal component analysis,the differences in semen quality among different types of pens were identified,and the relationships between various indicators affecting semen traits were explored.The results demonstrated that the sperm motility and normal morphology of confined boars were significantly lower than those of large pigs(P<0.001),and the semen density was significantly lower than that of large pigs(P<0.05);Compared with the control group,the dynamic parameters of male pig sperm in the restricted range showed no significant difference except for LIN(%)(P>0.05),while other parameters showed extremely significant differences(P<0.001);The proportion of bent tail,curled tail,distal cytoplasmic droplet,proximal cytoplasmic droplet,and midline interruption in the sperm of confined boars was significantly higher than that of large pigs(P<0.001);The sperm abnormalities of confined boars were significantly negatively correlated with VSL,STR,LIN,DSL,and positively correlated with BCF;There is a highly significant positive correlation between the sperm curling tail and the sperm kinematic parameters in the Da Lan boar.The results of principal component analysis showed that the cumulative contribution rates of the first five principal components in the limit bar and the large column were 76.941%and 75.334%,respectively.These findings demonstrate that the overall semen quality of large pen boars was better than that of limited pen boars,and the normal sperm morphology rate,linear velocity of sperm movement,and forward orientation were important indicators for measuring semen quality characteristics.