Low-Cost Federated Broad Learning for Privacy-Preserved Knowledge Sharing in the RIS-Aided Internet of Vehicles

High-efficiency and low-cost knowledge sharing can improve the decision-making ability of autonomous vehicles by mining knowledge from the Internet of Vehicles(IoVs).However,it is challenging to ensure high efficiency of local data learning models while preventing privacy leakage in a high mobility environment.In order to protect data privacy and improve data learning efficiency in knowledge sharing,we propose an asynchronous federated broad learning(FBL)framework that integrates broad learning(BL)into federated learning(FL).In FBL,we design a broad fully connected model(BFCM)as a local model for training client data.To enhance the wireless channel quality for knowledge sharing and reduce the communication and computation cost of participating clients,we construct a joint resource allocation and reconfigurable intelligent surface(RIS)configuration optimization framework for FBL.The problem is decoupled into two convex subproblems.Aiming to improve the resource scheduling efficiency in FBL,a double Davidon–Fletcher–Powell(DDFP)algorithm is presented to solve the time slot allocation and RIS configuration problem.Based on the results of resource scheduling,we design a reward-allocation algorithm based on federated incentive learning(FIL)in FBL to compensate clients for their costs.The simulation results show that the proposed FBL framework achieves better performance than the comparison models in terms of efficiency,accuracy,and cost for knowledge sharing in the IoV.

BLS-identification:A device fingerprint classification mechanism based on broad learning for Internet of Things

The popularity of the Internet of Things(IoT)has enabled a large number of vulnerable devices to connect to the Internet,bringing huge security risks.As a network-level security authentication method,device fingerprint based on machine learning has attracted considerable attention because it can detect vulnerable devices in complex and heterogeneous access phases.However,flexible and diversified IoT devices with limited resources increase dif-ficulty of the device fingerprint authentication method executed in IoT,because it needs to retrain the model network to deal with incremental features or types.To address this problem,a device fingerprinting mechanism based on a Broad Learning System(BLS)is proposed in this paper.The mechanism firstly characterizes IoT devices by traffic analysis based on the identifiable differences of the traffic data of IoT devices,and extracts feature parameters of the traffic packets.A hierarchical hybrid sampling method is designed at the preprocessing phase to improve the imbalanced data distribution and reconstruct the fingerprint dataset.The complexity of the dataset is reduced using Principal Component Analysis(PCA)and the device type is identified by training weights using BLS.The experimental results show that the proposed method can achieve state-of-the-art accuracy and spend less training time than other existing methods.

Spontaneous Paravesical and Broad Ligament Hematoma after Vaginal Delivery Had Uterine Artery Embolization after Evacuating the Hematoma

Broad ligament hematoma is typically seen during cesarean section due to rupture of branches of uterine and vaginal vessels and it’s rare to be seen post-normal vaginal delivery. Addressing puerperal hematomas postpartum presents considerable challenges for obstetric care providers. While hematomas such as those affecting the vulva, vulvovaginal region, or paravaginal area are frequently encountered, retroperitoneal hematomas are rare and notably pose a greater risk to the life of the patient. The medical literature contains scant case reports on retroperitoneal hematomas, with no consensus on a definitive treatment approach. Pelvic arterial embolization has emerged as both a sensible and increasingly preferred method for treating these hematomas recently, but its application is contingent upon the patient maintaining hemodynamic stability and the availability of a specialized interventional embolization unit. In our case, we are presenting a very rare case of a 31-year-old primigravida female with a history of in vitro fertilization pregnancy. She delivered a normal vaginal delivery at 31 weeks gestation. Unfortunately, she experienced multiple complications intrapartum, including preeclampsia and placental abruption. These complications increased her risk of developing a broad ligament hematoma.

High efficient broad-spectrum Bradyrhizobium elkanii Y63-1

Soybean(Glycine max),the primary source of high-quality plant protein,plays a crucial role as a grain and oil crop in China.Harnessing the full potential of symbiotic nitrogen fixation in soybean production holds immense significance for agriculture and ecology alike.Zhongdou 63,a newly developed early-maturing summer soybean cultivar in 2021,exhibits remarkable traits such as high yield,superior quality,multi-resistance,and wide adaptability.In this study,eight distinct rhizobia strains from diverse regions were meticulously screened to identify highly effective strains specifically suited for Zhongdou 63.The aboveground biomass,plant height,chlorophyll content,root length,nodule number,and nodule dry weight of Zhongdou 63 were measured and the data were subjected to statistical analysis.The results demonstrated that Y63-1 is a predominant strain of Zhongdou 63.Subsequently,we conducted further investigations on the broad-spectrum nodulation characteristics of Y63-1.Ten representative soybean cultivars were individually inoculated with Y63-1 and subsequently analyzed for nodule numbers and nodule dry weight in their symbiotic systems with rhizobia.The findings revealed that Y63-1 effectively formed nodules with all ten soybean varieties tested.In summary,our current study identified highly efficient broad-spectrum Bradyrhizobium elkanii strain Y63-1 as the predominant strain in Zhongdou 63 and provided a theoretical foundation for enhancing yield potential not only in Zhongdou 63 but also in other varieties through inoculation with highly efficient rhizobia in production.

Mulching broad ridges with a woven polypropylene fabric increases the growth and yield of young pear trees ‘Yuluxiang' in the North China Plain

Unlike traditional ridging, mulching broad ridges with a woven polypropylene fabric (WPF) can reduce soil evaporation during the drought season and avoid long saturation time in the root zone of pear trees during the rainy season. In this study, field experiments were conducted from 2017 to 2020 in a pear orchard in the North China Plain to investigate the effects of mulching broad ridges (0.3 m in height and 2 m in width) with WPF on soil temperature and moisture, nitrogen leaching, vegetative and reproductive growth of young pear trees(Pyrus bretschneideri Rehd.‘Yuluxiang’). The experiments involved two treatments, namely, control (traditional no-ridge planting without mulching) and mulching broad ridges with WPF (RM treatment). The results showed that the RM treatment increased soil moisture and temperature and decreased nitrogen leaching, resulting in vigorous growth of the young pear trees. Moreover, the RM treatment increased the tree trunk cross-sectional area and height of the young pear trees by 37%and 8%in 2020, respectively. The nitrate nitrogen content at the soil layer depth of 0-30 cm was significantly higher in the RM than that in control. Furthermore, the RM treatment significantly increased the fruit yield due to larger tree size. In addition, compared with control, significantly higher fruit soluble solid content of RM treatment was detected in 2020. High precipitation (423 mm) occurred during fruit enlargement stage in 2020, RM treatment decreased the rainfall infiltration in the ridge and the soil moisture in root region, resulting in the improvement of fruit quality, compared with control.Therefore, mulching broad ridges with WPF can be implemented to increase soil moisture during drought season, soil temperature, and nitrate nitrogen content, thereby improving the growth and fruit yield of young pear trees. Additionally, it can reduce soil moisture in the root zone during the rainy season and improve the fruit quality of the trees. Finally, it can reduce nitrate nitrogen leaching, thereby reducing environmental pollution.

Target tracking method of Siamese networks based on the broad learning system

Target tracking has a wide range of applications in intelligent transportation,real‐time monitoring,human‐computer interaction and other aspects.However,in the tracking process,the target is prone to deformation,occlusion,loss,scale variation,background clutter,illumination variation,etc.,which bring great challenges to realize accurate and real‐time tracking.Tracking based on Siamese networks promotes the application of deep learning in the field of target tracking,ensuring both accuracy and real‐time performance.However,due to its offline training,it is difficult to deal with the fast motion,serious occlusion,loss and deformation of the target during tracking.Therefore,it is very helpful to improve the performance of the Siamese networks by learning new features of the target quickly and updating the target position in time online.The broad learning system(BLS)has a simple network structure,high learning efficiency,and strong feature learning ability.Aiming at the problems of Siamese networks and the characteristics of BLS,a target tracking method based on BLS is proposed.The method combines offline training with fast online learning of new features,which not only adopts the powerful feature representation ability of deep learning,but also skillfully uses the BLS for re‐learning and re‐detection.The broad re‐learning information is used for re‐detection when the target tracking appears serious occlusion and so on,so as to change the selection of the Siamese networks search area,solve the problem that the search range cannot meet the fast motion of the target,and improve the adaptability.Experimental results show that the proposed method achieves good results on three challenging datasets and improves the performance of the basic algorithm in difficult scenarios.

Flexible broadband WS_(2)/Si optical position-sensitive detector with high sensitivity and fast speed

Si-based optical position-sensitive detectors(PSDs)have stimulated the interest of researchers due to their wide range of practical applications.However,due to the rigidity and fragility of Si crystals,the applications of flexible PSDs have been limited.Therefore,we presented a flexible broadband PSD based on a WS_(2)/Si heterostructure for the first time.A scalable sputtering method was used to deposit WS_(2)thin films onto the etched ultrathin crystalline Si surface.The fabricated flexible PSD device has a broad spectral response in the wavelength range of 450-1350 nm,with a high position sensitivity of~539.8 mV·mm^(−1)and a fast response of 2.3μs,thanks to the strong light absorption,the built-in electrical field at the WS_(2)/Si interface,and facilitated transport.Furthermore,mechanical-bending tests revealed that after 200 mechanical-bending cycles,the WS_(2)/Si PSDs have excellent mechanical flexibility,stability,and durability,demonstrating the great potential in wearable PSDs with competitive performance.

A broad learning-based comprehensive defence against SSDP reflection attacks in IoTs

The proliferation of Internet of Things(IoT)rapidly increases the possiblities of Simple Service Discovery Protocol(SSDP)reflection attacks.Most DDoS attack defence strategies deploy only to a certain type of devices in the attack chain,and need to detect attacks in advance,and the detection of DDoS attacks often uses heavy algorithms consuming lots of computing resources.This paper proposes a comprehensive DDoS attack defence approach which combines broad learning and a set of defence strategies against SSDP attacks,called Broad Learning based Comprehensive Defence(BLCD).The defence strategies work along the attack chain,starting from attack sources to victims.It defends against attacks without detecting attacks or identifying the roles of IoT devices in SSDP reflection attacks.BLCD also detects suspicious traffic at bots,service providers and victims by using broad learning,and the detection results are used as the basis for automatically deploying defence strategies which can significantly reduce DDoS packets.For evaluations,we thoroughly analyze attack traffic when deploying BLCD to different defence locations.Experiments show that BLCD can reduce the number of packets received at the victim to 39 without affecting the standard SSDP service,and detect malicious packets with an accuracy of 99.99%.

Closed loop ileus caused by a defect in the broad ligament:A case report

BACKGROUND Closed loop ileus caused by entrapment of bowel in a defect of the broad ligament is a rarity.Only a few cases have been reported in the literature.CASE SUMMARY We present the case of a 44-year-old,healthy patient with no prior history of abdominal surgery who developed a closed loop ileus due to an internal hernia secondary to a defect in the right broad ligament.She first presented to the emergency department with diarrhea and vomiting.As she had had no previous abdominal surgery,she was diagnosed with probable gastroenteritis and discharged.The patient subsequently returned to the emergency department due to a lack of improvement in her symptoms.Blood tests showed an elevated white blood cell count and a closed loop ileus was diagnosed on an abdominal computer tomography scan.Diagnostic laparoscopy revealed an internal hernia entrapped in a 2 cm large defect in the right broad ligament.The hernia was reduced and the ligament defect was closed using a running,barbed suture.CONCLUSION Bowel incarceration through an internal hernia may present with misleading symptoms and laparoscopy may reveal unexpected findings.

Broadband and continuous wave pumped second-harmonic generation from microfiber coated with layered GaSe crystal

The conversion-efficiency for second-harmonic(SH)in optical fibers is significantly limited by extremely weak second-order nonlinearity of fused silica,and pulse pump lasers with high peak power are widely employed.Here,we propose a simple strategy to efficiently realize the broadband and continuous wave(CW)pumped SH,by transferring a crystalline GaSe coating onto a microfiber with phase-matching diameter.In the experiment,high efficiency up to 0.08%W-1mm-1 is reached for a C-band pump laser.The high enough efficiency not only guarantees SH at a single frequency pumped by a CW laser,but also multi-frequencies mixing supported by three CW light sources.Moreover,broadband SH spectrum is also achieved under the pump of a superluminescent light-emitting diode source with a 79.3 nm bandwidth.The proposed scheme provides a beneficial method to the enhancement of various nonlinear parameter processes,development of quasi-monochromatic or broadband CW light sources at new wavelength regions.

Music Emotion Recognition Based on Feature Fusion Broad Learning Method

With the rapid development in the field of artificial intelligence and natural language processing(NLP),research on music retrieval has gained importance.Music messages express emotional signals.The emotional classification of music can help in conveniently organizing and retrieving music.It is also the premise of using music for psychological intervention and physiological adjustment.A new chord-to-vector method was proposed,which converted the chord information of music into a chord vector of music and combined the weight of the Mel-frequency cepstral coefficient(MFCC) and residual phase(RP) with the feature fusion of a cochleogram.The music emotion recognition and classification training was carried out using the fusion of a convolution neural network and bidirectional long short-term memory(BiLSTM).In addition,based on the self-collected dataset,a comparison of the proposed model with other model structures was performed.The results show that the proposed method achieved a higher recognition accuracy compared with other models.

Nutritional Value and Application of Broad Bean Nutrition Flour as a Combination of Staple Food

This paper introduced the definition and importance of combination of sta-ple food. With broad bean nutrition flour as a sample, the preparation of the combi-nation of staple food was also described. In addition, the main nutritional value of broad bean nutrition flour was introduced. Compared with those of other single flours, the nutritional value of broad bean nutrition flour was improved. Moreover, the nutrients in the broad bean nutrition flour would not be destroyed during the processing and preparation of staple food, and the processed steamed bread and raw noodle are more characteristic. The application value and prospects of broad bean nutrition flour, as a combination of staple food, were further discussed.

FTIR Spectroscopic Study of Broad Bean Diseased Leaves

[Objective] The aim was to indentify diseased leaves of broad bean by vibra- tional spectroscopy. [Method] In this paper, broad bean rust, fusarium rhizome rot, broad bean zonate spot, yellow leaf curl virus and normal leaves were studied using Fourier transform infrared spectroscopy combined with chemometrics. [Result] The spectra of the samples were similar, only with minor differences in absorption inten- sity of several peaks. Second derivative analyses show that the significant difference of all samples was in the range of 1 200-700 cm2. The data in the range of 1 200- 700 cm＇ were selected to evaluate correlation coefficients, hierarchical cluster analy- sis （HCA） and principal component analysis （PCA）. Results showed that the correla- tion coefficients are larger than 0.928 not only between the healthy leaves, but also between the same diseased leaves. The values between healthy and diseased leaves, and among diseased leaves, are all declined. HCA and PCA yielded about 73.3% and 82.2% accuracy, respectively. [Conclusion] This study demonstrated that FTIR techniques might be used to detect crop diseases.

Effects of Facilities Condition on Occurrence of Freeze Injury and Fresh Pod Yield of Broad Bean

Using Tongxian No.2 as material, the effects of different film-covering time, different sowing time and different planting density on the occurrence of freeze injury and yield of fresh broad beans were investigated. The randomized block design was adopted. The results showed that with the delayed film covering, the incidence of mild freeze injury and number of headless seedlings were increased correspondingly, but the yield was increased; with the delayed sowing, the branch number per plant, effective branch number per plant, incidence of mild freeze injury and number of headless seedlings were all reduced, and the broad beans, sowed on September 30 th, obtained the highest yield; planting density showed on effect on the occurrence of freeze injury, and the yield was increased with the increase of planting density. Under the same film-covering time, the incidence of freeze injury was reduced with the delayed sowing time and it showed no changes when planting density was changed, but the yield was increased with the increase of planting density and it was highest when broad bean seeds were sowed on September 30th;under the same sowing time, the incidence of freeze injury was increased with the delayed film-covering time and it showed no changes when planting density was changed, and the yield was increased with the delayed film-covering time and increased planting density; under the same planting density, the incidence of freeze injury was increased with the delayed film-covering time but was reduced with the delayed sowing time, and the yield was increased with the delayed film-covering time and it was highest when the broad bean seeds were sowed on September30 th. Under same film-covering time and sowing time, the total branch number per plant and effective branch number per plant were reduced, but the yield was increased with the increase of planting density; under same film-covering time and planting density, the incidence of freeze injury was reduced with the delayed sowing time, and the yield was highest when broad bean seeds were sowed on September30th; under same sowing time and planting density, the incidence of freeze injury and the yield were all increased with the delayed film-covering time.

Detection of Broad Bean Diseases by Fourier Transform Infrared Spectroscopy Combined with Curve Fitting

Fourier transform infrared （FTIR） spectroscopy was used to study diseased leaves in broad bean. Results showed that the infrared spectra of different broad bean diseased leaves were similar, which were mainly made up of the vibrational absorption bands of protein,lipid and polysaccharide.There were minor differences in-cluding the spectral peak position, peak shape and the absorption intensity in the range of 1 800-1 300 cm-1. There were obvious differences among their second derivative spectra in the range of 1 800-1 300 cm-1. After the procedure of the Fourier self-deconvolution and curve fitting of health bean leaves and broad bean diseased leaves in the range of 1 700-1 500 cm-1, three sub-peaks were obtained at 1 550 cm-1 （protein amide Ⅱ band）, 1 605 cm-1 （lignin） and 1 650 cm-1 （protein amide I band）.The ratios of relative areas of the bands of amide Ⅱ, lignin, and amide I were 38.86%, 28.68% and 32.47% in the spectra of healthy leaves, respec-tively. It was distinguished from the diseased leaves （chocolate spot leaf： 15.42%, 42.98% and 41.61%, ring spot leaf：32.39%, 35.63% and 31.98%, rust leaf： 13.97%, 46.40% and 39.65%, yel owing leaf curl disease leaf： 24.01%,36.55% and 39.44%）. For sub-peak area ratios （A1 563/A1 605, A1 650/A1 605 and A1 563/A1 654）, those of four kinds of diseased leaves were smal er than that of healthy leaves, and there were also differences among four kinds of diseased leaves. The results proved that FTIR combining with curve fitting might be a potential y useful tool for detecting different kinds of broad bean diseases.

从H1N1血凝素序列提取的沙门氏菌传递的COBRA-HA1抗原对甲型流感亚型产生广谱保护作用

A universal vaccine is in high demand to address the uncertainties of antigenic drift and the reduced effectiveness of current influenza vaccines.In this study,a strategy called computationally optimized broadly reactive antigen(COBRA)was used to generate a consensus sequence of the hemagglutinin globular head portion(HA1)of influenza virus samples collected from 1918 to 2021 to trace evolutionary changes and incorporate them into the designed constructs.Constructs carrying different HA1regions were delivered into eukaryotic cells by Salmonella-mediated bactofection using a Semliki Forest virus RNA-dependent RNApolymerase(RdRp)-based eukaryotic expression system,pJHL204.Recombinant protein expression was confirmed by Western blot and immunofluorescence assays.Mice immunized with the designed constructs produced a humoral response,with a significant increase in immunoglobulin G(IgG)levels,and a cell-mediated immune response,including a 1.5-fold increase in CD4^(+) and CD8^(+)T cells.Specifically,constructs #1 and #5 increased the production of interferon-γ(IFN-γ)producing CD4^(+)and CD8^(+)T cells,skewing the response toward the T helper type 1 cell(Th1)pathway.Additionally,interleukin-4(IL-4)-producing T cells were upregulated 4-fold.Protective efficacy was demonstrated,with up to 4-fold higher production of neutralizing antibodies and a hemagglutination inhibition titer>40 against the selected viral strains.The designed constructs conferred a broadly protective immune response,resulting in a notable reduction in viral titer and minimal inflammation in the lungs of mice challenged with the influenza A/PR8/34,A/Brisbane/59/2007,A/California/07/2009,KBPV VR-92,and NCCP 43021 strains.This discovery revolutionizes influenza vaccine design and delivery;Salmonella-mediated COBRA-HA1 is a highly effective in vivo antigen presentation strategy.This approach can effectively combat seasonal H1N1 influenza strains and potential pandemic outbreaks.

Multi-surrogate framework with an adaptive selection mechanism for production optimization

Data-driven surrogate models that assist with efficient evolutionary algorithms to find the optimal development scheme have been widely used to solve reservoir production optimization problems.However,existing research suggests that the effectiveness of a surrogate model can vary depending on the complexity of the design problem.A surrogate model that has demonstrated success in one scenario may not perform as well in others.In the absence of prior knowledge,finding a promising surrogate model that performs well for an unknown reservoir is challenging.Moreover,the optimization process often relies on a single evolutionary algorithm,which can yield varying results across different cases.To address these limitations,this paper introduces a novel approach called the multi-surrogate framework with an adaptive selection mechanism(MSFASM)to tackle production optimization problems.MSFASM consists of two stages.In the first stage,a reduced-dimensional broad learning system(BLS)is used to adaptively select the evolutionary algorithm with the best performance during the current optimization period.In the second stage,the multi-objective algorithm,non-dominated sorting genetic algorithm II(NSGA-II),is used as an optimizer to find a set of Pareto solutions with good performance on multiple surrogate models.A novel optimal point criterion is utilized in this stage to select the Pareto solutions,thereby obtaining the desired development schemes without increasing the computational load of the numerical simulator.The two stages are combined using sequential transfer learning.From the two most important perspectives of an evolutionary algorithm and a surrogate model,the proposed method improves adaptability to optimization problems of various reservoir types.To verify the effectiveness of the proposed method,four 100-dimensional benchmark functions and two reservoir models are tested,and the results are compared with those obtained by six other surrogate-model-based methods.The results demonstrate that our approach can obtain the maximum net present value(NPV)of the target production optimization problems.

Effects of Pb on Growth and Development of Broad Bean Roots

[Objective] This study aimed to investigate the influence of Pb2＋ on the growth and development of broad bean roots. [Method] The effects of Pb2＋ solution of different concentrations on root length, color, bending and mitotic index frequency of root tip cells of broad bean were measured and observed. [Result] Pb2＋ at concentration lower than 20 mg/L promoted the growth and development of roots, increased the cell mitotic indexes, but had little influence on root color and bending. When the Pb2＋ concentration was higher than 20 mg/L, the root growth was inhibited; the root color gradually turned deeper; the roots bended, but the cell mitotic index was decreased. [Conclusion] Pb2＋ promoted the growth of broad bean at low concentration but inhibited the growth at high concentration, and the influence was related to Pb2＋ concentration and time.

长期生态学研究中的经典实验途径——Lawcs和Gilbrt的Broad balk实验

没有长期实验领域的资料而去讨论长期实验是困难的,最早的实验在Rothamsted进行,在那里,测定了植物的氮素来源和化肥及农家肥对小麦产量的影响,该实验始于1843年,与类似的芜％Barnfild实验一样,随后变得更复杂而最终终止了。Barnfield实验以其简单的形式,仍然存在着而基本上没有改变,就象那时能够想象一样,该实验对“长期实验”做出了贡献,简单的设计可能是其成功的关键。当随后R.A.Fi sher试图分析时,这导致他设计新的野外实验统计和随机因子设计(Fisher 1925,1935),并变成了后来田间分析的基础;因为Broadbalk没有反复的实验,现在也不能重复;已经产生了如此多的不同且可改变的变化,就象后来土壤分析所示的,这是在做到最初目的之后继续取样的。存在(Survival)已是实验最显著的成就之一,一起进行的有费歇尔(Fi的叮i＊)的田间统计和对土壤、植物营养、/h麦产量、病害、昆虫、线虫、微生物和杂草的研究(例子见Bawden)。

A Federated Bidirectional Connection Broad Learning Scheme for Secure Data Sharing in Internet of Vehicles

Data sharing in Internet of Vehicles(IoV)makes it possible to provide personalized services for users by service providers in Intelligent Transportation Systems(ITS).As IoV is a multi-user mobile scenario,the reliability and efficiency of data sharing need to be further enhanced.Federated learning allows the server to exchange parameters without obtaining private data from clients so that the privacy is protected.Broad learning system is a novel artificial intelligence technology that can improve training efficiency of data set.Thus,we propose a federated bidirectional connection broad learning scheme(FeBBLS)to solve the data sharing issues.Firstly,we adopt the bidirectional connection broad learning system(BiBLS)model to train data set in vehicular nodes.The server aggregates the collected parameters of BiBLS from vehicular nodes through the federated broad learning system(FedBLS)algorithm.Moreover,we propose a clustering FedBLS algorithm to offload the data sharing into clusters for improving the aggregation capability of the model.Some simulation results show our scheme can improve the efficiency and prediction accuracy of data sharing and protect the privacy of data sharing.