Graph neural network-based scheduling for multi-UAV-enabled communications in D2D networks

In this paper,we jointly design the power control and position dispatch for Multi-Unmanned Aerial Vehicle(UAV)-enabled communication in Device-to-Device(D2D)networks.Our objective is to maximize the total transmission rate of Downlink Users(DUs).Meanwhile,the Quality of Service(QoS)of all D2D users must be satisfied.We comprehensively considered the interference among D2D communications and downlink transmissions.The original problem is strongly non-convex,which requires high computational complexity for traditional optimization methods.And to make matters worse,the results are not necessarily globally optimal.In this paper,we propose a novel Graph Neural Networks(GNN)based approach that can map the considered system into a specific graph structure and achieve the optimal solution in a low complexity manner.Particularly,we first construct a GNN-based model for the proposed network,in which the transmission links and interference links are formulated as vertexes and edges,respectively.Then,by taking the channel state information and the coordinates of ground users as the inputs,as well as the location of UAVs and the transmission power of all transmitters as outputs,we obtain the mapping from inputs to outputs through training the parameters of GNN.Simulation results verified that the way to maximize the total transmission rate of DUs can be extracted effectively via the training on samples.Moreover,it also shows that the performance of proposed GNN-based method is better than that of traditional means.

Outage Probability Analysis for D2D-Enabled Heterogeneous Cellular Networks with Exclusion Zone:A Stochastic Geometry Approach

Interference management is one of the most important issues in the device-to-device(D2D)-enabled heterogeneous cellular networks(HetCNets)due to the coexistence of massive cellular and D2D devices in which D2D devices reuse the cellular spectrum.To alleviate the interference,an efficient interference management way is to set exclusion zones around the cellular receivers.In this paper,we adopt a stochastic geometry approach to analyze the outage probabilities of cellular and D2D users in the D2D-enabled HetCNets.The main difficulties contain three aspects:1)how to model the location randomness of base stations,cellular and D2D users in practical networks;2)how to capture the randomness and interrelation of cellular and D2D transmissions due to the existence of random exclusion zones;3)how to characterize the different types of interference and their impacts on the outage probabilities of cellular and D2D users.We then run extensive Monte-Carlo simulations which manifest that our theoretical model is very accurate.

Influence of Stand-Alone Vertical Gas Vents on Aeration and Denitrification of Organic Municipal Waste Assessed by Two-Dimensional (2D) Lysimeters

Landfilled organic waste, in the presence of oxygen, can undergo aerobic decomposition facilitated by heterotrophic microorganisms. Aerobic degradation of solid waste can quickly consume available oxygen thus curtailing further degradation. The aim of this study was the investigation of a low-cost method of replenishing oxygen consumed in landfilled waste. Three 2D lysimeters were established to investigate the effectiveness of stand-alone, vertical ventilation pipes inserted into waste masses. Two different configurations of ventilation were tested with the third lysimeter acting as an unventilated control. Lysimeters were left uninsulated and observed over the course of 6 months with regular collection of gas and leachate samples. Lysimeters were then simulated for Oxygen (O2) and Nitrous oxide (N2O) to analyze the denitrification contributions of each. The experiment revealed that a single ventilation pipe can increase the mean oxygen level of a 1.7 m × 1.0 m area by up to 13.5%. It also identified that while increasing the density of ventilation pipes led to increased O2 levels, this increase was not significant at the 0.05 probability level. A single vent averaged 13.67% O2 while inclusion of an additional vent in the same area only increased the average to 14.59%, a 6.7% increase. Simulation helped to verify that lower ventilation pipe placement density may be more efficient as in addition to the effect on oxygenation, denitrification efficiency may increase. Simulations of N2O production estimated between 8% - 20% more N2O being generated with lower venting density configurations.

基于GADF与2D CNN-改进SVM的道岔故障诊断方法研究

针对道岔故障特征不易提取以及道岔故障诊断准确率较低的问题,提出一种格拉姆角差场(Gramian Angular Difference Fields, GADF)与二维卷积神经网络(Two Dimensional Convolutional Neural Network, 2D CNN)-改进支持向量机(Support Vector Machine, SVM)的道岔故障诊断组合方法。首先,结合现场实际应用情况,选取道岔设备正常转换与典型故障的转辙机功率曲线,建立转辙机功率曲线样本数据库;采用GADF编码将一维转辙机功率曲线信号转换为具有时间相关性的二维特征图,分别选择16×16、32×32以及64×64大小的特征图并提取图像数据。其次,在LeNet-5模型的基础上设计2D CNN网络结构,并将图像数据输入至基于2D CNN的道岔故障特征提取模型中,经多层的卷积层、池化层以及全连接层提取特征指标,建立道岔故障诊断样本数据库。最后,通过北方苍鹰优化(Northern Goshawk Optimization, NGO)算法优化SVM算法的惩罚因子与核函数方差,构建基于NGO-SVM的道岔故障诊断模型。实验结果分析表明,将转辙机功率曲线数据经GADF编码为64×64大小的特征图,并通过2D CNN模型提取道岔典型特征数据,较其他数据处理方法具有较高的故障诊断准确率,同时提高了故障诊断实时性;将建立的道岔故障诊断样本数据库输入至NGO-SVM道岔故障诊断模型,其故障诊断准确率高达97.5%,较其他故障诊断模型具有更好的故障诊断性能,为道岔故障诊断提供了一种新方法,对现场道岔设备的日常维修具有一定的指导意义。

基于2D人体图像特征学习的女西装合体性判别

为解决服装远程定制和网络购衣过程中服装合体性无法预先判断的问题,文章提出利用2D人体图像判断女西装合体性。首先采集了462名18~25周岁青年女性的2D图像及3D尺寸,建立人体数据集。并制作10件虚拟女西装,建立服装数据集。其次提取用于女西装合体性判别的服装特征、2D人体图像关键点距离特征和整体特征。基于感性工学原理,采用专家评价法获取女西装合体性评价标签。最后利用贝叶斯分类器建立女西装合体性判别模型。研究结果显示,将文章提出的特征应用于基于贝叶斯算法的女西装合体性判别模型判别准确率可达到84.8%,说明提出的服装合体性评价指标有效,可以用于基于2D图像的服装合体性预测。

H-CRAN中IRS辅助的D2D系统资源分配与RCG波束成形优化

以异构云无线电接入网(heterogeneous cloud radio access network,H-CRAN)中智能反射面(intelli‐gent reflecting surface,IRS)辅助的端到端(device-to-device,D2D)通信系统为背景,研究了该系统中以和速率最大化为目标的资源分配与黎曼共轭梯度(Riemannian conjugate gradient,RCG)波束成形优化方法。以最大化系统和速率为优化目标,构造子信道复用系数、发射功率门限以及IRS反射系数模约束等多约束优化问题。对于该非线性混合整数规划问题,提出了一种基于相对信道强度的延迟接受算法,以获得信道复用系数。随后将目标优化问题分解为两个子问题进行交替优化。对于发射功率优化子问题,使用逐次凸逼近(successive convex approximation,SCA)方法进行求解。对于IRS波束成形子问题,将IRS相移约束转化为复圆流形后,采用RCG算法进行求解。仿真结果表明,当IRS反射阵源数为50、基站最大发射功率为46 dBm时,与现有信道分配方案和随机信道分配方案相比,所提信道分配方案的和速率性能分别提高了5.2 bit/(s·Hz)和14.6 bit/(s·Hz)。与无IRS通信场景相比,部署IRS的和速率性能显著提高约31.2 bit/(s·Hz)。

基于TM2D1介导铁死亡探讨丙泊酚抑制结直肠癌的分子机制

目的:基于丙泊酚通过TM2D1介导铁死亡抑制结直肠癌的分子机制。方法:测定TM2D1在正常和CRC组织中的表达,将人CRC SW480细胞暴露于50μm丙泊酚中,检测细胞活力。用过表达TM2D1的载体转染SW480细胞并用丙泊酚处理,并评估异丙酚处理SW480细胞,观察细胞活力、集落形成、细胞增殖、铁水平、ROS生成和和铁死亡。结果:TM2D1在结直肠癌组织中高表达。异丙酚对SW480细胞活力有抑制作用,TM2D1在丙泊酚作用下显著下调,丙泊酚还能抑制结直肠癌细胞增殖和集落形成,提高细胞铁和ROS水平。此外,丙泊酚还改变了与铁死亡相关的蛋白的表达,包括CRC细胞中CHAC1和PTGS2的表达上调,以及GPX4的表达抑制,TM2D1过表达阻断了异丙酚对CRC细胞的作用,差异有统计学意义(P<0.05)。结论:丙泊酚可能通过下调TM2D1的表达引发CRC细胞铁死亡。

基于D2D的车联网资源分配与模式选择方案

针对当前设备到设备(D2D:Device-to-Device)通信在车联网端应用时只考虑提高D2D的复用模式的资源利用率或系统遍历容量,而未将D2D的其他模式纳入到系统内的问题,提出一种兼顾资源分配与模式选择的算法。通过优先复用模式的方式提高通信资源利用率,对不满足复用模式的D2D用户对(D-UE:Device-to-device UsErs)采用蜂窝模式,采取和蜂窝用户(C-UE:Cellular UsErs)相同的方案满足D-UE的基础需求,考虑D-UE用户与蜂窝用户和基站(BS:Base Station)距离、D-UE间的信噪比等因素对D-UE分配不同的资源接入模式。理论计算与仿真结果均表明所提资源分配与模式选择方案能有效提升系统容量及D-UE的服务质量(QoS:Quality of Service)。

miR-647、miR-122、KMT2D表达与前列腺癌临床病理特征及预后的相关性分析

目的探讨前列腺癌组织中微小RNA-647(miR-647)、微小RNA-122(miR-122)、组蛋白赖氨酸甲基转移酶2D(KMT2D)与其临床病理特征及预后的相关性。方法选取100例前列腺癌患者作为研究对象,均于术中采集肿瘤组织及癌旁正常组织标本送检,统计肿瘤组织及癌旁正常组织内miR-647、miR-122、KMT2D表达水平差异;分析miR-647、miR-122、KMT2D表达与肿瘤分期、肿瘤直径、淋巴结转移、微血管侵犯等的关系;并随访2年,比较复发转移与未复发转移患者miR-647、miR-122、KMT2D表达间差异。结果肿瘤组织内miR-647表达水平为(0.62±0.13),低于癌旁正常组织的(1.02±0.19),miR-122、KMT2D表达水平为(6.45±1.24)、(1.45±0.23),低于对照组的(1.23±0.22)、(0.89±0.12)(P<0.05);肿瘤分期Ⅲ期、肿瘤直径≥3 cm、有淋巴结转移及有微血管侵犯患者miR-647表达水平低于肿瘤分期Ⅰ~Ⅱ期、肿瘤直径<3 cm、无淋巴结转移及无微血管侵犯患者,miR-122、KMT2D表达水平高于肿瘤分期Ⅰ~Ⅱ期、肿瘤直径<3 cm、无淋巴结转移及无微血管侵犯患者(P<0.05);100例患者术后随访2年,共出现24例复发转移,复发转移率为24.00%(24/100);复发转移患者miR-647表达水平为(0.51±0.11),低于未复发转移患者的(0.73±0.15),miR-122、KMT2D表达水平为(8.42±1.39)、(1.89±0.32),高于未复发转移患者的(4.68±1.02)、(1.12±0.15)(P<0.05)。结论前列腺癌组织内miR-647、miR-122、KMT2D存在不同程度表达,均与肿瘤分期、肿瘤直径及淋巴结转移等关系密切,且可明显影响患者预后,还需临床高度重视。

移动机器人2D激光SLAM算法仿真与实现

随着无人驾驶技术的迅速发展,同步定位与建图技术因其精度高、稳定性好的优点备受人们关注。基于激光雷达传感器,选择主流的Gmapping和Cartographer算法,搭建实验环境,对两种算法进行对比仿真与实验建图,并对两种算法的建图效果进行深度的分析。基于滤波器的Gmapping算法计算量小,依赖于里程计信息,适用于小尺度、低特征环境中;基于图优化的Cartographer算法累计误差低,精度高,适用于精度和稳定性要求较高的场合。

基于改进白鲸优化算法的D2D通信功率控制

D2D(Device-to-Device)通信作为未来移动通信网络的关键技术,为用户提供了直接通信的便利性和资源共享的高效性.然而,D2D通信的功率控制一直是影响通信质量和系统性能的关键问题.为解决这一问题,将精英反向学习、自适应权重两种策略引入到白鲸优化算法(Beluga Whale Optimization,BWO)中,并利用莱维飞行的随机步长策略来增加算法寻优的多样性,提出了基于改进白鲸优化算法的D2D通信功率控制方法.该方法利用最优解的信息引导搜索过程,可提高搜索效率和全局收敛,并能够有效提高通信效率和系统稳定性.为了验证所提出方法的有效性,开展了大量的数值仿真实验.结果显示,基于改进白鲸优化算法的D2D通信功率控制方法在增加系统吞吐量、减少干扰方面有显著的改善.同时,提出的算法相对于已有的算法有着更出色的收敛性与鲁棒性,在不同通信环境和参数设置下都能表现出更稳定的性能.

基于改进NSGA-Ⅲ的D2D协同MEC多目标优化研究

在当前的移动边缘计算(Mobile Edge Computing,MEC)模型中,由于任务是直接上传到MEC服务器执行,存在边缘服务器的计算压力大、空闲移动设备上的资源未得到充分利用等问题。使用边缘网络中的空闲设备进行协同计算,能够实现用户闲置资源的合理利用,增强MEC的计算能力。因此,提出了一种利用终端直通(Device-to-Device,D2D)进行协同计算的部分卸载MEC模型(D2D Collaborative MEC for Partial Offloading,DCM-PO)。在该模型中,除本地计算和MEC服务器计算外,还能将部分任务上传到空闲D2D设备进行辅助计算。首先,以最小化边缘网络的时延、能耗和费用为目标建立多目标优化问题。然后,在多染色体混合编码、自适应交叉率和变异率等方面对基于参考点的非支配排序遗传算法(Non-dominated Sorting Genetic AlgorithmⅢ,NSGA-Ⅲ)进行改进,使之适合DCM-PO模型中的多目标优化问题求解。最后,仿真结果表明,相比基准MEC模型,DCM-PO模型在多项性能指标上有明显优势。

Two/Quasi-two-dimensional perovskite-based heterostructures:construction,properties and applications

Two-dimensional(2D)/quasi-2D organic-inorganic halide perovskites are regarded as naturally formed multiple quantum wells with inorganic layers isolated by long organic chains,which exhibit layered structure,large exciton binding energy,strong nonlinear optical effect,tunable bandgap via changing the layer number or chemical composition,improved environmental stability,and excellent optoelectronic properties.The extensive choice of long organic chains endows 2D/quasi-2D perovskites with tunable electron-phonon coupling strength,chirality,or ferroelectricity properties.In particular,the layered nature of 2D/quasi-2D perovskites allows us to exfoliate them to thin plates to integrate with other materials to form heterostructures,the fundamental structural units for optoelectronic devices,which would greatly extend the functionalities in view of the diversity of 2D/quasi-2D perovskites.In this paper,the recent achievements of 2D/quasi-2D perovskite-based heterostructures are reviewed.First,the structure and physical properties of 2D/quasi-2D perovskites are introduced.We then discuss the construction and characterizations of 2D/quasi-2D perovskite-based heterostructures and highlight the prominent optical properties of the constructed heterostructures.Further,the potential applications of 2D/quasi-2D perovskite-based heterostructures in photovoltaic devices,light emitting devices,photodetectors/phototransistors,and valleytronic devices are demonstrated.Finally,we summarize the current challenges and propose further research directions in the field of 2D/quasi-2D perovskite-based heterostructures.

Co-Ru alloy nanoparticles decorated onto two-dimensional nitrogen doped carbon nanosheets towards hydrogen/oxygen evolution reaction and oxygen reduction reaction

Constructing highly-efficient electrocatalysts toward hydrogen evolution reaction(HER)/oxygen evolution reaction(OER)/oxygen reduction reaction(ORR)with excellent stability is quite important for the development of renewable energy-related applications.Herein,Co-Ru based compounds supported on nitrogen doped two-dimensional(2D)carbon nanosheets(NCN)are developed via one step pyrolysis procedure(Co-Ru/NCN)for HER/ORR and following low-temperature oxidation process(Co-Ru@RuO_(x)/NCN)for OER.The specific 2D morphology guarantees abundant active sites exposure.Furthermore,the synergistic effects arising from the interaction between Co and Ru are crucial in enhancing the catalytic performance.Thus,the resulting Co-Ru/NCN shows remarkable electrocatalytic performance for HER(70 mV at 10 mA cm^(-2))in 1 M KOH and ORR(half-wave potential E_(1/2)=0.81 V)in 0.1 M KOH.Especially,the Co-Ru@RuO_(x)/NCN obtained by oxidation exhibits splendid OER performance in both acid(230 mV at 10 mA cm^(-2))and alkaline media(270 mV at 10 mA cm^(-2))coupled with excellent stability.Consequently,the fabricated two-electrode water-splitting device exhibits excellent performance in both acidic and alkaline environments.This research provides a promising avenue for the advancement of multifunctional nanomaterials.

Two-dimensional laser-induced periodic surface structures formed on crystalline silicon by GHz burst mode femtosecond laser pulses

Femtosecond laser pulses with GHz burst mode that consist of a series of trains of ultrashort laser pulses with a pulse interval of several hundred picoseconds offer distinct features in material processing that cannot be obtained by the conventional irradiation scheme of femtosecond laser pulses(single-pulse mode).However,most studies using the GHz burst mode femtosecond laser pulses focus on ablation of materials to achieve high-efficiency and high-quality material removal.In this study,we explore the ability of the GHz burst mode femtosecond laser processing to form laser-induced periodic surface structures(LIPSS)on silicon.It is well known that the direction of LIPSS formed by the single-pulse mode with linearly polarized laser pulses is typically perpendicular to the laser polarization direction.In contrast,we find that the GHz burst mode femtosecond laser(wavelength:1030 nm,intra-pulse duration:220 fs,intra-pulse interval time(intra-pulse repetition rate):205 ps(4.88 GHz),burst pulse repetition rate:200 kHz)creates unique two-dimensional(2D)LIPSS.We regard the formation mechanism of 2D LIPSS as the synergetic contribution of the electromagnetic mechanism and the hydrodynamic mechanism.Specifically,generation of hot spots with highly enhanced electric fields by the localized surface plasmon resonance of subsequent pulses in the bursts within the nanogrooves of one-dimensional LIPSS formed by the preceding pulses creates 2D LIPSS.Additionally,hydrodynamic instability including convection flow determines the final structure of 2D LIPSS.

2D MoS_(2)及其异质结构在生物传感领域的应用

简要介绍了多相态二维(2D)MoS_(2)纳米材料的结构、制备和相态调控策略。MoS_(2)具有可调节的电子转移和光学特性,可以通过制备过程参数的调制对产物相态调控。不同相态的MoS_(2)性质差异巨大,因而在传感器上的应用不同。详细介绍了基于不同相态MoS_(2)及其异质结构的2D纳米材料在生物传感领域的代表性应用。重点剖析了不同相态MoS_(2)在不同类型传感器中所起的作用。生物传感技术的发展对于疾病的诊断和临床治疗,以及环境和食品安全监测等领域都至关重要。希望可以为高性能MoS_(2)基生物传感器的研发和构建提供参考。

磁共振联合血清KMT2D检测在膀胱癌诊断中的临床意义

目的分析磁共振联合血清组蛋白赖氨酸甲基转移酶2D(KMT2D)检测对膀胱癌的诊断价值。方法以本院154例膀胱癌患者作为膀胱癌组(BC组),根据临床金标准分为肌层浸润性膀胱癌94例(MIBC组)和非肌层浸润性膀胱癌60例(NMIBC组)。对照组为同期100例健康体检者。比较各组血清KMT2D表达水平差异,受试者工作特征(ROC)曲线分析血清KMT2D对MIBC膀胱癌诊断价值。磁共振、血清KMT2D及二者联合诊断与临床“金标准”的一致性采用Kappa检验。结果BC组血清KMT2D表达水平低于对照组,MIBC组KMT2D表达水平低于NMIBC组,差异有统计学意义(P<0.05)。血清KMT2D诊断MIBC膀胱癌的截断值为0.68,曲线下面积为0.855。磁共振检查诊断MIBC膀胱癌与临床“金标准”检查一致性较好(Kappa=0.744,P<0.05)。磁共振检查联合血清KMT2D诊断MIBC膀胱癌与临床“金标准”结果相比的Kappa值为0.932(P<0.05)。磁共振检查联合血清KMT2D诊断MIBC膀胱癌的灵敏度、准确率及特异度优于单独诊断。结论磁共振联合血清KMT2D检测对MIBC膀胱癌诊断价值较高,有助于术前明确膀胱癌浸润程度。

Deep Learning Accelerates the Discovery of Two- Dimensional Catalysts for Hydrogen Evolution Reaction

Two-dimensional materials with active sites are expected to replace platinum as large-scale hydrogen production catalysts.However,the rapid discovery of excellent two-dimensional hydrogen evolution reaction catalysts is seriously hindered due to the long experiment cycle and the huge cost of high-throughput calculations of adsorption energies.Considering that the traditional regression models cannot consider all the potential sites on the surface of catalysts,we use a deep learning method with crystal graph convolutional neural networks to accelerate the discovery of high-performance two-dimensional hydrogen evolution reaction catalysts from two-dimensional materials database,with the prediction accuracy as high as 95.2%.The proposed method considers all active sites,screens out 38 high performance catalysts from 6,531 two-dimensional materials,predicts their adsorption energies at different active sites,and determines the potential strongest adsorption sites.The prediction accuracy of the two-dimensional hydrogen evolution reaction catalysts screening strategy proposed in this work is at the density-functional-theory level,but the prediction speed is 10.19 years ahead of the high-throughput screening,demonstrating the capability of crystal graph convolutional neural networks-deep learning method for efficiently discovering high-performance new structures over a wide catalytic materials space.

昆虫杆状病毒表达的猪圆环病毒2d型Cap蛋白-VLP对仔猪的免疫保护效果

猪圆环病毒2d基因型(Porcine circovirus type 2d,PCV2d)为猪场当前流行的PCV2优势基因型,为了研究针对PCV2d基因型的病毒样颗粒(Virus-like particle,VLP)疫苗,提高猪圆环病毒病的防控效果。本研究利用杆状病毒表达系统表达了能够自组装为VLP的PCV2d-Cap蛋白,然后将9头21日龄健康仔猪随机分为VLP组、攻毒对照组和空白对照组(n=3)。VLP组的每头仔猪经颈部肌肉注射400μg Cap蛋白-佐剂复合物(PCV2d-VLP),攻毒组注射等体积的PBS与佐剂混合物,空白组注射等体积的PBS。共接种2次,每次间隔14 d。于第2次免疫后21 d,VLP组和攻毒对照组的仔猪通过鼻腔感染PCV2 HBDX-2018株(106TCID50/头),评价PCV2d-VLP诱导的免疫效果。结果表明PCV2d-VLP能够诱导仔猪产生高水平的特异性IgG抗体和中和抗体,刺激IFN-γ水平升高,引起外周血淋巴细胞增殖能力增强,降低病毒经鼻腔和直肠的排毒率及病毒血症阳性率,减轻腹股沟淋巴结和脾脏的病理损伤,降低组织中的病毒载量。上述研究表明,应用杆状病毒表达的PCV2d-Cap蛋白能自组装为VLP,并能诱导仔猪产生良好的免疫保护效应,具有进一步开发为PCV2d-VLP疫苗的潜力。

2D/3D多模态医学图像配准算法研究

2D/3D多模态配准在医学影像导航手术中起着重要作用,主要用于提供术前三维图像和术中二维图像的实时信息,帮助医生精准定位病灶,规划手术路径,从而提高手术的安全性和效率。提出了一种2D/3D多模态医学图像配准算法,首先利用Swin Transformer优秀的特征提取能力,构建了初始姿态估计模型,实现姿态参数的快速预测;接着,为了提升整个配准方法的鲁棒性,引入基于Grangeat关系的粗配准方法;最后设计了基于梯度下降的精配准模块,以提升整个配准过程的精确性,且在该模块将Sobel微分算子与归一化互相关结合,提升了参数优化过程中的灵敏度。实验结果表明,所提配准方法在正位和侧位配准中的误差满足配准要求,配准成功率有显著提升。