The characteristics of edge computing make it have broad military application prospects.FL(Federated Learning)is introduced into edge computing.Considering the limited resources of IoT devices,FL accuracy and device energy consumption need to be taken into account.A framework combin-ing deep reinforcement learning,federated learning,and self attention mechanism(DRL-FLSL)is pro-posed to select devices and allocate resources to them,with the goal of balancing FL accuracy and device energy consumption.This framework introduces LSTM(Long Short Term Memory)to predict network state and adds a multi head self attention mechanism for more accurate information extraction.The simulation experimental results show that DRL-FLSL has super training effects and can effectively balance FL accu-racy and equipment energy consumption.

The article presents a method for evaluating network security state using an improved fuzzy evaluation membership function and BP neural networks. The method involves constructing a BP neural network model based on network security behavior data and introducing an enhanced fuzzy evaluation membership function that more accurately describes the relationship between network behavior features and security states. Through training and adjusting the model, the method enables accurate evaluation and prediction of network security states, effectively identifying and predicting security threats in networks. Experimental results demonstrate that the method has high accuracy and reliability, making it potentially valuable in the field of network security.

针对物联网与区块链融合场景下分布式网络共识的安全性和效率性难题,提出了一种基于多维信用评分的改进实用拜占庭容错机制(practical byzantine fault tolerance,PBFT)的共识算法.首先通过多维角度对服务节点进行量化评分,然后按照分值从高到低将服务节点分为三种类型的节点并选择出参与共识的节点,最后在共识的过程中进行双重的验证确认,在高效率完成共识的同时,保证分布式网络的安全性.分析结果表明多维信用评分算法(multi-dimensional credit consensus,MdCCA)具有很好的安全性,而且算法的执行效率也得到了提高.

In the face of huge Internet of Things (IoT) application scenarios, although trusted access and management of IoT devices can be realized through a single blockchain, there are problems of increased time consumption and low throughput. To solve the above problems, this paper proposes an IoT device authentication strategy based on master-slave chain structure. Firstly, the master-slave chain network structure of IoT device authentication is proposed based on blockchain technology. Secondly, the reputation evaluation mechanism is introduced to estimate the reputation value of slave chain nodes and IoT devices, so as to provide basis for the selection of consensus nodes. Thirdly, by selecting the consensus node with high reputation value and optimizing the consensus stage, the Practical Byzantine Fault Tolerance (PBFT) consensus algorithm is improved to increase the consensus throughput and improve the consensus efficiency of the authentication process. Finally, the authentication smart contract is designed to ensure fair and credible implementation of authentication rules. Security analysis shows that these strategies have high security against mainstream IoT attacks, reduce the delay of authentication request, and improve the efficiency and stability of authentication.

In recent years, it has become a new trend to integrate blockchain and the Internet of Things (IoT) in order to deal with the challenges of network security, privacy protection and identity recognition of the IoT. Data consistency algorithm, as the key support technology after the integration, has become a hot research issue. Due to the limited resources of IoT devices, the consensus algorithm of ...

伴随中国老龄化社会的逐步到来,利用大数据和人工智能技术匹配养老资源供给与居家养老需求成为亟待解决的问题.分析目前居家养老需求,建立居家养老服务数据模型,采用大数据技术构建居家养老服务分类组别,提出居家养老服务复合判别分析算法,设计面向服务架构的养老服务发现与推荐算法,采用爬取网络数据和仿真生成数据相结合方式验证了所提出算法的有效性.

针对工科研究生面向对象分析与设计课程融入思政元素问题,分析课程思政实施方案,提出基于建构主义学习原理的课程思政二元融合方法,介绍具体实施途径,阐述课程思政融入点,通过3个教学案例探讨课程思政融入方法,最后说明课程思政二元融合方法优势与实践效果.

在新工科和"金课"建设背景下,以OBE理念审视当前的工科大学英语教学现状和应用需求,提出当前工科大学英语教学"6个没有改变"的问题.基于OBE理念,提出在目标层实现学习成果产出与应用、在方法层践行新型教学理论与模式、在工具层构建"线上线下"协同教学课堂的三层教学结构.在三层教学结构的基础上,设计基于任务驱动和问题解决的"1+X"教学模式,构建"线上线下"协同教学课堂,实现具有持续改进能力的新型教学模式.

指挥信息系统的服务共享和开放导致其面临更多的服务资源毁伤和信息欺骗攻击风险,有效的攻击风险评估亟待解决.针对指挥信息系统应用层攻防中信息不完备性和不确定性带来的攻击效能评估困难问题,建立指标因素数量为2+5+17的指挥信息系统应用层攻击3级评估指标体系,提出双因子主观、客观赋权法确定评估指标权重,应用反向传播人工神经网络理论建立具有学习机制和持续改进能力的模糊隶属函数,设计了单一攻击效能和多种攻击效能的模糊综合评估方法.实验结果表明:该方法不仅能够实现单一攻击效果的模糊量化和多种攻击的模糊排序,而且指标集更完整、赋权方法更综合、模糊隶属函数更具适用性.

目前恶意软件的安全威胁越来越严重,提高恶意软件的识别准确率已成为亟待解决的问题.针对朴素贝叶斯方法恶意软件识别准确率不高的问题,提出一种利用萤火虫算法改进加权贝叶斯的恶意软件识别方法,以恶意软件的行为数据作为特征,通过萤火虫算法不断地迭代来优化样本属性的权值,将权值带入加权贝叶斯模型中识别恶意软件,通过对virusshare网站的1300个样本进行实际检测,相比于朴素贝叶斯和互信息加权贝叶斯恶意软件识别方法,其平均识别准确率分别提高了17％和6％,表明新方法具有更好的识别效果.

目前,指挥控制系统呈现网络化、服务化发展趋势.将服务质量(Quality of Service,QoS)约束条件作为Web服务选择的依据时,为了缩小Skyline Web服务集,提高服务选择的效率,提出一种面向Skytine Web服务的新型服务选择方法.基于代表性Skyline服务选择原理,结合效用函数与服务顺序结构组合技术,设计Web服务的优化选择算法,实现Skyline服务集的最大选择概率.基于开放的QWS数据集,针对服务选择效用值、服务选择执行时间指标进行实验,实验结果表明:相比于传统Skyline方法服务选择算法,新算法的服务选择执行时间至少减少了25％,服务选择平均效用值约提升10％.

针对目前专业学位硕士的计算机安全课程教学重理论轻实践、教学内容裁剪困难、学习成果评价困难等问题,分析课程技术更新较快、知识综合性强以及实践性较强等特点,基于建构主义、任务驱动教学法的基本原理,提出实践任务驱动下的计算机安全课程1+X教学方法,阐述课程内容、实验案例、课程体系以及基于建构主义的实践任务驱动下的1+X教学模式,构建基于任务驱动四化法的1+X教学技术.

随着互联网的发展和普及,互联网中恶意代码的安全威胁越来越严重,提高恶意代码的识别准确率已成为急需解决的问题.因此,本文在虚拟化环境中的静态行为跟踪和特征分析的基础上,引入基于信息增益的N-gram语义特征提取方法和文本频率特征提取方法,对恶意代码进行多元语义切分,映射为恶意代码的Op-code操作码特征,先进行对处理之后的特征数据集运用分类算法进行分类检测和分析,之后再结合机器学习分类方法,实现恶意代码样本的有效归属判别.

为了降低硬实时周期性任务主副版本容错调度的副版本调整开销,提出了一种BEDF-NENF容错调度算法.采用反向最早截止期优先(BEDF)策略为副版本预分配处理器时间,运行时则采用零调整最早通知时间优先(NENF)策略调度主版本.结果表明,BEDF-NENF算法能够按照最后机会策略调度副版本.当主版本错误概率不大于0.05时,BEDF-NENF算法的副版本调整平均比较次数和副版本调整时间比率均为0,与BEDF-RM算法、BEDF-EDF算法、BEDF-ENF算法的主版本完成率之差约为1％.BEDF-NENF算法不仅能够取得与同类调度算法接近的主版本完成率,而且能够通过省略副版本重新调整操作来降低调度的复杂性,节省调度时间.

近年来,各行业已投入使用的信息化软硬件设施种类数目众多,每年花费的各类维护开销庞大.以交通行业为例,指出当前交通信息化设施维护的困境,探索交通信息化设施维护定额的研究框架,为信息化系统正常运转及上级决策提供依据.

研究了不同引气剂和再生粗集料取代率对再生混凝土工作性能的影响.结果表明:在适宜掺量下AE-1可减小再生混凝土坍落度的损失;AE-2和AE-3可显著增加再生混凝土初始坍落度,但保坍性能没有明显提高;再生混凝土的坍落度损失值随着再生粗集料取代率增大而增大.

恶意软件的日益增长是对网络世界最大的威胁,基于签名的检测对于恶意软件检测率较低,局限性大,因此提出基于机器学习的恶意软件检测技术来代替传统的签名检测.根据沙箱中提取软件的特征类型包括注册表和A PI函数调用,并量化数据,使用机器学习的模型对此数据进行分类识别,并取得了较好的分类效果.

The traditional performance analysis of real-time systems relied on the input variable of worst-case execution time, which turned to be too pessimistic. Aiming at the problem of remarkably redundant design in real-time, a new model to characterize variable workload was created including workload curves, inverse workload curves and workload ratio curves. In the proposed model, event type, number and distribution were used as decision-variable, and relevant algorithm was proposed to solve the above model. In addition, the realistic applications were analyzed in mix scheduling based on VWM ( variable workload model) . The result indicates that the VWM can remarkably reduce execution workload of tasks, thus reducing the resource requirement of real-time systems.

Task workload analysis for primary-alternate fault-tolerance is usually performed based on WCET (Worst-Case Execution Time), which causes excessive redundancy of processor resource reservation in conventional real-time systems. To address this issue, a novel primary-alternate fault-tolerant model based on a variable workload is built, theorems for both the alternate schedulability and the primary execution success ratio is proposed, and their correctness are proved. Motivated by these theorems, a new algorithm called BCEVW (BCE with Variable Workload) is proposed through improving the algorithm BCE. Further, a new error probability model for the primary version is proposed, which is closer to the practice than does the conventional model. Using two metrics including schedulability of a task set and execution success ratio of the primary version, simulation results show that (1) for scheduling task sets in the alternate version, backwards-EDF algorithm performs better than backwards-RM algorithm, and the necessary and sufficient conditions for its schedulability is that the processor utilization is no more than 1; (2) in the case of variable workload, BCEVW algorithm can significantly improve the schedulability of task sets in the primary version; (3) for the new error probability model, the results of the proposed scheduling algorithm show that there is an obvious linear relationship between the error probability and execution success ratio of the primary version, and further the greater processor utilization of the primary version is, the more significant linear relationship is.

鉴于SOAP协议本身固有的安全脆弱性可致网络及服务面临泛洪攻击的威胁,在深入研究提供网络系统安全及可靠性建设所需的效能评估技术基础上,针对灰色评估法的局限性,将模糊评估法引入SOAP泛洪攻击效能评估,建立SOAP泛洪攻击效能模糊评估模型,其核心为攻击评判矩阵的确立、隶属度矩阵的定制、指标权重的配置以及综合评估的实施.基于建立的攻击效能模糊评估模型,利用采集测量的样本数据进行了实例验证.验证结果表明:该方法更具适用性,不仅可以提供攻击效能的确定性评估结论,而且可以合理地区分攻击的有效程度.