技术领域technical field
本发明涉及无线传感器网络(Wireless Sensor Network,WSN)及其可靠性评测领域，尤其涉及一种测评无线传感器网络可靠性的方法及系统。The present invention relates to the field of wireless sensor network (Wireless Sensor Network, WSN) and its reliability evaluation, in particular to a method and system for evaluating the reliability of wireless sensor network.
背景技术Background technique
工业无线传感器网络(Industrial Wireless Sensor Network,以下简称工业WSN)是由具有无线通信与计算能力的传感器节点构成、部署在工业现场环境为某种工业应用提供解决方案的自组织分布式网络智能系统，是满足工业应用高可靠、低能耗、硬实时等特殊需求的一类无线传感器网络技术。它集成了传感器、无线通信和分布式信息处理技术，是一种全新的计算模式。在过去的十年中，工业无线技术和传感器技术有了巨大的发展，无线传感器网络在工业应用的优势已经越来越明显。与传统的有线系统相比，工业无线传感器网络优势明显：(1)低成本；(2)高可靠、易维护；(3)高灵活、易使用。利用工业无线传感器网络，人们可以以较低的投资和使用成本实现对工业全流程的“泛在感知”，获取传统上由于成本原因无法在线监测的重要工业过程参数，并以此为基础实施优化控制，来达到提高产品质量和节能降耗的目标。目前，如何设计搭建稳定、可靠的工业无线传感器网络是该领域研究热点之一。Industrial Wireless Sensor Network (Industrial Wireless Sensor Network, hereinafter referred to as industrial WSN) is composed of sensor nodes with wireless communication and computing capabilities, and is deployed in an industrial field environment to provide solutions for certain industrial applications. A self-organizing distributed network intelligent system, It is a type of wireless sensor network technology that meets the special needs of industrial applications such as high reliability, low energy consumption, and hard real-time. It integrates sensors, wireless communication and distributed information processing technology, and is a new computing model. In the past decade, industrial wireless technology and sensor technology have developed tremendously, and the advantages of wireless sensor networks in industrial applications have become more and more obvious. Compared with traditional wired systems, industrial wireless sensor networks have obvious advantages: (1) low cost; (2) high reliability and easy maintenance; (3) high flexibility and easy use. Using industrial wireless sensor networks, people can realize the "ubiquitous perception" of the entire industrial process with low investment and use costs, obtain important industrial process parameters that traditionally cannot be monitored online due to cost reasons, and implement optimization based on this control to achieve the goal of improving product quality and saving energy. At present, how to design and build a stable and reliable industrial wireless sensor network is one of the research hotspots in this field.
虽然在过去的十年中，工业无线技术有了巨大的发展，但是由于需要满足一些特殊的要求，无线传感器网络在工业中的真正应用仍然尚不成熟。这其中工业无线传感器网络的可靠性研究是一个关键技术难题。基于无线通信的工业传感器网络的可靠性易受到背景噪声、路径损耗、多路干扰、物理冲突等因素影响而变得不稳定。经过多年的研究，科学工作者提出了许多可靠的应用方案与协议算法用于提高工业无线传感器网络的可靠性，但是仅仅从单个方面改善网络的协议对提高网络的可靠性效果不是特别明显，而工业无线传感器网络的可靠性测试与评估为衡量和完善网络性能提供了便利有效的手段。如果这项工作不完善，该网络的实际应用就具有极大的风险，既浪费了工业成本，又损害了工业效益，造成严重的后果。Although the industrial wireless technology has developed tremendously in the past decade, the real application of wireless sensor networks in industry is still immature due to the need to meet some special requirements. Among them, the reliability research of industrial wireless sensor network is a key technical problem. The reliability of industrial sensor networks based on wireless communication is easily affected by factors such as background noise, path loss, multi-path interference, and physical conflicts, and becomes unstable. After years of research, scientists have proposed many reliable application schemes and protocol algorithms to improve the reliability of industrial wireless sensor networks, but only improving the network protocol from a single aspect is not particularly effective in improving the reliability of the network. The reliability test and evaluation of industrial wireless sensor network provides a convenient and effective means for measuring and improving network performance. If this work is not perfect, the actual application of the network will have great risks, which will waste industrial costs and damage industrial benefits, resulting in serious consequences.
但是，目前现有的对工业WSN的可靠性测试与评估方法普遍存在如下缺陷：(1)尽管经过多年的研究，科学工作者针对传统的ad-hoc网络提出了许多可靠性评估方法，但是这些方法却难以按照工业应用的要求，针对网络各性能指标，根据评估方法对网络的不稳定因素进行分析，评估以及量化其满足工业应用的程度。因此，这些方法难以用于对工业无线传感器网络的可靠性进行衡量和比较，从而也难以为工业无线传感器网络的设计提供辅助评定。(2)在工业无线传感器网络中，网络所处的监测环境往往十分恶劣，开发者无法在现场收集数据测试和评估网络的可靠性，这就迫切的需要一种手段可以人为的模拟这些恶劣环境的干扰，即将某些故障注入到无线传感器网络中，另外还可以实时的记录并存储当前网络的运行信息以及对注入故障的反应信息，并加以分析，从而评估并提高整个工业WSN的可靠性。但是通过对大量文献研究发现，目前这一领域还没有受到足够的关注，仅有的针对工业WSN的故障注入(Fault Injection,FI)方法不是采用仿真的方法实现就是所注入的故障无法有效地模拟工业WSN在实际应用时所遇到的情况。(3)在研究工业WSN容错性的过程中，目前普遍遇到的瓶颈问题是没有一个有效地向工业WSN注入故障的机制或方法，也就无法真正地实际验证工业WSN的容错性，只能通过仿真的方式对工业WSN的容错性进行评价和验证，所以迫切的需要一个有效地故障注入方法用于评估和提高工业WSN的可靠性和容错性。However, the existing reliability testing and evaluation methods for industrial WSN generally have the following defects: (1) Although after years of research, scientists have proposed many reliability evaluation methods for traditional ad-hoc networks, but these However, it is difficult to analyze the unstable factors of the network according to the requirements of industrial applications, and to evaluate and quantify the degree to which it meets industrial applications according to the evaluation method for each performance index of the network. Therefore, these methods are difficult to measure and compare the reliability of industrial wireless sensor networks, and thus it is difficult to provide auxiliary evaluation for the design of industrial wireless sensor networks. (2) In industrial wireless sensor networks, the monitoring environment of the network is often very harsh, and developers cannot collect data on site to test and evaluate the reliability of the network, which urgently needs a means to artificially simulate these harsh environments Injection of certain faults into the wireless sensor network, and real-time recording and storage of current network operation information and response information to injected faults, and analysis, so as to evaluate and improve the reliability of the entire industrial WSN. However, through the research of a large number of literatures, it is found that this field has not received enough attention at present, and the only fault injection (Fault Injection, FI) method for industrial WSN is either realized by simulation or the injected fault cannot be effectively simulated. The situation encountered in the actual application of industrial WSN. (3) In the process of researching the fault tolerance of industrial WSN, the bottleneck problem commonly encountered at present is that there is no mechanism or method for effectively injecting faults into industrial WSN, and it is impossible to actually verify the fault tolerance of industrial WSN. The fault tolerance of industrial WSN is evaluated and verified through simulation, so an effective fault injection method is urgently needed to evaluate and improve the reliability and fault tolerance of industrial WSN.
发明内容Contents of the invention
为了解决现有技术的不足，本发明实施例的第一方面提供了一种测评无线传感器网络可靠性的方法，该方法能够有针对性地对无线传感器网络的可靠性进行测评，提高了无线传感器网络的可靠性和稳定性。In order to solve the deficiencies of the prior art, the first aspect of the embodiments of the present invention provides a method for evaluating the reliability of wireless sensor networks. Network reliability and stability.
本发明实施例的第一方面提供的一种测评无线传感器网络可靠性的方法，包括：The first aspect of the embodiments of the present invention provides a method for evaluating the reliability of a wireless sensor network, including:
采集当前无线传感器网络中节点网络状态信息；Collect node network status information in the current wireless sensor network;
从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重；Screen out the evaluation indicators of the wireless sensor network from the node network status information and determine the weight of each evaluation index;
根据无线传感器网络的测评指标及其权重，构建出无线传感器网络可靠性的测评模型；According to the evaluation index and weight of wireless sensor network, the evaluation model of wireless sensor network reliability is constructed;
向无线传感器网络人为注入模拟现场干扰的故障信息，并将实时获取的无线传感器网络中节点网络状态信息输入至所述测评模型内，由所述测评模型输出当前无线传感器网络的可靠性和容错性测评结果。Artificially inject fault information that simulates on-site interference into the wireless sensor network, and input the real-time acquired node network status information in the wireless sensor network into the evaluation model, and the evaluation model outputs the reliability and fault tolerance of the current wireless sensor network Assessment results.
结合本发明实施例的第一方面，本发明实施例的第一方面的第一种实施方式中，所述节点网络状态信息包括拓扑结构、数据包成功率、丢包率、电量及缓冲区使用情况。In combination with the first aspect of the embodiments of the present invention, in the first implementation manner of the first aspect of the embodiments of the present invention, the node network status information includes topology, data packet success rate, packet loss rate, power consumption, and buffer usage Condition.
结合本发明实施例的第一方面，本发明实施例的第一方面的第二种实施方式中，在从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重之前，还包括采用粗糙集理论对采集的当前无线传感器网络中节点网络状态信息进行去噪和去冗余处理。In combination with the first aspect of the embodiments of the present invention, in the second implementation manner of the first aspect of the embodiments of the present invention, before screening out the evaluation indicators of the wireless sensor network from the node network status information and determining the weight of each evaluation index, It also includes denoising and deredundancy processing of the collected node network status information in the current wireless sensor network by using the rough set theory.
从一系列已有数据中，寻找其规律或规则，预测问题的方向是粗糙集(RoughSets,RS)的基本思想。就本发明而言，在获得的通信节点海量运行信息中，存在着大量不精确、不完整和不确定的粗糙信息。传统数据处理方法无法精确确定数据中哪些信息是冗余的，哪些信息是有用的及其作用大小。本发明采用粗糙集理论作为前端处理系统，用以对海量节点网络状态信息进行预处理，从而得到简化、精确的数据，而且采用模糊理论对模糊的可靠性进行量化，为可靠性评估奠定基础。From a series of existing data, it is the basic idea of rough sets (RoughSets, RS) to find its laws or rules and predict the direction of the problem. As far as the present invention is concerned, there are a large amount of imprecise, incomplete and uncertain rough information in the mass operating information of the obtained communication nodes. Traditional data processing methods cannot accurately determine which information in the data is redundant, which information is useful, and its role. The invention adopts the rough set theory as the front-end processing system to preprocess massive node network status information to obtain simplified and accurate data, and uses fuzzy theory to quantify fuzzy reliability, laying the foundation for reliability evaluation.
结合本发明实施例的第一方面，本发明实施例的第一方面的第三种实施方式中，从节点网络状态信息中筛选出无线传感器网络的测评指标的过程中，采用聚类分析法将节点网络状态信息进行分类，进而得到各个测评指标。In combination with the first aspect of the embodiments of the present invention, in the third implementation manner of the first aspect of the embodiments of the present invention, in the process of screening out the evaluation indicators of the wireless sensor network from the node network status information, the cluster analysis method is used to The node network status information is classified, and then various evaluation indicators are obtained.
由于不同规范的可靠性指标之间存在程度不同的相似性，需要采用聚类分析等数学方法将不同规范的指标进行分类，并对可靠性的贡献进行量化，转化为可以进行综合的一个归一化的相对数，将不同性质和量纲的指标统一起来。此外，海量节点网络状态信息中异常数据的发现及评估模型中权值的确定，也需要聚类分析算法做支撑。Due to the different degrees of similarity between the reliability indicators of different specifications, it is necessary to use mathematical methods such as cluster analysis to classify the indicators of different specifications, quantify the contribution of reliability, and transform it into a normalized model that can be synthesized. It is a relative number that can be transformed to unify indicators of different properties and dimensions. In addition, the discovery of abnormal data in the network state information of massive nodes and the determination of weights in the evaluation model also require the support of cluster analysis algorithms.
结合本发明实施例的第一方面，本发明实施例的第一方面的第四种实施方式中，根据无线传感器网络的测评指标及其权重，并利用离散型Hopfield神经网络构建出无线传感器网络可靠性的测评模型。In combination with the first aspect of the embodiment of the present invention, in the fourth implementation manner of the first aspect of the embodiment of the present invention, according to the evaluation index and weight of the wireless sensor network, and using the discrete Hopfield neural network to construct a reliable wireless sensor network Sexual evaluation model.
其中，Hopfield神经网络即利用其联想记忆的能力逐渐趋近于某个储存的平衡点，当状态不再改变时，此时平衡点所对应的便是待求的性能分类等级。Among them, the Hopfield neural network uses its associative memory ability to gradually approach a certain stored balance point. When the state does not change, the balance point corresponds to the performance classification level to be sought.
本发明实施例的第二方面提供了一种测评无线传感器网络可靠性的系统。The second aspect of the embodiments of the present invention provides a system for evaluating the reliability of a wireless sensor network.
本发明实施例的第二方面提供的一种测评无线传感器网络可靠性的系统，包括：A second aspect of an embodiment of the present invention provides a system for evaluating the reliability of a wireless sensor network, including:
节点网络状态信息采集模块，其用于采集当前无线传感器网络中节点网络状态信息；A node network state information collection module, which is used to collect node network state information in the current wireless sensor network;
测评指标筛选模块，其用于从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重；An evaluation indicator screening module, which is used to filter out the evaluation indicators of the wireless sensor network from the node network status information and determine the weight of each evaluation indicator;
测评模型构建模块，其用于根据无线传感器网络的测评指标及其权重，构建出无线传感器网络可靠性的测评模型；An evaluation model building module, which is used to construct an evaluation model for the reliability of the wireless sensor network according to the evaluation indicators and weights of the wireless sensor network;
测评结果输出模块，其用于向无线传感器网络人为注入模拟现场干扰的故障信息，并将实时获取的无线传感器网络中节点网络状态信息输入至所述测评模型内，由所述测评模型输出当前无线传感器网络的可靠性和容错性测评结果。The evaluation result output module is used to artificially inject fault information simulating on-site interference into the wireless sensor network, and input the real-time obtained node network status information in the wireless sensor network into the evaluation model, and the evaluation model outputs the current wireless Reliability and Fault Tolerance Evaluation Results of Sensor Networks.
结合本发明实施例的第二方面，本发明实施例的第二方面的第一种实施方式中，所述节点网络状态信息包括拓扑结构、数据包成功率、丢包率、电量及缓冲区使用情况。In combination with the second aspect of the embodiments of the present invention, in the first implementation manner of the second aspect of the embodiments of the present invention, the node network status information includes topology, data packet success rate, packet loss rate, power consumption, and buffer usage Condition.
结合本发明实施例的第二方面，本发明实施例的第二方面的第二种实施方式中，该系统还包括去噪和去冗余模块，其用于采用粗糙集理论对采集的当前无线传感器网络中节点网络状态信息进行去噪和去冗余处理。With reference to the second aspect of the embodiments of the present invention, in the second implementation manner of the second aspect of the embodiments of the present invention, the system further includes a denoising and de-redundancy module, which is used to use the rough set theory to analyze the collected current wireless The node network status information in the sensor network is denoised and deredundantly processed.
从一系列已有数据中，寻找其规律或规则，预测问题的方向是粗糙集(RoughSets,RS)的基本思想。就本发明而言，在获得的通信节点海量运行信息中，存在着大量不精确、不完整和不确定的粗糙信息。传统数据处理方法无法精确确定数据中哪些信息是冗余的，哪些信息是有用的及其作用大小。本发明采用粗糙集理论作为前端处理系统，用以对海量节点网络状态信息进行预处理，从而得到简化、精确的数据，而且采用模糊理论对模糊的可靠性进行量化，为可靠性评估奠定基础。From a series of existing data, it is the basic idea of rough sets (RoughSets, RS) to find its laws or rules and predict the direction of the problem. As far as the present invention is concerned, there are a large amount of imprecise, incomplete and uncertain rough information in the mass operating information of the obtained communication nodes. Traditional data processing methods cannot accurately determine which information in the data is redundant, which information is useful, and its role. The invention adopts the rough set theory as the front-end processing system to preprocess massive node network status information to obtain simplified and accurate data, and uses fuzzy theory to quantify fuzzy reliability, laying the foundation for reliability evaluation.
结合本发明实施例的第二方面，本发明实施例的第二方面的第三种实施方式中，在所述测评指标筛选模块中，采用聚类分析法将节点网络状态信息进行分类，进而得到各个测评指标。In combination with the second aspect of the embodiments of the present invention, in the third implementation manner of the second aspect of the embodiments of the present invention, in the evaluation index screening module, the cluster analysis method is used to classify the node network status information, and then obtain various evaluation indicators.
由于不同规范的可靠性指标之间存在程度不同的相似性，需要采用聚类分析等数学方法将不同规范的指标进行分类，并对可靠性的贡献进行量化，转化为可以进行综合的一个归一化的相对数，将不同性质和量纲的指标统一起来。此外，海量节点网络状态信息中异常数据的发现及评估模型中权值的确定，也需要聚类分析算法做支撑。Due to the different degrees of similarity between the reliability indicators of different specifications, it is necessary to use mathematical methods such as cluster analysis to classify the indicators of different specifications, quantify the contribution of reliability, and transform it into a normalized model that can be synthesized. It is a relative number that can be transformed to unify indicators of different properties and dimensions. In addition, the discovery of abnormal data in the network state information of massive nodes and the determination of weights in the evaluation model also require the support of cluster analysis algorithms.
结合本发明实施例的第二方面，本发明实施例的第二方面的第四种实施方式中，在所述测评模型构建模块中，根据无线传感器网络的测评指标及其权重，并利用离散型Hopfield神经网络构建出无线传感器网络可靠性的测评模型。In combination with the second aspect of the embodiments of the present invention, in the fourth implementation manner of the second aspect of the embodiments of the present invention, in the evaluation model construction module, according to the evaluation indicators and their weights of the wireless sensor network, and using discrete Hopfield neural network constructs the evaluation model of wireless sensor network reliability.
其中，Hopfield神经网络即利用其联想记忆的能力逐渐趋近于某个储存的平衡点，当状态不再改变时，此时平衡点所对应的便是待求的性能分类等级。Among them, the Hopfield neural network uses its associative memory ability to gradually approach a certain stored balance point. When the state does not change, the balance point corresponds to the performance classification level to be sought.
与现有技术相比，本发明的有益效果是：Compared with prior art, the beneficial effect of the present invention is:
(1)本发明在实验环境中向无线传感器网络人为注入模拟现场干扰的故障，并通过观察注入故障后网络的反应以评价网络的可靠性和容错性的系统结构和实现方法，通过分析故障注入后的网络性能，可以有针对性地对网络机制做出改进来提高网络的可靠性和稳定性。(1) The present invention artificially injects the fault of simulating on-site interference to the wireless sensor network in the experimental environment, and by observing the reaction of the network after injecting the fault to evaluate the system structure and implementation method of the reliability and fault tolerance of the network, by analyzing the fault injection After the final network performance, the network mechanism can be improved in a targeted manner to improve the reliability and stability of the network.
(2)通过详细分析无线传感器网络的协议栈和拓扑结构，收集并整理无线传感器网络物理层、MAC层和网络层的性能参数，分析网络物理设备、链路连接、网络路由等性能的影响因素，建立无线传感器网络的性能参数体系和可靠性模型具有一定的创新性。(2) Through detailed analysis of the protocol stack and topology of the wireless sensor network, collect and organize the performance parameters of the physical layer, MAC layer and network layer of the wireless sensor network, and analyze the factors affecting the performance of network physical equipment, link connections, network routing, etc. , it is innovative to establish the performance parameter system and reliability model of wireless sensor network.
(3)本发明采用粗糙集理论作为前端处理系统，结合无线传感器网络节点性能参数属性的具体情况，利用组合数学知识改进归纳属性约简算法，提高约简的效率，然后与改进的离线Hopfield神经网络相结合进行可靠性评估，提高了测评结果的准确性。(3) the present invention adopts rough set theory as the front-end processing system, in conjunction with the concrete situation of wireless sensor network node performance parameter attribute, utilizes combinatorial mathematics knowledge to improve the induction attribute reduction algorithm, improves the efficiency of reduction, then with the improved off-line Hopfield neural network The combination of the network and the reliability evaluation improves the accuracy of the evaluation results.
附图说明Description of drawings
构成本申请的一部分的说明书附图用来提供对本申请的进一步理解，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。The accompanying drawings constituting a part of the present application are used to provide further understanding of the present application, and the schematic embodiments and descriptions of the present application are used to explain the present application, and do not constitute improper limitations to the present application.
图1是本发明实施例的一种测评无线传感器网络可靠性的方法流程图；FIG. 1 is a flowchart of a method for evaluating the reliability of a wireless sensor network according to an embodiment of the present invention;
图2是粗糙集理论的思想；Figure 2 is the idea of rough set theory;
图3是测评模型建立的步骤；Figure 3 is the steps of establishing the evaluation model;
图4是故障注入方法示意图；4 is a schematic diagram of a fault injection method;
图5是无线传感器网络可靠性层次图；Fig. 5 is a hierarchical diagram of wireless sensor network reliability;
图6是无线传感器网络可靠性的多层次系统；Figure 6 is a multi-level system of wireless sensor network reliability;
图7是本发明实施例的测评无线传感器网络可靠性的系统结构示意图。Fig. 7 is a schematic structural diagram of a system for evaluating reliability of a wireless sensor network according to an embodiment of the present invention.
具体实施方式detailed description
应该指出，以下详细说明都是例示性的，旨在对本申请提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本申请所属技术领域的普通技术人员通常理解的相同含义。It should be pointed out that the following detailed description is exemplary and intended to provide further explanation to the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本申请的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and/or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and/or combinations thereof.
图1是本发明实施例的一种测评无线传感器网络可靠性的方法流程图。Fig. 1 is a flowchart of a method for evaluating the reliability of a wireless sensor network according to an embodiment of the present invention.
如图1所示，本发明实施例的一种测评无线传感器网络可靠性，包括步骤1～步骤4。As shown in FIG. 1 , an evaluation of the reliability of a wireless sensor network in an embodiment of the present invention includes steps 1 to 4.
其中，步骤1：采集当前无线传感器网络中节点网络状态信息。Among them, step 1: collect the network status information of nodes in the current wireless sensor network.
在步骤中，所述节点网络状态信息包括拓扑结构、数据包成功率、丢包率、电量及缓冲区使用情况。In the step, the node network status information includes topology, data packet success rate, packet loss rate, power consumption and buffer usage.
步骤2：从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重。Step 2: Filter out the evaluation indicators of the wireless sensor network from the node network status information and determine the weight of each evaluation indicator.
在从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重之前，还包括采用粗糙集理论对采集的当前无线传感器网络中节点网络状态信息进行去噪和去冗余处理。Before screening out the evaluation indicators of the wireless sensor network from the node network state information and determining the weight of each evaluation index, it also includes using the rough set theory to denoise and de-redundantly process the collected node network state information in the current wireless sensor network .
基于粗糙集理论数据处理的基本思想是：①将实际收集的原始运行数据构成信息系统，如果收集的数据为连续量，则需要先通过适当的量化方法进行离散化，最终形成量化决策表；②通过知识的约简方法，得到由性能特征和模式构成的简化知识集合，利用分类质量和分类精度衡量简化知识的性能；③选择其中性能较好的简化知识形成最终的网络性能评估知识。The basic idea of data processing based on rough set theory is: ① The actual collected original operating data constitutes an information system. If the collected data is a continuous quantity, it needs to be discretized by an appropriate quantitative method first, and finally a quantitative decision table is formed; ② Through the knowledge reduction method, a simplified knowledge set composed of performance characteristics and patterns is obtained, and the performance of the simplified knowledge is measured by classification quality and classification accuracy; ③ Select the simplified knowledge with better performance to form the final network performance evaluation knowledge.
一般而言，首先要获得评估样本数据库，将需要评估对象的运行信息用粗糙集理论来表达，获得包含属性和值的二维关系型数据表，最终可以获得原始评估事例决策表。图2给出了粗糙集理论方法表达不确定性知识的基本思想。Generally speaking, the first step is to obtain the evaluation sample database, express the operation information of the evaluation object using rough set theory, obtain a two-dimensional relational data table containing attributes and values, and finally obtain the original evaluation case decision table. Figure 2 shows the basic idea of rough set theory to express uncertainty knowledge.
从一系列已有数据中，寻找其规律或规则，预测问题的方向是粗糙集(RoughSets,RS)的基本思想。就本发明而言，在获得的通信节点海量运行信息中，存在着大量不精确、不完整和不确定的粗糙信息。传统数据处理方法无法精确确定数据中哪些信息是冗余的，哪些信息是有用的及其作用大小。本发明采用粗糙集理论作为前端处理系统，用以对海量节点网络状态信息进行预处理，从而得到简化、精确的数据，而且采用模糊理论对模糊的可靠性进行量化，为可靠性评估奠定基础。From a series of existing data, it is the basic idea of rough sets (RoughSets, RS) to find its laws or rules and predict the direction of the problem. As far as the present invention is concerned, there are a large amount of imprecise, incomplete and uncertain rough information in the mass operating information of the obtained communication nodes. Traditional data processing methods cannot accurately determine which information in the data is redundant, which information is useful, and its role. The invention adopts the rough set theory as the front-end processing system to preprocess massive node network status information to obtain simplified and accurate data, and uses fuzzy theory to quantify fuzzy reliability, laying the foundation for reliability evaluation.
从节点网络状态信息中筛选出无线传感器网络的测评指标的过程中，采用聚类分析法将节点网络状态信息进行分类，进而得到各个测评指标。In the process of screening out the evaluation index of wireless sensor network from the node network state information, the node network state information is classified by cluster analysis method, and then each evaluation index is obtained.
由于不同规范的可靠性指标之间存在程度不同的相似性，需要采用聚类分析等数学方法将不同规范的指标进行分类，并对可靠性的贡献进行量化，转化为可以进行综合的一个归一化的相对数，将不同性质和量纲的指标统一起来。此外，海量节点网络状态信息中异常数据的发现及评估模型中权值的确定，也需要聚类分析算法做支撑。Due to the different degrees of similarity between the reliability indicators of different specifications, it is necessary to use mathematical methods such as cluster analysis to classify the indicators of different specifications, quantify the contribution of reliability, and transform it into a normalized model that can be synthesized. It is a relative number that can be transformed to unify indicators of different properties and dimensions. In addition, the discovery of abnormal data in the network state information of massive nodes and the determination of weights in the evaluation model also require the support of cluster analysis algorithms.
步骤3：根据无线传感器网络的测评指标及其权重，构建出无线传感器网络可靠性的测评模型。Step 3: According to the evaluation indicators and weights of the wireless sensor network, construct the evaluation model of the reliability of the wireless sensor network.
根据无线传感器网络的测评指标及其权重，并利用离散型Hopfield神经网络构建出无线传感器网络可靠性的测评模型。According to the evaluation index and weight of wireless sensor network, and using discrete Hopfield neural network, the evaluation model of wireless sensor network reliability is constructed.
其中，Hopfield神经网络即利用其联想记忆的能力逐渐趋近于某个储存的平衡点，当状态不再改变时，此时平衡点所对应的便是待求的性能分类等级。Among them, the Hopfield neural network uses its associative memory ability to gradually approach a certain stored balance point. When the state does not change, the balance point corresponds to the performance classification level to be sought.
离线Hopfield神经网络的设计思路是：The design idea of the offline Hopfield neural network is:
在无线传感器网络各层性能指标参数中挑选若干个能够反映网络可靠性的指标参数；将可靠性等级分为五个等级：很强(1级)、较强(2级)、一般(3级)、较差(4级)及很差(5级)，比较每个分类评估等级对应多个指标模型，将若干个典型的分类评估等级所对应的评估指标设计为离散型Hopfield神经网络的平衡点，Hopfield神经网络学习过程即为典型的分类等级评估指标逐渐趋近于Hopfield神经网络的平衡点的过程。Select several index parameters that can reflect the reliability of the network among the performance index parameters of each layer of the wireless sensor network; the reliability level is divided into five levels: very strong (level 1), strong (level 2), general (level 3 ), poor (level 4) and very poor (level 5), compare multiple index models corresponding to each classification evaluation level, and design the evaluation indicators corresponding to several typical classification evaluation levels as the balance of discrete Hopfield neural network point, the Hopfield neural network learning process is a process in which the typical classification level evaluation index gradually approaches the equilibrium point of the Hopfield neural network.
学习完成后，Hopfield神经网络储存的平衡点即为各个分类等级所对应的评估指标。当有待分类的WSN评估指标输入时，Hopfield神经网络即利用其联想记忆的能力逐渐趋近于某个储存的平衡点，当状态不再改变时，此时平衡点所对应的便是待求的分类等级。测评模型建立的步骤如图3所示。After the learning is completed, the balance point stored by the Hopfield neural network is the evaluation index corresponding to each classification level. When the WSN evaluation index to be classified is input, the Hopfield neural network uses its associative memory ability to gradually approach a certain stored equilibrium point. When the state does not change, the equilibrium point corresponds to the desired classification level. The steps of establishing the evaluation model are shown in Figure 3.
离线Hopfield神经网络的学习规则：利用外积法得到离线Hopfield神经网络的权系矩阵。对于一给定的需记忆的样本向量{t1,t2,...tN}，如果tk的状态为+1或-1，则其连接权值的学习可以利用“外积规则”，即利用外积法设计离线型Hopfield。Learning rules of the offline Hopfield neural network: use the outer product method to obtain the weight matrix of the offline Hopfield neural network. For a given sample vector {t1 ,t2 ,...tN } that needs to be memorized, if the state of tk is +1 or -1, the learning of its connection weights can use the "outer product rule" , that is, use the outer product method to design the off-line Hopfield.
具体地，基于层次分析法对工业无线传感器网络可靠性进行评估的模型，通过建立多层次的评价指标体系，对单项指标采用递推算法给出目标层对评语的隶属度。引入打分法，给出工业无线传感器网络可靠性的代数值。由数值的大小给出优劣评定，是对模糊性、相对性指标评价的一种可靠而有效的方法。Specifically, based on the AHP model for evaluating the reliability of industrial wireless sensor networks, a multi-level evaluation index system is established, and a recursive algorithm is used for a single index to give the membership degree of the target layer to the comment. The scoring method is introduced to give the algebraic value of the reliability of industrial wireless sensor networks. It is a reliable and effective method to evaluate fuzzy and relative indicators by giving the evaluation of the advantages and disadvantages based on the magnitude of the value.
建立评判对象指标集：U＝{U1,U2,L,Um}，其中Ui是U中的一个性能对可靠性的贡献度。Uij＝{Uij1,Uij2,L,Uijm}，是Ui中第j个性能的指标集，Uijk是Ui中第j个指标的指标集中的一个指标。多层结构如图5所示。Establish an evaluation object index set: U={U1 , U2 , L, Um }, where Ui is the contribution of a performance in U to reliability. Uij = {Uij1 , U ij2, L, Uijm }, which is the index set of the jth performance in Ui , and Uijk is an index in the index set of the jth index in Ui . The multilayer structure is shown in Figure 5.
建立A＝{A1,A2,L,Am}为U中各因素的权重，是U上模糊子集，且满足Ai≥0，i＝1,2,L,m。其根据不同的情况，运用层次分析法，通过两两比较建立判断矩阵，确定网络性能的指标权值。Establish A={A1 ,A2 ,L,Am } as the weight of each factor in U, which is a fuzzy subset on U, and satisfies Ai ≥ 0, i=1, 2, L, m. According to different situations, it uses analytic hierarchy process to establish a judgment matrix through pairwise comparison to determine the index weight of network performance.
设评判集：V＝{V1,V2,L,Vn}，Vj(j＝1,2,L,n)表示由高到低的评价等级，分为“1级、2级、3级、4级、5级”五个等级。Set the evaluation set: V={V1 ,V2 ,L,Vn }, Vj (j=1,2,L,n) represents the evaluation level from high to low, divided into "level 1, level 2, Level 3, Level 4, Level 5" five levels.
R为单因素的评价矩阵，R＝(Rij)m*n，Rij表示因素Ui被评为Vj的隶属度，并注意保持其归一化。设n为有效咨询次数，yij为因素Ui被评为Vj的次数，则有Rij＝yij/n。R is a single-factor evaluation matrix, R=(Rij )m*n , Rij represents the degree of membership of factor Ui rated as Vj , and attention should be paid to maintaining its normalization. Let n be the number of effective consultations, and yij be the number of times that factor Ui is rated as Vj , then Rij =yij /n.
利用模糊矩阵的合成运算，得综合评价模型为：Using the synthetic operation of fuzzy matrix, the comprehensive evaluation model is obtained as follows:
B＝AoR＝(B1,B2,L,Bn)B=AoR=(B1 ,B2 ,L,Bn )
上式中，j-1,2,L,n。∧表示Ai与Rij比较取最小值，∨表示要(Ai∧Rij)的几个最小值中取最大值。则做归一化处理：j＝1,2,L,n。In the above formula, j-1,2,L,n. ∧ means to compare Ai with Rij to take the minimum value, and ∨ means to take the maximum value among several minimum values of (Ai ∧ Rij ). Then do normalization: j=1,2,L,n.
设C＝(C1,C2,L,Cn)T是一分数集。其中Cj(j＝1,2,l,n)表示第j级评语的分数。这里采用百分制等差打分法：C＝{90,70,50,30,10}T。Let C=(C1 ,C2 ,L,Cn )T be a fraction set. Wherein Cj (j=1,2,l,n) represents the score of the jth level comment. Here, the percentile scoring method is adopted: C={90,70,50,30,10}T .
计算评价结果D。D是一得分值D＝B*C或D＝B'*C。Calculate the evaluation result D. D is a score value D=B*C or D=B'*C.
对于工业无线传感器网络可靠性的多层次系统，则由底层向上递推计算上一层次指标的评价结果，根据算出的各得分值，再向上递推直到目标层，如图6所示。For the multi-level system of industrial wireless sensor network reliability, the evaluation results of the upper-level indicators are recursively calculated from the bottom layer, and then recursively upward to the target layer according to the calculated scores, as shown in Figure 6.
对于不同的比较对象，设有S＝1,2,L,n个对象，将算得的D值记作Ds，则网络可靠性一目了然。For different comparison objects, assuming S=1, 2, L, n objects, the calculated D value is recorded as Ds , then the reliability of the network is clear at a glance.
步骤4：向无线传感器网络人为注入模拟现场干扰的故障信息，并将实时获取的无线传感器网络中节点网络状态信息输入至所述测评模型内，由所述测评模型输出当前无线传感器网络的可靠性和容错性测评结果。Step 4: Artificially inject fault information simulating on-site interference into the wireless sensor network, and input the real-time acquired node network status information in the wireless sensor network into the evaluation model, and the evaluation model outputs the reliability of the current wireless sensor network and fault-tolerance evaluation results.
具体方法为：采用的故障注入方法如图4所示。首先进行注入故障前的工作：The specific method is as follows: the adopted fault injection method is shown in Fig. 4 . First do the work before injecting the fault:
①确定要注入WSN节点的故障类型，包括故障命令及故障功能；① Determine the fault type to be injected into the WSN node, including fault commands and fault functions;
②完成对应故障类型的故障执行机制；② Complete the failure execution mechanism corresponding to the failure type;
③确定故障的激活时间即向WSN注入各种故障的时间。③ Determining the activation time of faults is the time to inject various faults into WSN.
这三部分组成一个故障集，然后通过故障注入节点将故障集一对一的注入到WSN节点中促使WSN节点发生故障。最后通过节点信息获取技术将各WSN节点的反应信息一起收集到上位计算机中进行分析。These three parts form a fault set, and then the fault set is injected into the WSN node one-to-one through the fault injection node to cause the WSN node to fail. Finally, through the node information acquisition technology, the response information of each WSN node is collected into the host computer for analysis.
随着无线传感器网络的深入研究，尽管研究人员提出了多个传感器节点上的协议栈但这些协议栈大体上包括物理层、数据链路层、网络层、传输层和应用层与互联网协议栈的五层协议相对应。本发明所采用的故障注入方式是在无线传感器网络协议栈中的应用层和物理层之间引进一个新的层—故障层(Fault Layer)，故障层对应的协议为故障协议(Fault Protocol)。故障协议实际上就是各种故障的执行机制，本发明提供了这样一个接口，它可以向网络注入故障序列，任何包含故障层的无线传感器网络接收到故障序列后就会执行故障协议中对应的故障函数从而达到注入故障的效果。这样即使不知道网络的内部运行机制，任何无线传感器网络协议栈中只要包含故障层，用本课题提供的接口都可以实现故障注入功能。在WSN的开发验证阶段，如果把故障的执行机制(故障协议)作为无线传感器网络协议栈的一部分连同其它协议一起开发，那么使用这种方法对无线传感器网络可靠性和容错性进行测试就成为一种普遍化的方法得以广泛使用。With the in-depth study of wireless sensor networks, although researchers have proposed a number of protocol stacks on sensor nodes, these protocol stacks generally include the physical layer, data link layer, network layer, transport layer and application layer and the Internet protocol stack. Corresponding to the five-layer protocol. The fault injection method adopted by the present invention is to introduce a new layer—Fault Layer—between the application layer and the physical layer in the wireless sensor network protocol stack, and the protocol corresponding to the fault layer is the Fault Protocol. The fault protocol is actually the execution mechanism of various faults. The present invention provides such an interface, which can inject fault sequences into the network, and any wireless sensor network containing the fault layer will execute the corresponding fault in the fault protocol after receiving the fault sequence. function to achieve the effect of injecting faults. In this way, even if the internal operation mechanism of the network is unknown, as long as the fault layer is included in any wireless sensor network protocol stack, the fault injection function can be realized by using the interface provided by this subject. In the development and verification stage of WSN, if the fault execution mechanism (fault protocol) is developed together with other protocols as part of the wireless sensor network protocol stack, then it becomes a must to use this method to test the reliability and fault tolerance of wireless sensor networks. A generalized approach is widely used.
故障注入节点向无线传感器网络节点注入的故障序列如表1所示，注入的故障可以有效地模拟WSN实际应用时遇到的故障和干扰。The fault sequence injected by the fault injection node into the wireless sensor network node is shown in Table 1. The injected fault can effectively simulate the fault and interference encountered in the actual application of WSN.
表1故障注入序列Table 1 Fault injection sequence
图7是本发明实施例的测评无线传感器网络可靠性的系统结构示意图。Fig. 7 is a schematic structural diagram of a system for evaluating reliability of a wireless sensor network according to an embodiment of the present invention.
如图7所示，本发明实施例的第二方面提供的一种测评无线传感器网络可靠性的系统，包括：As shown in Figure 7, the second aspect of the embodiment of the present invention provides a system for evaluating the reliability of a wireless sensor network, including:
节点网络状态信息采集模块，其用于采集当前无线传感器网络中节点网络状态信息；A node network state information collection module, which is used to collect node network state information in the current wireless sensor network;
测评指标筛选模块，其用于从节点网络状态信息中筛选出无线传感器网络的测评指标并确定各个测评指标的权重；An evaluation indicator screening module, which is used to filter out the evaluation indicators of the wireless sensor network from the node network status information and determine the weight of each evaluation indicator;
测评模型构建模块，其用于根据无线传感器网络的测评指标及其权重，构建出无线传感器网络可靠性的测评模型；An evaluation model building module, which is used to construct an evaluation model for the reliability of the wireless sensor network according to the evaluation indicators and weights of the wireless sensor network;
测评结果输出模块，其用于向无线传感器网络人为注入模拟现场干扰的故障信息，并将实时获取的无线传感器网络中节点网络状态信息输入至所述测评模型内，由所述测评模型输出当前无线传感器网络的可靠性和容错性测评结果。The evaluation result output module is used to artificially inject fault information simulating on-site interference into the wireless sensor network, and input the real-time obtained node network status information in the wireless sensor network into the evaluation model, and the evaluation model outputs the current wireless Reliability and Fault Tolerance Evaluation Results of Sensor Networks.
其中，所述节点网络状态信息包括拓扑结构、数据包成功率、丢包率、电量及缓冲区使用情况。Wherein, the node network state information includes topology structure, data packet success rate, packet loss rate, battery power and buffer usage.
结合本发明实施例的第二方面，本发明实施例的第二方面的第二种实施方式中，该系统还包括去噪和去冗余模块，其用于采用粗糙集理论对采集的当前无线传感器网络中节点网络状态信息进行去噪和去冗余处理。With reference to the second aspect of the embodiments of the present invention, in the second implementation manner of the second aspect of the embodiments of the present invention, the system further includes a denoising and de-redundancy module, which is used to use the rough set theory to analyze the collected current wireless The node network status information in the sensor network is denoised and deredundantly processed.
从一系列已有数据中，寻找其规律或规则，预测问题的方向是粗糙集(RoughSets,RS)的基本思想。就本发明而言，在获得的通信节点海量运行信息中，存在着大量不精确、不完整和不确定的粗糙信息。传统数据处理方法无法精确确定数据中哪些信息是冗余的，哪些信息是有用的及其作用大小。本发明采用粗糙集理论作为前端处理系统，用以对海量节点网络状态信息进行预处理，从而得到简化、精确的数据，而且采用模糊理论对模糊的可靠性进行量化，为可靠性评估奠定基础。From a series of existing data, it is the basic idea of rough sets (RoughSets, RS) to find its laws or rules and predict the direction of the problem. As far as the present invention is concerned, there are a large amount of imprecise, incomplete and uncertain rough information in the mass operating information of the obtained communication nodes. Traditional data processing methods cannot accurately determine which information in the data is redundant, which information is useful, and its role. The invention adopts the rough set theory as the front-end processing system to preprocess massive node network status information to obtain simplified and accurate data, and uses fuzzy theory to quantify fuzzy reliability, laying the foundation for reliability evaluation.
具体地，在所述测评指标筛选模块中，采用聚类分析法将节点网络状态信息进行分类，进而得到各个测评指标。Specifically, in the evaluation index screening module, a cluster analysis method is used to classify the node network status information, and then each evaluation index is obtained.
由于不同规范的可靠性指标之间存在程度不同的相似性，需要采用聚类分析等数学方法将不同规范的指标进行分类，并对可靠性的贡献进行量化，转化为可以进行综合的一个归一化的相对数，将不同性质和量纲的指标统一起来。此外，海量节点网络状态信息中异常数据的发现及评估模型中权值的确定，也需要聚类分析算法做支撑。Due to the different degrees of similarity between the reliability indicators of different specifications, it is necessary to use mathematical methods such as cluster analysis to classify the indicators of different specifications, quantify the contribution of reliability, and transform it into a normalized model that can be synthesized. It is a relative number that can be transformed to unify indicators of different properties and dimensions. In addition, the discovery of abnormal data in the network state information of massive nodes and the determination of weights in the evaluation model also require the support of cluster analysis algorithms.
具体地，在所述测评模型构建模块中，根据无线传感器网络的测评指标及其权重，并利用离散型Hopfield神经网络构建出无线传感器网络可靠性的测评模型。Specifically, in the evaluation model construction module, the evaluation model of the reliability of the wireless sensor network is constructed by using the discrete Hopfield neural network according to the evaluation index and weight of the wireless sensor network.
其中，Hopfield神经网络即利用其联想记忆的能力逐渐趋近于某个储存的平衡点，当状态不再改变时，此时平衡点所对应的便是待求的性能分类等级。Among them, the Hopfield neural network uses its associative memory ability to gradually approach a certain stored balance point. When the state does not change, the balance point corresponds to the performance classification level to be sought.
本发明在实验环境中向无线传感器网络人为注入模拟现场干扰的故障，并通过观察注入故障后网络的反应以评价网络的可靠性和容错性的系统结构和实现方法，通过分析故障注入后的网络性能，可以有针对性地对网络机制做出改进来提高网络的可靠性和稳定性。The present invention artificially injects faults simulating on-site interference into the wireless sensor network in the experimental environment, and evaluates the system structure and implementation method of network reliability and fault tolerance by observing the network response after the fault injection, and analyzes the network after the fault injection Performance, the network mechanism can be improved in a targeted manner to improve the reliability and stability of the network.
通过详细分析无线传感器网络的协议栈和拓扑结构，收集并整理无线传感器网络物理层、MAC层和网络层的性能参数，分析网络物理设备、链路连接、网络路由等性能的影响因素，建立无线传感器网络的性能参数体系和可靠性模型具有一定的创新性。By analyzing the protocol stack and topology of the wireless sensor network in detail, collecting and organizing the performance parameters of the physical layer, MAC layer and network layer of the wireless sensor network, and analyzing the factors affecting the performance of network physical equipment, link connections, network routing, etc., the establishment of wireless The performance parameter system and reliability model of the sensor network have certain innovations.
本发明采用粗糙集理论作为前端处理系统，结合无线传感器网络节点性能参数属性的具体情况，利用组合数学知识改进归纳属性约简算法，提高约简的效率，然后与改进的离线Hopfield神经网络相结合进行可靠性评估，提高了测评结果的准确性。The present invention adopts the rough set theory as the front-end processing system, combines the specific situation of the performance parameter attribute of the wireless sensor network node, uses the knowledge of combination mathematics to improve the induction attribute reduction algorithm, improves the efficiency of the reduction, and then combines it with the improved off-line Hopfield neural network Reliability evaluation is carried out to improve the accuracy of evaluation results.
上述虽然结合附图对本发明的具体实施方式进行了描述，但并非对本发明保护范围的限制，所属领域技术人员应该明白，在本发明的技术方案的基础上，本领域技术人员不需要付出创造性劳动即可做出的各种修改或变形仍在本发明的保护范围以内。Although the specific implementation of the present invention has been described above in conjunction with the accompanying drawings, it does not limit the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, those skilled in the art do not need to pay creative work Various modifications or variations that can be made are still within the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710670681.6ACN107295537A (en) | 2017-08-08 | 2017-08-08 | A kind of method and system for wireless sensor network reliability of testing and assessing |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710670681.6ACN107295537A (en) | 2017-08-08 | 2017-08-08 | A kind of method and system for wireless sensor network reliability of testing and assessing |
| Publication Number | Publication Date |
|---|---|
| CN107295537Atrue CN107295537A (en) | 2017-10-24 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710670681.6APendingCN107295537A (en) | 2017-08-08 | 2017-08-08 | A kind of method and system for wireless sensor network reliability of testing and assessing |
| Country | Link |
|---|---|
| CN (1) | CN107295537A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133083A (en)* | 2017-12-08 | 2018-06-08 | 中国航天标准化研究所 | A kind of space man-machine system safety effects Factor Selection method |
| CN108683547A (en)* | 2018-05-29 | 2018-10-19 | 中科边缘智慧信息科技（苏州）有限公司 | A kind of wireless sensor network configuration method of software definition |
| CN110430546A (en)* | 2019-07-12 | 2019-11-08 | 重庆电子工程职业学院 | A kind of anti-interference optimization system of wireless sensor network |
| CN111077886A (en)* | 2019-12-31 | 2020-04-28 | 上海申铁信息工程有限公司 | Station fault real-time monitoring system |
| CN111210361A (en)* | 2019-12-30 | 2020-05-29 | 国网江苏省电力有限公司信息通信分公司 | Routing planning method for power communication network based on reliability prediction and particle swarm optimization |
| CN113423328A (en)* | 2019-02-22 | 2021-09-21 | 吴宇 | Physical sign parameter detection system and reliability evaluation method of physical sign parameters |
| CN115604739A (en)* | 2019-03-08 | 2023-01-13 | 江苏海洋大学（Cn） | A System for Wireless Sensor Network Security |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090113232A1 (en)* | 2007-10-31 | 2009-04-30 | Electronics And Telecommunications Research Institute | Apparatus and method for managing wireless sensor network |
| CN101867960A (en)* | 2010-06-08 | 2010-10-20 | 江苏大学 | A Comprehensive Performance Evaluation Method for Wireless Sensor Networks |
| CN104820716A (en)* | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN106446016A (en)* | 2016-08-29 | 2017-02-22 | 中国电力科学研究院 | Parallel association rule mining-based method for predicting running reliability of power distribution network |
| CN106550387A (en)* | 2016-10-25 | 2017-03-29 | 华侨大学 | A kind of wireless sensor network routing layer QoS evaluating method |
| CN106650944A (en)* | 2016-12-07 | 2017-05-10 | 大连理工大学 | Metallurgical enterprise converter gas scheduling method based on knowledge |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090113232A1 (en)* | 2007-10-31 | 2009-04-30 | Electronics And Telecommunications Research Institute | Apparatus and method for managing wireless sensor network |
| CN101867960A (en)* | 2010-06-08 | 2010-10-20 | 江苏大学 | A Comprehensive Performance Evaluation Method for Wireless Sensor Networks |
| CN104820716A (en)* | 2015-05-21 | 2015-08-05 | 中国人民解放军海军工程大学 | Equipment reliability evaluation method based on data mining |
| CN106446016A (en)* | 2016-08-29 | 2017-02-22 | 中国电力科学研究院 | Parallel association rule mining-based method for predicting running reliability of power distribution network |
| CN106550387A (en)* | 2016-10-25 | 2017-03-29 | 华侨大学 | A kind of wireless sensor network routing layer QoS evaluating method |
| CN106650944A (en)* | 2016-12-07 | 2017-05-10 | 大连理工大学 | Metallurgical enterprise converter gas scheduling method based on knowledge |
| Title |
|---|
| 陈仕鸿: ""基于离散型Hopfield神经网络的台风灾情评估模型"", 《自然灾害学报》* |
| 黄旭: ""无线传感器网络性能测试与智能故障诊断技术研究"", 《中国博士学位论文全文数据库信息科技辑》* |
| 黄福员: ""一种金融风险预警指标体系的构建方法与应用"", 《科技与产业》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133083A (en)* | 2017-12-08 | 2018-06-08 | 中国航天标准化研究所 | A kind of space man-machine system safety effects Factor Selection method |
| CN108133083B (en)* | 2017-12-08 | 2021-09-03 | 中国航天标准化研究所 | Method for screening safety influence factors of space man-machine system |
| CN108683547A (en)* | 2018-05-29 | 2018-10-19 | 中科边缘智慧信息科技（苏州）有限公司 | A kind of wireless sensor network configuration method of software definition |
| CN113423328A (en)* | 2019-02-22 | 2021-09-21 | 吴宇 | Physical sign parameter detection system and reliability evaluation method of physical sign parameters |
| CN115604739A (en)* | 2019-03-08 | 2023-01-13 | 江苏海洋大学（Cn） | A System for Wireless Sensor Network Security |
| CN110430546A (en)* | 2019-07-12 | 2019-11-08 | 重庆电子工程职业学院 | A kind of anti-interference optimization system of wireless sensor network |
| CN111210361A (en)* | 2019-12-30 | 2020-05-29 | 国网江苏省电力有限公司信息通信分公司 | Routing planning method for power communication network based on reliability prediction and particle swarm optimization |
| CN111077886A (en)* | 2019-12-31 | 2020-04-28 | 上海申铁信息工程有限公司 | Station fault real-time monitoring system |
| Publication | Publication Date | Title |
|---|---|---|
| CN107295537A (en) | A kind of method and system for wireless sensor network reliability of testing and assessing | |
| CN114048600B (en) | Digital twin-driven multi-model fusion industrial system anomaly detection method | |
| CN106779069A (en) | A kind of abnormal electricity consumption detection method based on neutral net | |
| CN109407654B (en) | A nonlinear causal analysis method for industrial data based on sparse deep neural network | |
| CN104298881B (en) | A kind of public transport environment dynamic change forecasting procedure based on Bayesian network model | |
| CN110488150A (en) | A kind of intelligent fault diagnosis method based on more algorithm fusions | |
| CN102269972A (en) | Method and device for compensating pipeline pressure missing data based on genetic neural network | |
| CN101718834B (en) | Method for analysing analog circuit fault propagation characteristic | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| CN117590130A (en) | Digital twinning-based power transmission line situation awareness method | |
| CN104503420B (en) | Non-linear process industry fault prediction method based on novel FDE-ELM and EFSM | |
| CN105956798A (en) | Sparse random forest-based method for assessing running state of distribution network device | |
| CN111273125A (en) | Fault diagnosis method of power cable channel based on RST-CNN | |
| CN110874685A (en) | A method and system for judging the running state of smart electric energy meter based on neural network | |
| CN103279803A (en) | Load modeling method and system based on comprehensive information theory and modern interior point theory | |
| CN107644297A (en) | A kind of energy-saving of motor system amount calculates and verification method | |
| CN113722195B (en) | Local area network operation evaluation system and method based on AHP hierarchical analysis method | |
| WO2023082788A1 (en) | Method and apparatus for predicting oxygen content in flue gas and load, method and apparatus for selecting prediction model, and method and apparatus for predicting flue gas emission | |
| CN114282714A (en) | Electric energy meter state analysis and evaluation method based on generalized regression neural network | |
| CN117613877A (en) | A power grid planning method and system based on multi-objective optimization | |
| CN113033898A (en) | Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network | |
| CN115438897A (en) | Industrial process product quality prediction method based on BLSTM neural network | |
| Hu et al. | Research on the fault identification method of oil pumping unit based on residual network | |
| CN117665442A (en) | A testing method and system for electric intelligent IoT locks | |
| CN115425768B (en) | PyTorch-based LCC-S type WPT system load and self-mutual inductance identification method and system |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20171024 | |
| RJ01 | Rejection of invention patent application after publication |