WO2020093904A1 - 无线传感器网络容错拓扑演化方法 - Google Patents

无线传感器网络容错拓扑演化方法 Download PDF

Info

Publication number
WO2020093904A1
WO2020093904A1 PCT/CN2019/113846 CN2019113846W WO2020093904A1 WO 2020093904 A1 WO2020093904 A1 WO 2020093904A1 CN 2019113846 W CN2019113846 W CN 2019113846W WO 2020093904 A1 WO2020093904 A1 WO 2020093904A1
Authority
WO
WIPO (PCT)
Prior art keywords
fault
tolerant
nodes
network
rhcs
Prior art date
Application number
PCT/CN2019/113846
Other languages
English (en)
French (fr)
Inventor
李光辉
胡世红
Original Assignee
江南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 江南大学 filed Critical 江南大学
Publication of WO2020093904A1 publication Critical patent/WO2020093904A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/04—Arrangements for maintaining operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/12—Discovery or management of network topologies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks

Definitions

  • the invention relates to a fault-tolerant topology evolution method of a wireless sensor network, and belongs to the field of topology evolution of a wireless sensor network.
  • Topology is an organizational structure of Wireless Sensor Networks (WSN) nodes. Two nodes that can communicate directly have a topological edge. If there is no topology evolution model, all sensor nodes will form an unorganized network with maximum power transmission, and The sensor node is usually powered by its own battery. When it is always in the state of maximum power consumption, it will inevitably cause rapid exhaustion of node energy, high network routing load, and short life cycle.
  • WSN Wireless Sensor Networks
  • topology evolution technology In order to extend the life cycle of sensor nodes as long as possible, in the field of sensor networks, topology evolution technology has become a more intensively studied problem in recent years, and the unique characteristics and strict constraints of wireless sensor networks make the study of this problem more challenging. Aiming at the problem of topology evolution in wireless sensor networks, there are currently many methods, based on energy perception, random walk, fitness and complex network theory. In addition, according to the topology evolution of the sensor network architecture, it can be divided into centralized and distributed, but the existing fault-tolerant topology evolution methods of wireless sensor networks need to improve the tolerance of comprehensive faults.
  • the technical problem to be solved by the present invention is to provide a fault-tolerant topology evolution method for wireless sensor networks based on Markov and scale-free networks.
  • the Markov and scale-free networks Wireless sensor network fault-tolerant topology evolution model (Scale-Free Topology Evolution Mechanism, SFTEM) first proposed a regular hexagonal clustering mechanism (Regular hexagonal clustering structure, RHCS), which was analyzed by Markov model to meet at least 1-fault tolerance, SFTEM Combining the reliability of RHCS with scale-free characteristics, a robust wireless sensor network is formed, which utilizes the synergy between reliable clustering schemes and topological evolution, and can tolerate comprehensive failures such as random and energy failures .
  • Regular hexagonal clustering structure RHCS
  • SFTEM regular hexagonal clustering structure
  • the present invention provides a fault-tolerant topology evolution method for wireless sensor networks based on Markov and scale-free networks, including:
  • RHCS regular hexagon clustering mechanism
  • the comprehensive failure probability is introduced into the scale-free topology construction rules to form a fault-tolerant topology evolution model of wireless sensor networks based on Markov and scale-free networks;
  • RHCS regular hexagon clustering mechanism
  • RFP Random Failure Probability
  • the fault-tolerant topology evolution model specifically includes:
  • the scale-free network formed is analyzed, and its network node degree distribution is analyzed to meet the power law characteristics.
  • the nodes with higher node degree are randomly removed, the removal rate is between 0.05 and 0.25, and the ratio of the number of nodes in the largest connected branch to the total number is the fault tolerance performance index.
  • a computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, any of the steps of the method is implemented.
  • a processor for running a program wherein the method according to any one of the items is executed when the program is running.
  • a fault-tolerant topology evolution method for wireless sensor networks based on Markov and scale-free networks Unlike traditional scale-free network topology models, the fault-tolerant topology evolution model (SFTEM) of wireless sensor networks based on Markov and scale-free networks proposed in this application first proposed a regular hexagon clustering mechanism (RHCS). The model analyzes that the mechanism satisfies at least 1-fault tolerance. SFTEM combines the reliability of RHCS with scale-free characteristics to form a robust wireless sensor network, which utilizes the synergy between a reliable clustering scheme and topological evolution , Can tolerate comprehensive failures such as random failures and energy failures.
  • RHCS regular hexagon clustering mechanism
  • FIG. 1A is a schematic diagram of a clustering mechanism based on regular hexagons in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • 1B is a second schematic diagram of a clustering mechanism based on regular hexagons in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • FIG. 1 Markov model of RHCS in the state of fault-free nodes in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • FIG. 3 Markov model of RHCS under the fault state of strong fault-tolerant nodes in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • Fig. 4 Markov model of RHCS in the fault state of common fault-tolerant nodes in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • FIG. 5 is a schematic diagram of the basic structure of the RHCS in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • 6A is one of the comparison schematic diagrams of fault tolerance performance in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • 6B is a second schematic diagram for comparing fault tolerance performance in the fault-tolerant topology evolution method of the wireless sensor network of the present invention.
  • the BA scale-free network proposed by Albert R and Barabasi AL has a wide range of applications in topological evolution.
  • This method has two characteristics. One is growth. The so-called growth means that the network scale is constantly increasing. In the network, the number of nodes in the network is constantly increasing; the second is the priority connection mechanism. This feature means that the new nodes constantly generated in the network are more inclined to connect with those nodes with greater connectivity.
  • the topology generated by this method has a strong tolerance to random node failures, but has a weak ability to tolerate node failures caused by malicious attacks. Therefore, the present invention attempts to propose a fault-tolerant topology evolution model (SFTEM) for wireless sensor networks based on Markov and scale-free networks.
  • SFTEM fault-tolerant topology evolution model
  • the model first constructs a regular hexagon clustering mechanism (RHCS) with fault-tolerant sensor nodes as the vertices of the hexagon; secondly, the random failure probability and energy failure probability of RHCS are analyzed by Markov to obtain the comprehensive failure probability; finally The comprehensive failure probability is introduced into the scale-free topology construction rules to form a fault-tolerant topology evolution model for wireless sensor networks based on Markov and scale-free networks.
  • RHCS regular hexagon clustering mechanism
  • Fault tolerance the ratio of the number of nodes in the largest connected branch to the total number.
  • this application proposes a fault-tolerant topology evolution model (SFTEM) for wireless sensor networks based on Markov and scale-free networks.
  • the invention proposes a fault-tolerant topology evolution model (SFTEM) for wireless sensor networks based on Markov and scale-free networks.
  • the model first proposes a regular hexagonal clustering mechanism (RHCS).
  • RHCS regular hexagonal clustering mechanism
  • the Markov model analyzes that the mechanism meets at least 1-fault tolerance.
  • SFTEM combines the reliability of RHCS with scale-free characteristics to form a robust wireless sensor network .
  • Using the synergy between a reliable clustering scheme and topological evolution can tolerate comprehensive failures such as random failures and energy failures.
  • the method of the invention improves the network fault tolerance and extends the network life.
  • the fault-tolerant topology evolution model of wireless sensor networks based on Markov and scale-free networks proposed in this application is mainly designed for fault-tolerant topology in wireless sensor networks, including two steps: clustering mechanism and inter-cluster topology evolution.
  • Step 1 Build a regular hexagon clustering mechanism RHCS
  • Node redundancy is one of the most effective methods to improve the fault tolerance of sensor nodes. Therefore, we call the duplex sensor node a fault-tolerant sensor node. In the fault-tolerant sensor node model, it is assumed that the backup node is dormant. Only when the active node is diagnosed as a fault, the backup node becomes the active node.
  • the failure rate ⁇ t of the sensor node can be expressed as an exponential distribution with the failure rate ⁇ t at time t s .
  • the node degree k represents the total number of nodes connected to the node.
  • the fault diagnosis accuracy factor c The coverage of the wireless sensor network is usually defined as the goodness and duration of the sensor node's ability to observe the physical space of the monitoring area.
  • the fault diagnosis accuracy factor c represents the probability that the active sensor node has been correctly diagnosed and replaced by the backup sensor node.
  • the factor c depends on the node degree k and the cumulative probability of sensor failure ⁇ t .
  • M ( ⁇ t ) is a function of ⁇ t and represents an adjustment parameter that can loosely correspond to the desired average node degree required to obtain good fault detection accuracy for a given ⁇ t .
  • Definition 5 If the node clustering mechanism arbitrarily deletes k nodes and still maintains the coverage of the clustering mechanism, the mechanism is said to have k-coverage fault tolerance.
  • Theorem 1 If the ratio between the transmission range expressed in r cs and the sensing range is not less than 2, then coverage means connectivity. Regular triangle The ratio is optimal.
  • Step 1.1 Construct a regular hexagon clustering mechanism that satisfies 1-coverage fault tolerance:
  • the fault-tolerant nodes are placed in a regular hexagon
  • the black circle area represents the perception range of a common fault-tolerance node
  • the red circle area represents the perception range of a strong fault-tolerance node.
  • Seven common fault-tolerant nodes form a regular hexagonal structure, one of which is located in the center of the hexagon. There are also strong fault-tolerant nodes located in the center of the hexagon, in addition, the distance d between adjacent fault-tolerant nodes is equal to As shown in Figure 1B, we analyzed the k-coverage clustering mechanism.
  • the green area of the model can represent 1-coverage fault tolerance
  • the yellow area represents 3 coverage fault tolerance
  • the red area represents 5-cover fault tolerance. Therefore, the model satisfies at least 1-coverage fault tolerance.
  • RHCS regular hexagon clustering mechanism
  • Step 1.2 RHCS random failure probability (RFP) analysis
  • RHCS As long as the fault-tolerant node works normally, RHCS is effective.
  • the random failure probability of RHCS is analyzed from the following two cases.
  • Step 1.3 RHCS energy failure probability (EFP) analysis
  • E tx E elec ⁇ l + ⁇ amp ⁇ l ⁇ d 2
  • E elec the data fusion energy consumption
  • ⁇ amp the amplifier power consumption
  • D the transmission radius of the node.
  • RHCS because the basic structure of node deployment in the mechanism is a regular hexagon, if RHCS is used as a cluster evolution topology, then the cluster structure in the topology will be the same, and there is no difference in energy failure probability.
  • the size of the model will change according to the distance d between nodes, as shown in Figure 5,
  • n c is the number of normally working nodes in RHCS.
  • the energy RHCS failure probability P e is:
  • Step 1.4 RHCS's comprehensive failure probability (JFP)
  • step 2 The comprehensive failure probability derived from the final analysis in step 1 will be introduced into the model building in step 2 below.
  • Step 2 Establish fault-tolerant topology evolution model (SFTEM)
  • wireless sensor networks tend to adopt a cluster structure to extend the life cycle of the network in many cases.
  • the wireless sensor network has obvious dynamic characteristics, including the increase of new nodes and new links, and the failure of nodes caused by environmental factors or energy consumption.
  • the evolution process refers to adding new fault-tolerant cluster heads to the network.
  • Sensor nodes in the network are divided into strong fault-tolerant nodes and common fault-tolerant nodes.
  • the common fault-tolerant node joins the network as a cluster member of the fault-tolerant cluster and establishes a communication relationship with the fixed cluster head of the fault-tolerant cluster.
  • strong fault-tolerant nodes join the network as cluster heads, they will establish links with cluster heads of other fault-tolerant clusters and use multi-hop communication to transmit data.
  • Step 2.1 Clustering scale-free topology evolution model
  • the JFP represented as P
  • the node degree k of the cluster head
  • the distance D between the cluster heads are used as evaluation criteria.
  • the fitness function F be the reciprocal of the product of P and D between cluster heads. The probability of selecting an existing cluster head in the network to connect with the newly added cluster head depends on the values of F and k.
  • the fitness function and the node degree determine the probability that the new cluster head is connected to it.
  • the fitness function combines the comprehensive failure probability of the fault-tolerant cluster and the distance between the cluster heads. It not only considers the overall failure probability of the fault-tolerant cluster, random failure and energy exhaustion failure, but also controls the energy consumption of the cluster head to forward data. That is, the smaller the distance, the lower the energy consumption.
  • the setting of the upper limit of the node degree affects the distribution of the network node degree, and plays a certain role in improving the energy consumption balance of the network. The following will analyze the dynamic characteristics of the evolution model to prove that the degree distribution of the network satisfies the power law of the scale-free network.
  • Step 2.2 Dynamic characteristics analysis
  • the distribution of the cluster head node degree has obvious heterogeneity, that is, a small number of cluster heads have the majority of the network as the connection, and the connection degree of most nodes is the lowest node degree.
  • ⁇ k> represents the average node degree of the cluster head in the local world.
  • equation 14 can be simplified to:
  • the network cluster head node degree distribution p (k) conforms to the power law distribution characteristics, and the power law index Therefore, the network generated by the topology evolution model proposed in this paper has cluster head node degree distribution that meets the characteristics of scale-free networks, and thus has the fault-tolerant capability of scale-free networks.
  • Step 3 Evaluate fault tolerance performance
  • Steps 3.1-3.2 Poisson's rule is used to randomly generate failed nodes, and the nodes with exhausted energy are removed to calculate the maximum number of connected branch nodes; then the cluster heads with high node degree are randomly removed, and the removal ratio is 0.05 to 0.25, calculate the maximum number of connected branch nodes.
  • the initial energy of the nodes is the same.
  • the number of fault-tolerant clusters m 0 is 4, and the number of newly added cluster head connection edges m is set to 3.
  • the average value of the results of 50 experiments is taken as the experimental result to ensure the accuracy of the experiment.
  • the specific experimental parameters are shown in Table 1.
  • failure nodes are randomly generated by Poisson's rule, and nodes with exhausted energy are removed after each round of operation, and the relationship between the maximum connected branch and the network running time is observed, as shown in FIG. 6A.
  • the cluster heads with high node degree are randomly removed, and the removal ratio is increased from 0.05 to 0.25, as shown in FIG. 6B.
  • model of the present invention has a higher tolerance to comprehensive faults and a higher node degree cluster head attack tolerance capability than traditional topology-based evolution models based on scale-free networks.
  • the comparative literature is as follows:
  • Some steps in the embodiments of the present invention may be implemented by software, and the corresponding software program may be stored in a readable storage medium, such as an optical disk or a hard disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Algebra (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明公开了一种无线传感器网络容错拓扑演化方法，属于无线传感器网络(WSN)拓扑演化领域。所述方法首先提出了一个正六边形的分簇机制(RHCS)，通过Markov模型分析该机制至少满足1-容错，SFTEM将RHCS的可靠性与无标度特性相结合，形成了一个鲁棒的无线传感器网络，它利用了可靠的分簇方案和拓扑演化之间的协同作用，能够容忍随机故障和能量故障等综合故障，根据实验结果可知，在容侵实验中，随机移除高节点度的簇头，移除比例由0.05提高到0.25，对综合故障的容忍能力以及对高节点度簇头攻击的容侵能力明显高于传统基于无标度网络的拓扑演化模型，延长了无线传感器网络的寿命。

Description

无线传感器网络容错拓扑演化方法 技术领域
本发明涉及无线传感器网络容错拓扑演化方法，属于无线传感器网络拓扑演化领域。
背景技术
拓扑作为无线传感器网络(Wireless sensor networks，WSN)节点的组织结构，可以直接通信的两个节点存在一条拓扑边，如果没有拓扑演化模型，所有传感器节点都会以最大功率传输形成无组织的网络，而传感器节点通常以自带电池供电，始终处于最大功耗状态下势必造成节点能量快速耗尽，网络路由负载高，生命周期短等问题。
为了尽可能长的延长传感器节点的生命周期，在传感器网络领域，拓扑演化技术成为近年较为深入研究的问题，而无线传感器网络独特的特点及严格的约束条件使得该问题的研究更具有挑战性。针对无线传感器网络中拓扑演化问题，目前已有很多种方法，基于能量感知、基于随机行走、基于适应度以及基于复杂网络理论的方法。此外，按照传感器网络体系结构拓扑演化又可以分为集中式的和分布式的，但是现有的无线传感器网络容错拓扑演化方法对于综合故障的容忍能力有待提高。
发明内容
本发明要解决的技术问题是提供一种基于Markov和无标度网络的无线传感器网络容错拓扑演化方法，不同于传统的无标度网络拓扑模型，本申请提出的基于Markov和无标度网络的无线传感器网络容错拓扑演化模型(Scale-Free Topology Evolution Mechanism，SFTEM)首先提出了一个正六边形的分簇机制(Regular hexagon clustering structure，RHCS)，通过Markov模型分析该机制至少满足1-容错，SFTEM将RHCS的可靠性与无标度特性相结合，形成了一个鲁棒的无线传感器网络，它利用了可靠的分簇方案和拓扑演化之间的协同作用，能够容忍随机故障和能量故障等综合故障。
为了解决上述技术问题，本发明提供了一种基于Markov和无标度网络的无线传感器网络容错拓扑演化方法，包括：
构造一种以容错传感器节点为六边形顶点的正六边形分簇机制(RHCS)，通过Markov分析了RHCS的随机失效概率以及能量故障概率，得出综合故障概率；
将综合故障概率引入无标度拓扑构建规则中，形成一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型；
设置随机故障节点，计算网络最大连通子图节点个数，评估容错性能。
可选的，所述构造一种以容错传感器节点为六边形顶点的正六边形分簇机制(RHCS)，通过Markov分析了RHCS的随机失效概率以及能量故障概率，得出综合故障概率；”具体包括：
将双工传感器节点设为容错传感器节点，把容错传感器节点放置为正六边形结构形成基本的分簇机制RHCS；
通过Markov分析RHCS的随机失效率(Random Failure Probability，RFP)；
采用经典的一阶无线通信能量消耗模型，分析RHCS的能量故障率(Energy Failure Probability，EFP)；
结合RFP和EFP，建立RHCS的综合故障概率(Joint Failure Probability，JFP)。
可选的，“将综合故障概率引入无标度拓扑构建规则中，形成一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型；”中，容错拓扑演化模型具体包括：
网络中的大部分节点只和很少节点连接，而有极少的节点与非常多的节点连接，其经典的无标度网络模型构建算法：
增长：从一个具有m 0个节点的联通网络开始，每次引入一个新的节点,并且连到m个已经存在的节点上，这里m<＝m 0；
择优连接：一个新的节点与一个已经存在的节点i相连的概率w与节点i的度k_i之间的关系为w＝k_i/(k_1+k_2+k_3+...+k_n)，其中n为网络中的节点的总个数；
形成的无标度网络，分析其网络节点度分布，满足幂律特性。
可选的，“设置随机故障节点，计算网络最大连通子图节点个数，评估容错性能。”中，所述评估容错性能具体包括：
随机失效容错性，以泊松规则随机地产生失效节点，每一轮运行后移除能量耗尽的节点，以最大连通分支中的节点个数所占总个数的比例为容错性能指标；
恶意攻击容错性，将节点度较高的节点随机去除，去除率在0.05～0.25之间，以最大连通分支中的节点个数所占总个数的比例为容错性能指标。
一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现任一项所述方法的步骤。
一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现任一项所述方法的步骤。
一种处理器，所述处理器用于运行程序，其中，所述程序运行时执行任一项所述的方法。
本发明的有益效果：
一种基于Markov和无标度网络的无线传感器网络容错拓扑演化方法。不同于传统的无标度网络拓扑模型，本申请提出的基于Markov和无标度网络的无线传感器网络容错拓扑演化模型(SFTEM)首先提出了一个正六边形的分簇机制(RHCS)，通过Markov模型分析该机制至少满足1-容错，SFTEM将RHCS的可靠性与无标度特性相结合，形成了一个鲁棒的无线传感器网络，它利用了可靠的分簇方案和拓扑演化之间的协同作用，能够容忍随机故障和能量故障等综合故障。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1A本发明无线传感器网络容错拓扑演化方法中的基于正六边形的分簇机制示意图之一。
图1B本发明无线传感器网络容错拓扑演化方法中的基于正六边形的分簇机制示意图之二。
图2本发明无线传感器网络容错拓扑演化方法中的无故障节点状态下RHCS的Markov模型。
图3本发明无线传感器网络容错拓扑演化方法中的强容错节点故障状态下RHCS的Markov模型。
图4本发明无线传感器网络容错拓扑演化方法中的普通容错节点故障状态下RHCS的Markov模型。
图5本发明无线传感器网络容错拓扑演化方法中的RHCS的基本结构示意图。
图6A本发明无线传感器网络容错拓扑演化方法中的容错性能对比示意图之一。
图6B本发明无线传感器网络容错拓扑演化方法中的容错性能对比示意图之二。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
Albert R和Barabasi AL提出的BA无标度网络在拓扑演化中具有广泛应用，该方法具有两个特性，其一是增长性，所谓增长性是指网络规模是在不断的增大的，在研究的网络当中，网络的节点是不断的增加的；其二就是优先连接机制，这个特性是指网络当中不断产生的新 的节点更倾向于和那些连接度较大的节点相连接。运用该方法产生的拓扑对随机节点失效具有强容忍能力，但容忍恶意攻击导致的节点失效能力较弱。因此，本发明尝试提出了一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型(SFTEM)。该模型首先构造了一种以容错传感器节点为六边形顶点的正六边形分簇机制(RHCS)；其次，通过Markov分析了RHCS的随机失效概率以及能量故障概率，得出综合故障概率；最后，将综合故障概率引入无标度拓扑构建规则中，形成一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型。通过对生成拓扑的容错实验，结果表明，本发明提出的模型提高了网络的综合容错性能，具有广阔的应用前景。
名词解释：
综合故障，节点随机失效的概率与能量耗尽失效的概率的乘积。
容错性，最大连通分支中的节点个数所占总个数的比例。
本申请提出的模型主要思想及创新之处如下：
应对目前大规模无线传感网络的拓扑演化，本申请提出了一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型(SFTEM)。本发明提出了一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型(SFTEM)。该模型首先提出一个正六边形的分簇机制(RHCS)，通过Markov模型分析该机制至少满足1-容错，SFTEM将RHCS的可靠性与无标度特性相结合，形成一个鲁棒的无线传感器网络，利用了可靠的分簇方案和拓扑演化之间的协同作用，能够容忍随机故障和能量故障等综合故障。本发明的方法提高了网络容错性，并延长了网络寿命。
本申请提出的基于Markov和无标度网络的无线传感器网络容错拓扑演化模型，主要就是针对无线传感器网络中容错拓扑设计的，包括分簇机制和簇间拓扑演化两个步骤。
步骤一：构建正六边形分簇机制RHCS
节点冗余是提高传感器节点容错能力的最有效方法之一。因此，我们将双工传感器节点称为容错传感器节点。在容错传感器节点模型中，假设备份节点处于休眠状态。仅当活动节点被诊断为故障时，备份节点才变为活动节点。
定义1：传感器节点的故障率λ t可以表示为一个时间为t s失效率为λ t的指数分布。
Figure PCTCN2019113846-appb-000001
定义2：节点度k表示与节点连接的节点总个数。
定义3：故障诊断精度因子c无线传感器网络的覆盖范围通常被定义为传感器节点能够观察监测区域的物理空间的良好程度和持续时间。
定义4：故障诊断精度因子c表示活动传感器节点已被正确诊断并被备份传感器节点替换的概率。因子c取决于节点度k和传感器失效的累积概率λ t。我们建立了具有经验关系的c(c≤1)模型：
Figure PCTCN2019113846-appb-000002
M(λ t)是λ t的函数，表示调整参数，该调整参数可以松散地对应于为获得给定λ t的良好故障检测精度所需的期望的平均节点度。
定义5：如果节点分簇机制任意删除k个节点，仍保持该分簇机制的覆盖范围，则该机制被称为具有k-覆盖容错性。
定理1：如果以r cs表示的传输范围和传感范围之间的比率不小于2，则覆盖意味着连通性。正则三角格型在
Figure PCTCN2019113846-appb-000003
比下是最优的。
步骤1.1：构建满足1-覆盖容错的正六边形分簇机制：
如图1A所示，容错节点按正六边形放置，黑色圆圈区域表示普通容错节点的感知范围，红色圆圈区域代表强容错节点的感知范围。其中强容错节点与普通容错节点的rcs满足
Figure PCTCN2019113846-appb-000004
七个普通容错节点形成一个规则的六边形结构，其中有一个位于六边形中心。还有强容错节点也位于六边形中心，此外，相邻容错节点之间的距离d都等于
Figure PCTCN2019113846-appb-000005
如图1B所示，我们分析了k覆盖分簇机制。
根据定义5，可以得到模型的绿色区域表示1-覆盖容错，黄色区域表示3覆盖容错，而红色区域表示5-覆盖容错。因此，该模型至少满足1-覆盖容错。我们称之为正六边形分簇机制(RHCS)。
步骤1.2：RHCS的随机失效概率(RFP)分析
首先，我们对RHCS做出以下假设：
①只要强容错节点正常工作，RHCS就是有效的。
②当强容错节点故障，但所有普通容错节点正常工作，那么RHCS依然有效。
③当强容错节点故障，一旦任何一个普通容错节点故障，则RHCS被认为失效。
下面从2种情况来分析RHCS的随机失效概率。
i：当RHCS中所有节点的初始状态均为无故障时，我们使用马尔可夫模型来分析该方案的可靠性，如图2所示。状态‘(7,1)’表示所有节点正常工作，‘7’表示7个普通容错节点，‘1’表示强容错节点。当强容错节点故障，状态将转变为‘(7,0)’；状态‘(6,1)’则表示7个普通容错节点失效了一个，强容错节点正常工作；当RHCS失效，则状态表示为‘(0,0)’。普通容 错节点故障率为λ FT，强容错节点的故障率为λ sFT。
假设λ sFT＝λ FT，通过Markov模型分析。我们得出该情况下RHCS的随机失效概率为：
Figure PCTCN2019113846-appb-000006
ii：当RHCS中存在节点的初始状态为故障时，如果是强容错节点故障，则RHCS被认为失效，其Markov模型如图3所示。
如果是普通容错节点失效，则其Markov模型如图4所示。
分析图3和图4，我们得到此种情况下RHCS的随机失效概率为：
Figure PCTCN2019113846-appb-000007
步骤1.3：RHCS的能量故障概率(EFP)分析
采用经典的一阶无线通信能量消耗模型，发送1bit信息所消耗的能量为E tx＝E elec·l+ε amp·l·d 2，其中E elec是数据融合能耗，ε amp是放大器功耗，d是节点的传输半径。接收1bit信息所消耗的能量为E r＝E elec·l，所以总能耗为E c＝E tx+E r。
对于RHCS，由于机制中节点部署的基本结构为正六边形，如果以RHCS为一个簇演化拓扑，那么拓扑中簇的结构将一样，能量失效概率就没有任何差别，所以结合实际部署需求，覆盖容错模型的大小将根据节点间距离d的大小而改变，如图5所示，
Figure PCTCN2019113846-appb-000008
由上述能量模型，RHCS的总能耗为：
E c＝n c·E nc＝2n c·E elec·l+n c·ε amp·l·d 2        (5)
其中，n c为RHCS中正常工作的节点个数。
若RHCS的初始总能量为E 0，则RHCS的能量失效概率P e为：
Figure PCTCN2019113846-appb-000009
由公式5公式6，则有，
Figure PCTCN2019113846-appb-000010
其中
Figure PCTCN2019113846-appb-000011
步骤1.4：RHCS的综合故障概率(JFP)
由上文对RHCS的分析，将RHCS的随机失效概率和能量失效概率结合起来，建立RHCS的综合故障概率模型：
①RHCS初始状态下无故障节点，RHCS的综合故障概率为：
Figure PCTCN2019113846-appb-000012
②RHCS初始状态下有故障节点，RHCS的综合故障概率为：
Figure PCTCN2019113846-appb-000013
步骤一最后分析得出的综合故障概率将引入到下述步骤二的模型建立中。
步骤二：建立容错拓扑演化模型(SFTEM)
无线传感器网络作为一种能量受限的分布式网络，在很多情况下都倾向于采用分簇结构来延长网络的生命周期。无线传感器网络具有明显的动态特性，包括新节点和新链路的增加，以及环境因素或能量消耗引起的节点失效。在这一部分中，基于容错簇RHCS中簇头的可靠分布，我们对无标度拓扑进行了改进。在这里，演化过程是指在网络中增加新的容错簇头。将网络中的传感器节点分为强容错节点和普通容错节点。普通容错节点作为容错簇的簇成员加入网络，并与容错簇的固定簇头建立通信关系。当强容错节点作为簇头加入网络时，它们将与其他容错簇的簇头建立链路，并使用多跳通信来传输数据。
步骤2.1：分簇无标度拓扑演化模型
当一个新的簇头加入网络时，以容错簇的JFP(表示为P)、簇头的节点度k和簇头之间的距离D作为评价标准。设适应度函数F是簇头之间P和D的乘积的倒数。选择网络中的现有簇头与新添加的簇头连接的概率取决于F和k的值。
同时，我们将k的阈值设为k max，这意味着簇头的最大连接数不能超过k max。具体的演变规则如下：
①网络初始化：初始时刻t＝0时，初始网络由m 0个容错簇和e 0条边组成，每个容错簇的簇头首至少存在一个与其他簇头连接的边。
②择优连接：在每个时间间隔内，添加一个容错簇头，选择现有容错簇的m个簇头进行连接，并根据RHCS的结构在簇中添加容错节点，形成一个新的容错簇。∏k i表示一个现有的容错簇头被选中连接的概率，并遵循以下规则：
Figure PCTCN2019113846-appb-000014
其中F i＝1/P(i)×D i，k i表示容错簇i的簇头节点度，P(i)表示第i个容错簇的JFP，D i表示新簇头与第i个容错簇的簇头之间的距离。显然，根据连接规则，当容错簇头的节点度为k max时，容错簇被选择连接的概率为零。
从以上演化模型的描述可以看出，适应度函数和节点度决定了新簇头与其相连的概率。其中，适应度函数结合了容错簇的综合故障概率和簇头间距离，既考虑了备选容错簇的整体 失效概率，随机失效和能量耗尽失效，也控制了簇头转发数据的能耗，即距离越小，能耗越低。同时，节点度的上限值设定，影响了网络节点度的分布情况，对于网络的能耗均衡起到一定的改善作用。下面将对演化模型的动态特性进行分析，证明网络的度分布满足无标度网络的幂律性。
步骤2.2：动态特性分析
借助平均场理论，我们分析网络中簇头节点度的分布情况。假设节点度k i是随时间连续变化的，则建立节点i的度k i(t)满足的动力学方程：
Figure PCTCN2019113846-appb-000015
由生成机制可知，网络簇头节点度的分布具有明显的异质性，即少数簇头拥有网络中大部分视为连接，大多数节点的连接度为最低节点度。
因此，在保证网络具有足够规模的条件下，可得
Figure PCTCN2019113846-appb-000016
对于M个簇头组成的局域世界Ω，有：
Figure PCTCN2019113846-appb-000017
其中，
Figure PCTCN2019113846-appb-000018
为适应度期望值，<k>表示局域世界簇头的的平均节点度。
由择优连接规则知，t个时间间隔完成后，网络共增加了mt条链路，每条链路连接两个节点，故新增节点度数为2mt，即
Figure PCTCN2019113846-appb-000019
将公式12和公式13代入公式11，得
Figure PCTCN2019113846-appb-000020
又因为k i(t＝t i)＝m，则公式14可简化为：
Figure PCTCN2019113846-appb-000021
求解上式微分方程，可得：
Figure PCTCN2019113846-appb-000022
所以节点i的节点度小于k的概率为：
Figure PCTCN2019113846-appb-000023
由公式16可得
Figure PCTCN2019113846-appb-000024
由上分析，网络簇头节点度分布p(k)符合幂律分布特性，且幂律指数
Figure PCTCN2019113846-appb-000025
所以本文提出的拓扑演化模型所产生的网络，其簇头节点度分布满足无标度网络的特征，从而具有无标度网络的容错能力。
步骤三：评估容错性能
步骤3.1-步骤3.2：以泊松规则随机地产生失效节点，并移除能量耗尽的节点，计算最大连通分支节点个数；接着随机移除高节点度的簇头，移除比例0.05由到0.25，计算最大连通分支节点个数。
实验以及结果分析：
假设传感器节点随机部署在二维平面区域中，节点的初始能量相同。
初始网络中，容错簇个数m 0为4，新加入簇头连接边数m设为3。每种实验均取50次实验的结果平均值作为实验结果，以保障实验的准确性。具体实验参数见表1。
表1：实验参数
Figure PCTCN2019113846-appb-000026
为了验证本申请提出的分簇拓扑模型(SFTEM)对综合故障的容忍能力以及对高节点度簇头攻击的容侵能力，与传统的BA模型、文献[1]中的Model 1和文献[2]中的Model 2进行 了对比仿真实验。四种拓扑模型初始网络相同，实验参数也相同。
容错实验中，以泊松规则随机地产生失效节点，每一轮运行后移除能量耗尽的节点，观察最大连通分支与网络运行时间的关系，如图6A所示。
在容侵实验中，随机移除高节点度的簇头，移除比例由0.05提高到0.25，如图6B所示。
可以看出本发明中的模型对综合故障的容忍能力以及对高节点度簇头攻击的容侵能力明显高于传统基于无标度网络的拓扑演化模型。
对比文献如下：
[1]Zheng,Gengzhong,and Q.Liu."Scale-free topology evolution for wireless sensor networks,"Computers & Electrical Engineering,vol.39,no.6,pp.1779-1788,Aug.2013.
[2]Fu,Xiuwen,et al."Topology upgrading method for energy balance in scale-free wireless sensor networks,"in Proc.Int.Conf.Network.Sens.Control(ICNSC),May.2017,pp.192-197.
本发明实施例中的部分步骤，可以利用软件实现，相应的软件程序可以存储在可读取的存储介质中，如光盘或硬盘等。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (7)

  1. 一种无线传感器网络容错拓扑演化方法，其特征在于，包括：
    构造一种以容错传感器节点为六边形顶点的正六边形分簇机制RHCS，通过Markov分析RHCS的随机失效概率以及能量故障概率，得出综合故障概率；
    将综合故障概率引入无标度拓扑构建规则中，形成一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型；
    设置随机故障节点，计算网络最大连通子图节点个数，评估容错性能。
  2. 根据权利要求1所述的无线传感器网络容错拓扑演化方法，其特征在于，所述构造一种以容错传感器节点为六边形顶点的正六边形分簇机制RHCS，通过Markov分析了RHCS的随机失效概率以及能量故障概率，得出综合故障概率，包括：
    将双工传感器节点设为容错传感器节点，把容错传感器节点放置为正六边形结构形成基本的分簇机制RHCS；
    通过Markov分析RHCS的随机失效率RFP；
    采用经典的一阶无线通信能量消耗模型，分析RHCS的能量故障率EFP；
    结合RFP和EFP，建立RHCS的综合故障概率JFP。
  3. 根据权利要求1所述的无线传感器网络容错拓扑演化方法，其特征在于，所述将综合故障概率引入无标度拓扑构建规则中，形成一种基于Markov和无标度网络的无线传感器网络容错拓扑演化模型，其中，容错拓扑演化模型的包括：
    增长：从一个具有m 0个节点的联通网络开始，每次引入一个新的节点，并且连到m个已经存在的节点上，这里m<＝m 0；
    择优连接：一个新的节点与一个已经存在的节点i相连的概率w与节点i的度k_i之间的关系为w＝k_i/(k_1+k_2+k_3+...+k_n)，其中n为网络中的节点的总个数；
    形成的无标度网络，分析其网络节点度分布，满足幂律特性。
  4. 根据权利要求1所述的无线传感器网络容错拓扑演化方法，其特征在于，“设置随机故障节点，计算网络最大连通子图节点个数，评估容错性能。”中，所述评估容错性能包括：
    随机失效容错性，以泊松规则随机地产生失效节点，每一轮运行后移除能量耗尽的节点，以最大连通分支中的节点个数所占总个数的比例为容错性能指标；
    恶意攻击容错性，将节点度较高的节点随机去除，去除率在0.05～0.25之间，以最大连通分支中的节点个数所占总个数的比例为容错性能指标。
  5. 一种计算机设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时实现权利要求1到4任一项所述方法的步骤。
  6. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现权利要求1到4任一项所述方法的步骤。
  7. 一种处理器，其特征在于，所述处理器用于运行程序，其中，所述程序运行时执行权利要求1到4任一项所述的方法。
PCT/CN2019/113846 2018-11-08 2019-10-29 无线传感器网络容错拓扑演化方法 WO2020093904A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811325318.1A CN109379756A (zh) 2018-11-08 2018-11-08 无线传感器网络容错拓扑演化方法
CN201811325318.1 2018-11-08

Publications (1)

Publication Number Publication Date
WO2020093904A1 true WO2020093904A1 (zh) 2020-05-14

Family

ID=65384263

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/113846 WO2020093904A1 (zh) 2018-11-08 2019-10-29 无线传感器网络容错拓扑演化方法

Country Status (2)

Country Link
CN (1) CN109379756A (zh)
WO (1) WO2020093904A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112882385A (zh) * 2021-01-12 2021-06-01 南京航空航天大学 基于隔离/移除故障个体的蜂群无人机可重构性评估方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109379756A (zh) * 2018-11-08 2019-02-22 江南大学 无线传感器网络容错拓扑演化方法
CN110187989B (zh) * 2019-05-24 2022-08-09 广东致盛技术有限公司 雾环境下基于Markov Chain的容错策略选择方法
CN110909769A (zh) * 2019-11-05 2020-03-24 浙江工业大学 一种基于度值的无标度网络分类算法的对抗攻击方法
CN113873611B (zh) * 2021-09-24 2023-10-31 天津津航计算技术研究所 一种无线网络簇头自适应竞争方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100260070A1 (en) * 2009-04-09 2010-10-14 Yigal Bejerano Localized construction of fault resilient high capacitywireless networks with bounded node degree
CN105959990A (zh) * 2016-04-25 2016-09-21 东莞市联洲知识产权运营管理有限公司 一种无线传感器网络拓扑控制方法
CN107333316A (zh) * 2017-08-24 2017-11-07 国网安徽省电力公司黄山供电公司 一种无线传感器网络容错拓扑构建方法
CN107580355A (zh) * 2017-08-22 2018-01-12 西安科技大学 无线传感器网络基于位置的均匀分簇层次路由算法
CN109379756A (zh) * 2018-11-08 2019-02-22 江南大学 无线传感器网络容错拓扑演化方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105554844B (zh) * 2016-01-22 2018-12-21 大连理工大学 一种无线传感器网络拓扑构建方法
CN106507426B (zh) * 2016-11-23 2019-10-25 西安电子科技大学 一种新型能耗均衡的传感器网络拓扑演化方法
CN106507506B (zh) * 2016-12-06 2018-06-12 无锡悟莘科技有限公司 一种无线传感网络组网方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100260070A1 (en) * 2009-04-09 2010-10-14 Yigal Bejerano Localized construction of fault resilient high capacitywireless networks with bounded node degree
CN105959990A (zh) * 2016-04-25 2016-09-21 东莞市联洲知识产权运营管理有限公司 一种无线传感器网络拓扑控制方法
CN107580355A (zh) * 2017-08-22 2018-01-12 西安科技大学 无线传感器网络基于位置的均匀分簇层次路由算法
CN107333316A (zh) * 2017-08-24 2017-11-07 国网安徽省电力公司黄山供电公司 一种无线传感器网络容错拓扑构建方法
CN109379756A (zh) * 2018-11-08 2019-02-22 江南大学 无线传感器网络容错拓扑演化方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HU , SHIHONG: "Fault-Tolerant Clustering Topology Evolution Mechanism of Wireless Sensor Networks Mechanism of Wireless Sensor Networks", IEEE ACCESS, vol. 6, 19 June 2018 (2018-06-19), pages 28085 - 28096, XP011685343 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112882385A (zh) * 2021-01-12 2021-06-01 南京航空航天大学 基于隔离/移除故障个体的蜂群无人机可重构性评估方法

Also Published As

Publication number Publication date
CN109379756A (zh) 2019-02-22

Similar Documents

Publication Publication Date Title
WO2020093904A1 (zh) 无线传感器网络容错拓扑演化方法
Hu et al. Fault-tolerant clustering topology evolution mechanism of wireless sensor networks
Chanak et al. Fuzzy rule-based faulty node classification and management scheme for large scale wireless sensor networks
Jiang et al. Towards clustering algorithms in wireless sensor networks-a survey
Asim et al. A self-managing fault management mechanism for wireless sensor networks
Tran et al. OCO: Optimized communication & organization for target tracking in wireless sensor networks
Sangaiah et al. SALA-IoT: Self-reduced internet of things with learning automaton sleep scheduling algorithm
Fu et al. Toward robust and energy-efficient clustering wireless sensor networks: A double-stage scale-free topology evolution model
Hu et al. TMSE: A topology modification strategy to enhance the robustness of scale-free wireless sensor networks
Sahoo et al. Vector method based coverage hole recovery in wireless sensor networks
Fu et al. Cascading failures analysis of wireless sensor networks with varying routing schemes
WO2022262127A1 (zh) 智慧物联网综合感知系统及方法
Venkataraman et al. A cluster-based approach to fault detection and recovery in wireless sensor networks
CN110087208B (zh) 一种基于斯坦纳树和凸多边形的wsn分区双连通恢复方法
Mishra et al. Qualitative analysis of intra-class and inter-class clustering routing and clusterization in wireless sensor network
Bhajantri et al. Genetic algorithm based node fault detection and recovery in distributed sensor networks
Chen et al. SOFM neural network based hierarchical topology control for wireless sensor networks
Bapu et al. Fault tolerance in wireless sensor networks–a survey
CN114363852A (zh) 一种增强网络抗毁性的异构节点优化部署方法
Beyens et al. High-density wireless sensor networks: a new clustering approach for prediction-based monitoring
Wang A dynamic evolution model of balanced energy consumption scale-free fault-tolerant topology based on fitness function for wireless sensor networks
Zhao et al. Topology management for reducing energy consumption and tolerating failures in wireless sensor networks
Raja et al. Distributed Fault Node Detection and Classification using Fuzzy Logic and Management Scheme for Wireless Sensor Networks
Liu et al. Region-Different Network Reconfiguration in Disjoint Wireless Sensor Networks for Smart Agriculture Monitoring
Kalpana et al. An Implementation of Energy Efficient Secured Routing Framework in WSN by Honey Badger Algorithm

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19882513

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19882513

Country of ref document: EP

Kind code of ref document: A1