WO2022222236A1 - 一种面向异构无线传感器网络的覆盖增强方法及系统 - Google Patents

一种面向异构无线传感器网络的覆盖增强方法及系统 Download PDF

Info

Publication number
WO2022222236A1
WO2022222236A1 PCT/CN2021/098084 CN2021098084W WO2022222236A1 WO 2022222236 A1 WO2022222236 A1 WO 2022222236A1 CN 2021098084 W CN2021098084 W CN 2021098084W WO 2022222236 A1 WO2022222236 A1 WO 2022222236A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
nodes
efficiency matrix
virtual
efficiency
Prior art date
Application number
PCT/CN2021/098084
Other languages
English (en)
French (fr)
Inventor
赵小强
文秦
李雄
刘敏
崔砚鹏
高心岗
常虹
曾耀平
付银娟
翟永智
姚引娣
廖焕敏
高强
习子辰
Original Assignee
西安邮电大学
西安碧海蓝天电子信息技术有限公司
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 西安邮电大学, 西安碧海蓝天电子信息技术有限公司 filed Critical 西安邮电大学
Publication of WO2022222236A1 publication Critical patent/WO2022222236A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/22—Traffic simulation tools or models
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18—Network planning tools
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present invention relates to the technical field of network coverage enhancement, in particular to a coverage enhancement method and system for heterogeneous wireless sensor networks.
  • WSNs Wireless Sensor Networks
  • Coverage reflects the ability of WSNs to perceive the monitoring area.
  • the coverage rate is often used to measure the coverage effect of WSNs. A higher coverage rate can ensure that the network can collect complete and effective data from the monitoring area.
  • WSNs usually work in complex scenes such as underwater and complex mountains, the nodes are often initially deployed by random throwing such as airdrops, which is difficult to form effective coverage.
  • the network is divided into homogeneous WSNs and heterogeneous WSNs.
  • the network composed of different types of nodes is called heterogeneous WSNs, and different types of nodes have differences in initial energy, unit movement energy consumption, and sensing range. The opposite is called isomorphic WSNs.
  • the theoretical research on homogeneous sensor networks has been relatively mature.
  • different types of nodes are usually put into the monitoring area, that is, the coverage of heterogeneous wireless sensor networks needs to be considered.
  • the commonly used methods mainly include the following two methods: extending the coverage algorithm of homogeneous wireless sensor networks to heterogeneous wireless sensor networks to solve; The geometric relationship between the sampling line and the sampling line is used to optimize the coverage effect; however, the above scheme only considers the coverage of the network unilaterally. Due to the complexity of the WSNs working environment, it is difficult to supply energy to the nodes. Therefore, how to maximize and equalize the remaining energy of each node under the premise of ensuring the coverage effect is also a key issue.
  • the current sampling-based coverage optimization algorithm for heterogeneous wireless sensor networks transforms the area coverage problem of heterogeneous WSNs by analyzing the positional relationship between the sampling line and the coordinates of the intersection points of the nodes, and taking the movement energy consumption of the node redeployment process as the objective function. for optimization mathematical problems. After multiple iterations, when each sampling straight line segment achieves better coverage, the entire area achieves coverage enhancement, and effectively controls the energy consumed in the process of node movement, but when the number of iterations increases, the network coverage has an optimization upper limit.
  • the purpose of the present invention is to provide a coverage enhancement method and system for heterogeneous wireless sensor networks, which saves the total remaining energy of nodes and improves the balance of remaining energy while realizing coverage optimization.
  • the present invention provides a coverage enhancement method for heterogeneous wireless sensor networks, including:
  • Initializing a network model of the monitoring area including the type, quantity, sensing radius and location of nodes in the monitoring area; the nodes are wireless sensors;
  • the monitoring area is divided by a honeycomb grid stacking method to obtain a honeycomb grid stacking
  • the uncovered grid points are determined according to the current virtual positions of the nodes, and the final positions of the nodes are determined by using the virtual forces of the uncovered grid points and the nodes.
  • the constructing the efficiency matrix of each of the nodes and the centroids of the honeycomb grids in the stacking of the honeycomb grids specifically includes:
  • An efficiency matrix is constructed from each of the efficiency values.
  • the efficiency matrix is expressed as:
  • Eff N ⁇ N represents the efficiency matrix
  • Eff i,j represents the efficiency value of the i-th node moving to the j-th cellular grid
  • i ⁇ [1,N], j ⁇ [1,N] N denotes the node number.
  • the use of the Hungarian algorithm to solve the efficiency matrix to obtain the virtual positions of each node after reassignment specifically includes:
  • the minimum number of lines is not equal to the number of nodes, find the minimum element in the area not covered by the line in the second efficiency matrix, subtract the minimum element from all elements in the area not covered by the line, and add the line to the Add the minimum element to the element at the intersection, and return to the step "find the minimum number of straight lines that can cover all zero elements in the second efficiency matrix";
  • the optimal allocation of each node is determined according to each of the zero elements, and the reassigned virtual position of each node is obtained.
  • determining the uncovered grid point according to the current virtual position of each node, and determining the final position of each node by using the virtual force of each uncovered grid point and each of the nodes specifically including:
  • the current virtual position of each node is determined as the final position of each node
  • the traversing each of the nodes, and updating the virtual position of each node by virtually moving each of the nodes specifically includes:
  • the invention also discloses a coverage enhancement system for heterogeneous wireless sensor networks, comprising:
  • the monitoring area discrete module is used to discretize the monitoring area into M grid points;
  • an initialization module for initializing a network model of the monitoring area, the network model including the type, quantity, sensing radius and location of nodes in the monitoring area; the nodes are wireless sensors;
  • the cellular grid stacking module is used to divide the monitoring area by adopting the cellular grid stacking method based on the network model to obtain the cellular grid stacking;
  • an efficiency matrix building module used for constructing the efficiency matrix of each of the nodes and the centroids of the honeycomb grids in the stacking of the honeycomb grids;
  • the redistribution module is used to solve the efficiency matrix by adopting the Hungarian algorithm to obtain the virtual positions of the redistributed nodes;
  • the final position determination module is used for determining the uncovered grid points according to the current virtual positions of the nodes, and determining the final positions of the nodes by using the virtual forces of the uncovered grid points and the nodes.
  • the efficiency matrix building module specifically includes:
  • a distance calculation unit used to calculate the distance from each of the nodes to the centroid of each of the honeycomb grids
  • a residual energy calculation unit configured to calculate the residual energy after each of the nodes is moved to each of the honeycomb grids according to the distance from each of the nodes to the center of mass of each of the honeycomb grids;
  • an efficiency value calculation unit configured to calculate the efficiency value of each of the nodes moving to each of the cellular grids according to the remaining energy of each of the nodes after moving to each of the cellular grids;
  • An efficiency matrix construction unit configured to construct an efficiency matrix according to each of the efficiency values.
  • the efficiency matrix is expressed as:
  • Eff N ⁇ N represents the efficiency matrix
  • Eff i,j represents the efficiency value of the i-th node moving to the j-th cellular grid
  • i ⁇ [1,N], j ⁇ [1,N] N denotes the node number.
  • the redistribution module specifically includes:
  • a first efficiency matrix obtaining unit configured to subtract the minimum value in this row from each element in each row of the efficiency matrix to obtain a first efficiency matrix
  • a second efficiency matrix obtaining unit configured to subtract the minimum value in the column from each element in each column of the first efficiency matrix to obtain a second efficiency matrix
  • the second efficiency matrix updating unit is configured to find the minimum element in the area not covered by the straight line in the second efficiency matrix if the minimum number of straight lines is not equal to the number of nodes, and replace all the elements in the area not covered by the straight line Subtract the minimum element from the element, add the minimum element to the element at the intersection of the straight lines, and return to the step "find the minimum number of straight lines that can cover all zero elements in the second efficiency matrix";
  • a zero element obtaining unit used for finding the zero element corresponding to each row and the zero element corresponding to each column in the second efficiency matrix if the minimum number of straight lines is equal to the number of nodes;
  • the reassignment unit is configured to determine the optimal assignment of each node according to each of the zero elements, and obtain the virtual position of each node after reassignment.
  • the final position determination module specifically includes:
  • the uncovered grid point obtaining unit is used to obtain the uncovered grid point coordinates according to the current position of each node;
  • a virtual moving unit for each node position used to traverse each of the nodes, update the virtual position of each node by virtually moving each of the nodes, and increase the number of iterations by 1;
  • Return unit used to return "obtain the coordinates of uncovered grid points according to the current virtual positions of each node" if the number of iterations does not reach the preset number of iterations;
  • a final position determination unit configured to determine the current virtual position of each node as the final position of each node if the number of iterations reaches a preset number of iterations
  • the virtual mobile unit of each node position specifically including:
  • the virtual force calculation subunit is used to calculate the virtual force between the ith node and each uncovered grid point;
  • the resultant force calculation subunit is used to calculate the resultant force that the ith node is subjected to all the uncovered grid points, and the resultant force is the sum of the virtual forces of each uncovered grid point to the ith node;
  • the virtual moving subunit is used to virtually move the ith node according to the resultant force received by the ith node to obtain the updated virtual position of each node.
  • the present invention discloses the following technical effects:
  • the invention preliminarily improves the coverage of the network by using the honeycomb grid stacking method to divide the monitoring area, further updates the position of each node through the virtual force of the uncovered grid points, further optimizes the coverage performance, and solves the problem by the Hungarian algorithm.
  • the efficiency matrix improves the residual total energy and residual energy balance of nodes in the network model.
  • FIG. 1 is a schematic flowchart of a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention
  • Fig. 2 is the grid discrete effect diagram of monitoring area of the present invention
  • Fig. 3 is a honeycomb grid stacking effect diagram of the present invention.
  • FIG. 4 is a schematic diagram of a bipartite graph matching model for the coverage enhancement problem of heterogeneous WSNs according to the present invention.
  • Fig. 5 is the Hungarian algorithm flow chart of the present invention.
  • Fig. 6 is the flow chart of the virtual force algorithm of uncovered grid points of the present invention.
  • FIG. 8 is a schematic diagram of the coverage effect of a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention.
  • Fig. 9 is the coverage effect diagram of COSH algorithm
  • FIG. 10 is a movement trajectory diagram of a node of a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention.
  • Figure 11 is the movement trajectory diagram of the COSH algorithm node
  • FIG. 12 is a comparison diagram of a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention and the residual energy of the COSH algorithm;
  • FIG. 13 is a schematic diagram of a honeycomb grid stacking method of the present invention.
  • FIG. 14 is a schematic structural diagram of a coverage enhancement system for heterogeneous wireless sensor networks according to the present invention.
  • the purpose of the present invention is to provide a coverage enhancement method and system for heterogeneous wireless sensor networks, which improves the remaining total energy of nodes and the balance of remaining energy.
  • FIG. 1 is a schematic flowchart of a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention. As shown in FIG. 1 , a coverage enhancement method for heterogeneous wireless sensor networks includes:
  • Step 101 Discrete the monitoring area into M grid points
  • Step 102 Initialize a network model of the monitoring area, where the network model includes the type, number, sensing radius and location (initial location) of nodes in the monitoring area; the nodes are wireless sensors.
  • Step 103 Based on the network model, adopt the cellular grid stacking method to divide the monitoring area to obtain cellular grid stacking.
  • Step 104 Construct an efficiency matrix of each of the nodes and the centroids of each honeycomb grid in the honeycomb grid stacking.
  • the construction of the efficiency matrix of each of the nodes and the centroids of the honeycomb grids in the stacking of the honeycomb grids specifically includes:
  • the residual energy of each of the nodes after moving to each of the cellular grids is calculated according to the distance from each of the nodes to the centroid of each of the cellular grids.
  • the efficiency value of each of the nodes moving to each of the cellular grids is calculated according to the remaining energy of each of the nodes after moving to each of the cellular grids.
  • An efficiency matrix is constructed from each of the efficiency values.
  • the efficiency matrix is expressed as:
  • Eff N ⁇ N represents the efficiency matrix
  • Eff i,j represents the efficiency value of the i-th node moving to the j-th cellular grid
  • i ⁇ [1,N], j ⁇ [1,N] N denotes the node number.
  • Step 105 Use the Hungarian algorithm to solve the efficiency matrix, and obtain the virtual position of each node after reassignment.
  • the Hungarian algorithm is used to solve the efficiency matrix, and the virtual position of each node after redistribution is obtained, which specifically includes:
  • a first efficiency matrix is obtained by subtracting the minimum value in the row from each element in each row of the efficiency matrix.
  • the minimum value in this column is subtracted from each element in each column of the first efficiency matrix to obtain a second efficiency matrix.
  • the minimum number of lines is not equal to the number of nodes, find the minimum element in the area not covered by the line in the second efficiency matrix, subtract the minimum element from all elements in the area not covered by the line, and add the line to the The element at the intersection is added with the minimum element to update the second efficiency matrix, and the step is returned to "find the minimum number of straight lines that can cover all zero elements in the second efficiency matrix".
  • the optimal allocation of each node is determined according to each of the zero elements, and the reassigned virtual position of each node is obtained.
  • Step 106 Determine the uncovered grid points according to the current virtual positions of the nodes, and use the virtual forces of the uncovered grid points and the nodes to determine the final positions of the nodes.
  • the determining the uncovered grid point according to the current virtual position of each node, and determining the final position of each node by using the virtual force of each uncovered grid point and each of the nodes specifically includes:
  • the coordinates of the uncovered grid points are obtained according to the virtual positions of the current nodes.
  • the current virtual position of each node is determined as the final position of each node.
  • the traversing each of the nodes, and updating the virtual position of each node by virtually moving each of the nodes specifically includes:
  • a coverage enhancement method for heterogeneous wireless sensor networks according to the present invention will be described in detail below.
  • a coverage enhancement method for heterogeneous wireless sensor networks includes the following steps:
  • Step1 Initialize the network model and node parameters.
  • Equation 1 The coordinates of the M grid points (grid centroids) are shown in Equation 1.
  • Grid m,1 represents the x-axis coordinate of the mth grid point
  • Grid m,2 represents the y-axis coordinate of the mth grid point.
  • the initialized node type is Boolean perception model, the total number is N, and the initial position coordinates of N nodes are shown in formula 2.
  • the node type is K
  • the number of different types of nodes is N 1 , N 2 ,...N k
  • the sensing radii of different types of nodes are R 1 , R 2 ,... R k
  • the initial energy of different types of nodes is Ener 1 , Ener 2 ,...Ener k
  • the energy consumption per unit distance for different types of nodes to move is e 1 , e 2 ,...e k , respectively.
  • the initial positions of N nodes are represented by a matrix as Position N ⁇ 2
  • Position i,1 represents the x-axis coordinate of the ith node
  • Position i,2 represents the y-axis coordinate of the ith node.
  • Step2 Use the cellular grid to divide the monitoring area.
  • the radius of N circumscribed circles is The honeycomb grids are stacked on the monitoring area with an area of L ⁇ B, and the stacking effect is shown in Figure 3.
  • centroid coordinates of the N honeycomb grids are shown in Equation 3.
  • Step3 "Three-step method" to construct the efficiency matrix of nodes and cellular grid centroid points and abstractly model the coverage enhancement problem of heterogeneous WSNs.
  • Equation 5 (2) Calculate the remaining energy after each node moves to each cellular grid, as shown in Equation 5, where REner i,j represents the remaining energy after node i moves to the jth cellular grid centroid, and Ener i represents node i The initial energy of , e i represents the energy consumption of node i moving unit distance.
  • Equation 6 is used to calculate the efficiency value of each node moving to each cellular grid.
  • the coverage enhancement problem of heterogeneous WSNs is transformed into the task assignment problem of N nodes moving to N cellular grid centroids.
  • the bipartite graph model is shown in Figure 4. Therefore, the coverage enhancement problem of heterogeneous WSNs can be abstracted as a mathematical model of the optimal task assignment problem, as shown in Equation 8. That is: find N elements located in different rows and different columns in the efficiency matrix Eff N ⁇ N , so that the sum F of the values of the N elements is the largest.
  • x i,j indicates whether the i-th node selects the j-th honeycomb grid. If it is selected, the value is 1, and if it is not selected, the value is 0.
  • Step4 Use the Hungarian algorithm to solve the efficiency matrix.
  • the basic idea of the Hungarian algorithm is to modify each row or column in the efficiency matrix until there is at least one zero element in different rows and different columns, so as to obtain a matching scheme corresponding to these zero elements.
  • the flowchart of the Hungarian algorithm is shown in Figure 5.
  • the efficiency matrix performs row subtraction and column subtraction. Specifically, the minimum value of this row is subtracted from each row of the efficiency matrix; the minimum value of this column is subtracted from each column in the new matrix.
  • Step5 Use the virtual force of the uncovered grid points to further determine the node position.
  • Equation 9 the virtual force calculation method between node i and uncovered grid point j is shown in Equation 9.
  • d ij represents the Euclidean distance between node i and uncovered grid point j
  • ⁇ ij is the direction angle from node i to uncovered grid point j
  • d th is the Euclidean distance between them.
  • Equation 10 The calculation method of the resultant force of node i on the uncovered grid points is shown in Equation 10.
  • xi and yi are the horizontal and vertical coordinates of node i , respectively, F xi and F yi are the components of the resultant force F i received by the node on the horizontal and vertical coordinates, and Max_Step is the single virtual movement step of the node.
  • step (5) If the N nodes have been traversed, go to step (5).
  • step (2) If the N nodes have not been traversed, go to step (2).
  • step (6) If the number of virtual force iterations is satisfied, go to step (6).
  • step (1) If the number of virtual force iterations is not satisfied, go to step (1).
  • Step6 The node is deployed for the second time.
  • the node moves from the initial position to the final position of the node obtained in step 5, and calculates the balance between the total energy consumption and the remaining energy of the node after the movement.
  • each node in the present invention determines whether it is the node position allocated by the Hungarian algorithm or the position moved by the node in each round of the subsequent virtual force algorithm, it is a virtual position (it only needs to be implemented in the algorithm, and does not require the real movement).
  • the node is moved only once, that is, the final position is calculated by the algorithm, and then the node is moved from the initial position to the final position.
  • Network coverage The probability that any grid point m is perceived by the ith node can be calculated by formula 12.
  • d im represents the Euclidean distance between the ith node and the mth grid point.
  • the grid point m is perceived by any node, the grid point is said to be covered, and the grid point that is not perceived by any node is called the uncovered grid point.
  • the coverage of the entire network is shown in Equation 13.
  • p m is the coverage of grid point m
  • M is the number of grid points after discretization.
  • Equation 14 The mobility energy consumption of node i after redeployment is shown in Equation 14.
  • Equation 15 Equation 15
  • Equation 16 The remaining energy of node i after redeployment is shown in Equation 16:
  • Equation 17 The residual energy balance of the entire network is shown in Equation 17:
  • the Boolean perception model means that the monitoring probability of the node to the target point in the monitoring area is constant.
  • the sensing range of node i is a circular area with the node coordinates as the center and the radius as Ri, where Ri is the sensing radius of node i .
  • Ri is the sensing radius of node i .
  • d ij is the Euclidean distance between node i and target point j.
  • Grid division of the monitoring area can effectively improve the uniformity of node deployment.
  • Regular triangles, regular quadrilaterals, regular pentagons and regular hexagons can be used to stack the monitoring areas, and the honeycomb grid (regular hexagon) has the highest Stacking efficiency can effectively reduce coverage redundancy.
  • the honeycomb grid stacking scheme is shown in Figure 13.
  • the present invention provides a coverage enhancement method for heterogeneous wireless sensor networks, and the method has the following advantages:
  • the present invention proposes a cellular grid division method for heterogeneous WSNs.
  • the weighted average of the sensing radii of different types of nodes is used as the circumcircle radius of the cellular grid to ensure the uniformity of node distribution after redeployment. , which initially improves the coverage of the network.
  • Step 5 Aiming at the defect that the coverage rate caused by cellular grid division has a threshold, a virtual force algorithm for uncovered grid points is proposed to further optimize the coverage performance.
  • FIG. 14 is a schematic structural diagram of a coverage enhancement system for heterogeneous wireless sensor networks. As shown in FIG. 14 , the present invention also discloses a coverage enhancement system for heterogeneous wireless sensor networks, including:
  • the monitoring area discretization module 201 is used to discretize the monitoring area into M grid points.
  • the initialization module 202 is used to initialize the network model of the monitoring area, and the network model includes the type, quantity, sensing radius and position of the nodes in the monitoring area; the nodes are wireless sensors; the nodes are wireless sensors.
  • the cellular grid stacking module 203 is configured to divide the monitoring area by adopting the cellular grid stacking method based on the network model to obtain the cellular grid stacking.
  • the efficiency matrix construction module 204 is configured to construct an efficiency matrix of each of the nodes and the centroids of each cellular grid in the cellular grid stacking.
  • the reassignment module 205 is configured to solve the efficiency matrix by using the Hungarian algorithm, and obtain the virtual positions of the reassigned nodes.
  • the final position determination module 206 is configured to determine the uncovered grid point according to the current virtual position of each node, and determine the final position of each node by using the virtual force of each uncovered grid point and each of the nodes.
  • the efficiency matrix building module 204 specifically includes:
  • a distance calculation unit used to calculate the distance from each of the nodes to the centroid of each of the honeycomb grids
  • a residual energy calculation unit configured to calculate the residual energy after each of the nodes is moved to each of the honeycomb grids according to the distance from each of the nodes to the center of mass of each of the honeycomb grids;
  • an efficiency value calculation unit configured to calculate the efficiency value of each of the nodes moving to each of the cellular grids according to the remaining energy of each of the nodes after moving to each of the cellular grids;
  • An efficiency matrix construction unit configured to construct an efficiency matrix according to each of the efficiency values.
  • the efficiency matrix is expressed as:
  • Eff N ⁇ N represents the efficiency matrix
  • Eff i,j represents the efficiency value of the i-th node moving to the j-th cellular grid
  • i ⁇ [1,N], j ⁇ [1,N] N denotes the node number.
  • the redistribution module 205 specifically includes:
  • a first efficiency matrix obtaining unit configured to subtract the minimum value in this row from each element in each row of the efficiency matrix to obtain a first efficiency matrix
  • a second efficiency matrix obtaining unit configured to subtract the minimum value in the column from each element in each column of the first efficiency matrix to obtain a second efficiency matrix
  • the second efficiency matrix updating unit is configured to find the minimum element in the area not covered by the straight line in the second efficiency matrix if the minimum number of straight lines is not equal to the number of nodes, and replace all the elements in the area not covered by the straight line Subtract the minimum element from the element, add the minimum element to the element at the intersection of the straight lines, and return to the step "find the minimum number of straight lines that can cover all zero elements in the second efficiency matrix";
  • a zero element obtaining unit used for finding the zero element corresponding to each row and the zero element corresponding to each column in the second efficiency matrix if the minimum number of straight lines is equal to the number of nodes;
  • the reassignment unit is configured to determine the optimal assignment of each node according to each of the zero elements, and obtain the virtual position of each node after reassignment.
  • the final position determination module 206 specifically includes:
  • the uncovered grid point obtaining unit is used to obtain the uncovered grid point coordinates according to the current position of each node;
  • Each node position virtual moving unit is used to traverse each described node, and by virtual moving each described node, the virtual position of each node is updated, and the number of iterations is increased by 1;
  • Return unit used to return "obtain the coordinates of uncovered grid points according to the current virtual positions of each node" if the number of iterations does not reach the preset number of iterations;
  • a final position determination unit configured to determine the current virtual position of each node as the final position of each node if the number of iterations reaches a preset number of iterations
  • the virtual mobile unit of each node position specifically including:
  • the virtual force calculation subunit is used to calculate the virtual force between the ith node and each uncovered grid point;
  • the resultant force calculation subunit is used to calculate the resultant force that the ith node is subjected to all the uncovered grid points, and the resultant force is the sum of the virtual forces of each uncovered grid point to the ith node;
  • the virtual moving subunit is used to virtually move the ith node according to the resultant force received by the ith node to obtain the updated virtual position of each node.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明提供一种面向异构无线传感器网络的覆盖增强方法及系统，该方法包括：将监测区域离散为M个网格点（101）；初始化监测区域的网络模型，该网络模型包括该监测区域中节点的类型、数量、感知半径和位置；该节点为无线传感器（102）；基于该网络模型，采用蜂窝网格堆砌方法对该监测区域进行划分，获得蜂窝网格堆砌（103）；构建各该节点与该蜂窝网格堆砌中各蜂窝网格质心的效率矩阵（104）；采用匈牙利算法求解该效率矩阵，获得重新分配后各节点的虚拟位置（105）；根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各该节点的虚拟力确定各节点的最终位置（106）。本发明实现覆盖率优化的同时，节约了节点的剩余总能量，提高了剩余能量均衡度。

Description

一种面向异构无线传感器网络的覆盖增强方法及系统
本申请要求于2021年4月20日提交中国专利局、申请号为202110425315.0、发明名称为“一种面向异构无线传感器网络的覆盖增强方法及系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及网络覆盖增强技术领域，特别是涉及一种面向异构无线传感器网络的覆盖增强方法及系统。
背景技术
无线传感器网络(Wireless Sensor Networks,WSNs)是由众多具备感知、通信和计算能力的微型传感器节点组成的多跳自组织网络，这些节点之间通过相互协作完成对监测区域的有效覆盖。覆盖体现了WSNs对监测区域的感知能力，常使用覆盖率衡量WSNs的覆盖效果，较高的覆盖率可以保证网络对监测区域进行完整有效的数据采集。然而，由于WSNs通常工作在水下、复杂山地等复杂场景中，节点往往以空投等随机抛洒的方式进行初始部署，难以形成有效地覆盖。
根据节点类型的差异将网络分为同构WSNs与异构WSNs。将由不同类型节点组成的网络称为异构WSNs，不同类型的节点在初始能量、单位移动能耗以及感知范围上均存在差异。反之称为同构WSNs。
当前，针对同构传感器网络的理论研究已经相对成熟。然而，在实际部署中，从成本以及监测需求方面考虑，通常向监测区域投放不同类型的节点，即需要考虑异构无线传感器网络的覆盖问题。针对异构无线传感器网络的覆盖问题，常用的方法主要包括以下两种：将同构无线传感器网络的覆盖算法扩展至异构无线传感器网络来解决；将区域覆盖问题映射为直线覆盖问题，构建节点与采样直线之间的几何关系来优化覆盖效果；然而上述方案仅仅单方面考虑了网络的覆盖问题。由于WSNs工作环境的复杂性，难以对节点进行能量补给，因此如何在保证覆盖效果的前提下，最大化与均衡化各节点的剩余能量也是问题关键。
目前基于采样的异构无线传感器网络覆盖优化算法，通过分析采样直线与节点的交点坐标之间的位置关系，以节点重部署过程的移动能耗为目标函数，将异构WSNs的区域覆盖问题转化为最优化数学问题。经过多次迭代，当各采样直线段达到较优覆盖时，整个区域即实现了覆盖增强，并且有效控制了节点移动过程中消耗的能量，但是当迭代次数增加时，网络覆盖率存在优化上限。
发明内容
基于此，本发明的目的是提供一种面向异构无线传感器网络的覆盖增强方法及系统，在实现覆盖率优化的同时，节约了节点的剩余总能量，提高了剩余能量均衡度。
为实现上述目的，本发明提供了一种面向异构无线传感器网络的覆盖增强方法，包括：
将监测区域离散为M个网格点；
初始化监测区域的网络模型，所述网络模型包括所述监测区域中节点的类型、数量、感知半径和位置；所述节点为无线传感器；
基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌；
构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵；
采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置；
根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
可选地，所述构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵，具体包括：
计算各所述节点到各所述蜂窝网格质心的距离；
根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量；
根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值；
根据各所述效率值构建效率矩阵。
可选地，所述效率矩阵表示为：
Figure PCTCN2021098084-appb-000001
其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
可选地，所述采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置，具体包括：
将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵；
将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵；
找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数；
若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉点处的元素加上最小元素，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”；
若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和每一列对应的零元素；
根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
可选地，所述根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置，具体包括：
根据当前各节点的虚拟位置获得未覆盖网格点坐标；
遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，迭代次数加1；
判断迭代次数是否达到预设迭代次数；
若未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”；
若达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置；
所述遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，具体包括：
计算第i个节点与各未覆盖网格点之间的虚拟力；
计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和；
根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
本发明还公开了一种面向异构无线传感器网络的覆盖增强系统，包括：
监测区域离散模块，用于将监测区域离散为M个网格点；
初始化模块，用于初始化监测区域的网络模型，所述网络模型包括所述监测区域中节点的类型、数量、感知半径和位置；所述节点为无线传感器；
蜂窝网格堆砌模块，用于基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌；
效率矩阵构建模块，用于构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵；
重新分配模块，用于采用匈牙利算法求解所述效率矩阵，获得重新分配后的各节点的虚拟位置；
最终位置确定模块，用于根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
可选地，所述效率矩阵构建模块，具体包括：
距离计算单元，用于计算各所述节点到各所述蜂窝网格质心的距离；
剩余能量计算单元，用于根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量；
效率值计算单元，用于根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值；
效率矩阵构建单元，用于根据各所述效率值构建效率矩阵。
可选地，所述效率矩阵表示为：
Figure PCTCN2021098084-appb-000002
其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
可选地，所述重新分配模块，具体包括：
第一效率矩阵获得单元，用于将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵；
第二效率矩阵获得单元，用于将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵；
最小直线数获得单元，用于找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数；
第二效率矩阵更新单元，用于若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉点处的元素加上最小元素，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”；
零元素获得单元，用于若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和每一列对应的零元素；
重新分配单元，用于根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
可选地，所述最终位置确定模块，具体包括：
未覆盖网格点获得单元，用于根据当前各节点位置的获得未覆盖网格点坐标；
各节点位置虚拟移动单元，用于遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，迭代次数加1；
返回单元，用于若迭代次数未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”；
最终位置确定单元，用于若迭代次数达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置；
各节点位置虚拟移动单元，具体包括：
虚拟力计算子单元，用于计算第i个节点与各未覆盖网格点之间的虚拟力；
合力计算子单元，用于计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和；
虚拟移动子单元，用于根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
根据本发明提供的具体实施例，本发明公开了以下技术效果：
本发明通过采用蜂窝网格堆砌方法对所述监测区域进行划分初步提升了网络的覆盖率，通过未覆盖网格点虚拟力对各节点位置进行进一步更新，进一步优化了覆盖性能，通过匈牙利算法求解效率矩阵，提升了网络模型中节点的剩余总能量以及剩余能量均衡度。
说明书附图
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例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性的前提下，还可以根据这些附图获得其他的附图。
图1为本发明一种面向异构无线传感器网络的覆盖增强方法流程示意图；
图2为本发明监测区域网格离散效果图；
图3为本发明蜂窝网格堆砌效果图；
图4为本发明异构WSNs覆盖增强问题二分图匹配模型示意图；
图5为本发明匈牙利算法流程图；
图6为本发明未覆盖网格点虚拟力算法流程图；
图7为本发明初始覆盖效果示意图；
图8为本发明一种面向异构无线传感器网络的覆盖增强方法覆盖效果示意图；
图9为COSH算法覆盖效果图；
图10为本发明一种面向异构无线传感器网络的覆盖增强方法节点移动轨迹图；
图11为COSH算法节点移动轨迹图；
图12为本发明一种面向异构无线传感器网络的覆盖增强方法和COSH算法剩余能量对比图；
图13为本发明蜂窝网格堆砌方法示意图；
图14为本发明一种面向异构无线传感器网络的覆盖增强系统结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的目的是提供一种面向异构无线传感器网络的覆盖增强方法及系统，提高了节点的剩余总能量以及剩余能量均衡度。
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合附图和具体实施方式对本发明作进一步详细的说明。
图1为本发明一种面向异构无线传感器网络的覆盖增强方法流程示意图，如图1所示，一种面向异构无线传感器网络的覆盖增强方法包括：
步骤101：将监测区域离散为M个网格点；
步骤102：初始化监测区域的网络模型，所述网络模型包括所述监测区域中节点的类型、数量、感知半径和位置(初始位置)；所述节点为无线传感器。
步骤103：基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌。
步骤104：构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵。
所述构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵，具体包括：
计算各所述节点到各所述蜂窝网格质心的距离。
根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量。
根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值。
根据各所述效率值构建效率矩阵。
所述效率矩阵表示为：
Figure PCTCN2021098084-appb-000003
其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
步骤105：采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置。
所述采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置，具体包括：
将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵。
将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵。
找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数。
若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉 点处的元素加上最小元素，实现对第二效率矩阵的更新，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”。
若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和每一列对应的零元素。
根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
步骤106：根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
所述根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置，具体包括：
根据当前各节点的虚拟位置获得未覆盖网格点坐标。
遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，迭代次数加1。
判断迭代次数是否达到预设迭代次数。
若未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”。
若达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置。
所述遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，具体包括：
计算第i个节点与各未覆盖网格点之间的虚拟力。
计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和。
根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
下面详细说明本发明一种面向异构无线传感器网络的覆盖增强方法。
一种面向异构无线传感器网络的覆盖增强方法包括以下步骤：
Step1：初始化网络模型及节点参数。
(1)初始化二维监测区域。将面积为L×B的监测区域均匀离散化为M个网格点，离散效果如图2所示。
M个网格点(网格质心)坐标如公式1所示。
Figure PCTCN2021098084-appb-000004
其中，Grid m,1表示第m个网格点x轴的坐标，Grid m,2表示第m个网格点y轴的坐标。
(2)初始化节点类型为布尔感知模型，总数量N，N个节点初始位置坐标如公式2所示。节点类型为K，不同类型节点数量分别为N 1,N 2,...N k，不同类型节点感知半径分别为R 1,R 2,...R k，不同类型节点初始能量分别为Ener 1,Ener 2,...Ener k，不同类型节点移动单位距离能耗分别为e 1,e 2,...e k。N个节点的初始位置用矩阵表示为Position N×2，Position i,1表示第i个节点的x轴坐标，Position i,2表示第i个节点的y轴坐标。
Figure PCTCN2021098084-appb-000005
Step2：使用蜂窝网格划分监测区域。
用N个外接圆半径为
Figure PCTCN2021098084-appb-000006
的蜂窝网格对面积为L×B的监测区域进行堆砌，堆砌效果如图3所示。
N个蜂窝网格的质心坐标如公式3所示。
Figure PCTCN2021098084-appb-000007
Step3：“三步法”构建节点与蜂窝网格质心点的效率矩阵并对异构WSNs覆盖增强问题抽象建模。
(1)计算各节点到各蜂窝网格质心的距离，如公式4所示。dis i,j为第i个节点到第j个蜂窝网格质心的距离。
Figure PCTCN2021098084-appb-000008
(2)计算各节点移动至各蜂窝网格后的剩余能量，如公式5所示，其中REner i,j表示节点i移动至第j个蜂窝网格质心后的剩余能量，Ener i代表节点i的初始能量，e i代表节点i移动单位距离的能耗。
REner i,j＝Ener i-e i×dis i,j        (公式5)
使用公式6计算各节点移动至各蜂窝网格的效率值。
Eff i,j＝ln(1+REner i,j)           (公式6)
由于对数函数的单调性，每个节点移动至每个蜂窝网格质心后的效率值与剩余能量值排序一致。但是剩余能量小的值将会大幅度减少，限制了后续使用匈牙利算法过程中对剩余能量较小的值进行任务指派，即在最小化移动总能耗的同时保证了剩余能量的均衡度。构建效率矩阵，如公式7所示。
Figure PCTCN2021098084-appb-000009
经过蜂窝网格划分，异构WSNs的覆盖增强问题转化为N个节点移动至N个蜂窝网格质心的任务指派问题，其二分图模型如图4所示。因此，异构WSNs的覆盖增强问题可抽象为最优任务指派问题的数学模型，如公式8所示。即：在效率矩阵Eff N×N 中寻找N个位于不同行不同列的元素，使得N个元素的值的和F最大。
Figure PCTCN2021098084-appb-000010
其中，x i,j表示第i个节点是否选择第j个蜂窝网格，若选择，则取值为1，不选择则取值为0。
Step4：使用匈牙利算法求解效率矩阵。
匈牙利算法适用于求解二分图最小匹配问题，而节点重部署中的剩余能量最大化属于最大匹配问题，因此转换效率矩阵为Eff N×N＝-Eff N×N。
匈牙利算法基本思想是修正效率矩阵中的每一行或列，直至不同行不同列都至少有一个零元素，从而得到与这些零元素相对应的匹配方案。匈牙利算法的流程图如图5所示。
具体步骤如下:
(1)效率矩阵进行行减和列减。具体而言，效率矩阵每行减去本行的最小值；在新的矩阵中每列减去本列最小值。
(2)画线，即指用最少的行线和列线将经过步骤(1)得到的新的效率矩阵中的0全部穿起来(即当前矩阵中的0都被最少的线所覆盖)。找出能够覆盖效率矩阵中所有零元素的最少直线数P。
(3)若最小直线数P＝N，则转步骤(5)；否则转步骤(4)。
(4)在效率矩阵未被直线覆盖的区域中找出最小元素，令此区域中所有元素减去最小元素，并令覆盖直线交叉点处的元素加上最小元素。得到新的效率矩阵并画线，转步骤(3)。
(5)找出每一行对应的零元素和每一列对应的零元素，根据零元素找到最优分配，确定基于匈牙利算法后节点的位置。
Step5：使用未覆盖网格点虚拟力进一步确定节点位置。
未覆盖网格点虚拟力算法流程图如图6所示。
具体步骤如下:
(1)计算当前迭代轮次覆盖率并统计未覆盖网格点数量。
(2)遍历各节点计算其与各未覆盖网格点之间的虚拟力。
具体而言，节点i与未覆盖网格点j之间的虚拟力计算方法如公式9所示。
Figure PCTCN2021098084-appb-000011
其中，d ij代表节点i与未覆盖网格点j之间的欧式距离，θ ij为节点i指向未覆盖网格点j的方向角，d th为二者之间的欧氏距离。
(3)计算节点受到的合力并进行位置更新。
节点i受到未覆盖网格点的合力的计算方法如公式10所示。
Figure PCTCN2021098084-appb-000012
节点i在合力的作用下按照公式11进行位置更新。
Figure PCTCN2021098084-appb-000013
x i与y i分别为节点i的横、纵坐标，F xi与F yi为节点受到的合力F i在横、纵坐标上的分量，Max_Step为节点单次虚拟移动步长。
(4)判断N个节点是否已经遍历完毕。
若N个节点已经遍历完毕，则进行步骤(5)。
若N个节点未遍历完毕，则进行步骤(2)。
(5)判断是否满足虚拟力最大迭代次数。
若满足虚拟力迭代次数，则转步骤(6)。
若不满足虚拟力迭代次数，则转步骤(1)。
(6)确定节点最终位置。
Step6：节点进行二次部署。
根据各节点的最终位置对节点进行二次部署。
节点从初始位置移动至步骤五得出的节点最终位置处，并计算移动后节点移动总能耗与剩余能量均衡度。
在本发明确定各节点最终位置的过程中，无论是匈牙利算法分配后的节点位置还是经过后续虚拟力算法节点每轮移动的位置，都是虚拟位置(只需要在算法中实现，不需要节点真实的移动)。节点只移动一次，即通过算法计算出了最终位置，然后将节点从初始位置移动到最终位置。
各名词解释说明如下：
(1)网络覆盖率：任意网格点m被第i个节点感知概率可通过公式12计算。d im表示第i个节点与第m个网格点的欧氏距离。
Figure PCTCN2021098084-appb-000014
若网格点m被任意节点感知，则称此网格点被覆盖，未被任一节点感知的网格点称为未覆盖网格点。整个网络的覆盖率如公式13所示。
Figure PCTCN2021098084-appb-000015
其中，p m为网格点m的覆盖率，M为离散化后的网格点数量。
(2)移动总能耗
重部署后节点i的移动能耗如公式14所示。
Emov i＝e i×d i        (公式14)
e i为节点i移动单位距离的能耗，d i为节点初始位置到最终位置的欧氏距离。整个网络的移动总能耗如公式15所示，N为网络中节点数量。
Figure PCTCN2021098084-appb-000016
(3)剩余能量均衡度
重部署后节点i的剩余能量如公式16所示：
Eres i＝Ener i-Emov i           (公式16)
整个网络的剩余能量均衡度如公式17所示：
Figure PCTCN2021098084-appb-000017
(4)布尔感知模型
布尔感知模型，即指节点对监测区域内的目标点的监测概率是恒定的。节点i的感知范围是以节点坐标为圆心，半径为R i的圆形区域，R i为节点i的感知半径。对于监测区域内任意一目标点j，节点i可成功监测到目标点j的概率如公式12所示。
Figure PCTCN2021098084-appb-000018
其中，d ij为节点i与目标点j之间的欧氏距离。
(5)蜂窝网格堆砌理论
对监测区域进行网格划分可以有效提高节点部署的均匀性，正三角形、正四边形、正五边形以及正六边形均可进行监测区域的堆砌，而蜂窝网格(正六边形)拥有最高的堆砌效率，可以有效地减少覆盖冗余。蜂窝网格堆砌方案如图13所示。
本发明提供一种面向异构无线传感器网络的覆盖增强方法，该方法有以下优点：
(1)本发明提出了一种面对异构WSNs的蜂窝网格划分方法，以不同类型节点感知半径的加权平均值作为蜂窝网格的外接圆半径，保证了重部署后节点分布的均匀性，初步提升了网络的覆盖率。步骤五针对蜂窝网格划分导致的覆盖率存在阈值这一缺陷，提出了未覆盖网格点虚拟力算法进一步优化了覆盖性能。
与COSH算法相比，覆盖率得到了明显提升。两种方案的覆盖效果如图7-9所示。
(2)通过“三步法”构建效率矩阵并通过匈牙利算法求解的思路，与COSH算法相比，可显著提升网络中节点的剩余总能量以及剩余能量均衡度，根据图7-9的显示了各种情况的覆盖效果，各节点的移动轨迹如图10-11所示，各无线传感器节点的剩余能量对比如图12所示。
图14为一种面向异构无线传感器网络的覆盖增强系统结构示意图，如图14所示，本发明还公开了一种面向异构无线传感器网络的覆盖增强系统，包括：
监测区域离散模块201，用于将监测区域离散为M个网格点。
初始化模块202，用于初始化监测区域的网络模型，所述网络模型包括所述监测 区域中节点的类型、数量、感知半径和位置；所述节点为无线传感器；所述节点为无线传感器。
蜂窝网格堆砌模块203，用于基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌。
效率矩阵构建模块204，用于构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵。
重新分配模块205，用于采用匈牙利算法求解所述效率矩阵，获得重新分配后的各节点的虚拟位置。
最终位置确定模块206，用于根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
所述效率矩阵构建模块204，具体包括：
距离计算单元，用于计算各所述节点到各所述蜂窝网格质心的距离；
剩余能量计算单元，用于根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量；
效率值计算单元，用于根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值；
效率矩阵构建单元，用于根据各所述效率值构建效率矩阵。
所述效率矩阵表示为：
Figure PCTCN2021098084-appb-000019
其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
所述重新分配模块205，具体包括：
第一效率矩阵获得单元，用于将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵；
第二效率矩阵获得单元，用于将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵；
最小直线数获得单元，用于找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数；
第二效率矩阵更新单元，用于若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉点处的元素加上最小元素，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”；
零元素获得单元，用于若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和每一列对应的零元素；
重新分配单元，用于根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
所述最终位置确定模块206，具体包括：
未覆盖网格点获得单元，用于根据当前各节点位置的获得未覆盖网格点坐标；
各节点位置虚拟移动单元，用于遍历各所述节点，通过虚拟移动各所述节点，对 各节点的虚拟位置进行更新，迭代次数加1；
返回单元，用于若迭代次数未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”；
最终位置确定单元，用于若迭代次数达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置；
各节点位置虚拟移动单元，具体包括：
虚拟力计算子单元，用于计算第i个节点与各未覆盖网格点之间的虚拟力；
合力计算子单元，用于计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和；
虚拟移动子单元，用于根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
本说明书中各个实施例采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似部分互相参见即可。
本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处。综上所述，本说明书内容不应理解为对本发明的限制。

Claims (10)

  1. 一种面向异构无线传感器网络的覆盖增强方法，其特征在于，包括：
    将监测区域离散为M个网格点；
    初始化监测区域的网络模型，所述网络模型包括所述监测区域中节点的类型、数量、感知半径和位置；所述节点为无线传感器；
    基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌；
    构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵；
    采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置；
    根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
  2. 根据权利要求1所述的面向异构无线传感器网络的覆盖增强方法，其特征在于，所述构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵，具体包括：
    计算各所述节点到各所述蜂窝网格质心的距离；
    根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量；
    根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值；
    根据各所述效率值构建效率矩阵。
  3. 根据权利要求2所述的面向异构无线传感器网络的覆盖增强方法，其特征在于，所述效率矩阵表示为：
    Figure PCTCN2021098084-appb-100001
    其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
  4. 根据权利要求1所述的面向异构无线传感器网络的覆盖增强方法，其特征在于，所述采用匈牙利算法求解所述效率矩阵，获得重新分配后各节点的虚拟位置，具体包括：
    将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵；
    将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵；
    找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数；
    若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉点处的元素加上最小元素，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”；
    若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和 每一列对应的零元素；
    根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
  5. 根据权利要求1所述的面向异构无线传感器网络的覆盖增强方法，其特征在于，所述根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置，具体包括：
    根据当前各节点的虚拟位置获得未覆盖网格点坐标；
    遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，迭代次数加1；
    判断迭代次数是否达到预设迭代次数；
    若未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”；
    若达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置；
    所述遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，具体包括：
    计算第i个节点与各未覆盖网格点之间的虚拟力；
    计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和；
    根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
  6. 一种面向异构无线传感器网络的覆盖增强系统，其特征在于，包括：
    监测区域离散模块，用于将监测区域离散为M个网格点；
    初始化模块，用于初始化监测区域的网络模型，所述网络模型包括所述监测区域中节点的类型、数量、感知半径和位置；所述节点为无线传感器；
    蜂窝网格堆砌模块，用于基于所述网络模型，采用蜂窝网格堆砌方法对所述监测区域进行划分，获得蜂窝网格堆砌；
    效率矩阵构建模块，用于构建各所述节点与所述蜂窝网格堆砌中各蜂窝网格质心的效率矩阵；
    重新分配模块，用于采用匈牙利算法求解所述效率矩阵，获得重新分配后的各节点的虚拟位置；
    最终位置确定模块，用于根据各节点当前的虚拟位置确定未覆盖网格点，利用各未覆盖网格点与各所述节点的虚拟力确定各节点的最终位置。
  7. 根据权利要求6所述的面向异构无线传感器网络的覆盖增强系统，其特征在于，所述效率矩阵构建模块，具体包括：
    距离计算单元，用于计算各所述节点到各所述蜂窝网格质心的距离；
    剩余能量计算单元，用于根据各所述节点到各所述蜂窝网格质心的距离计算各所述节点移动至各所述蜂窝网格后的剩余能量；
    效率值计算单元，用于根据各所述节点移动至各所述蜂窝网格后的剩余能量计算各所述节点移动至各所述蜂窝网格的效率值；
    效率矩阵构建单元，用于根据各所述效率值构建效率矩阵。
  8. 根据权利要求7所述的面向异构无线传感器网络的覆盖增强系统，其特征在于，所述效率矩阵表示为：
    Figure PCTCN2021098084-appb-100002
    其中，Eff N×N表示效率矩阵，Eff i,j表示第i个节点移动至第j个蜂窝网格的效率值，i∈[1,N]，j∈[1,N]，N表示节点个数。
  9. 根据权利要求6所述的面向异构无线传感器网络的覆盖增强系统，其特征在于，所述重新分配模块，具体包括：
    第一效率矩阵获得单元，用于将所述效率矩阵中每行中各元素均减去本行中的最小值，得到第一效率矩阵；
    第二效率矩阵获得单元，用于将所述第一效率矩阵中每列中各元素均减去本列中的最小值，得到第二效率矩阵；
    最小直线数获得单元，用于找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数；
    第二效率矩阵更新单元，用于若最小直线数不等于节点数量，则在所述第二效率矩阵中未被直线覆盖的区域中找出最小元素，将所述未被直线覆盖的区域中所有元素减去最小元素，并将直线交叉点处的元素加上最小元素，返回步骤“找出能够覆盖所述第二效率矩阵中所有零元素的最小直线数”；
    零元素获得单元，用于若最小直线数等于节点数量，则找出所述第二效率矩阵中每一行对应的零元素和每一列对应的零元素；
    重新分配单元，用于根据各所述零元素确定各节点的最优分配，获得重新分配后的各节点的虚拟位置。
  10. 根据权利要求6所述的面向异构无线传感器网络的覆盖增强系统，其特征在于，所述最终位置确定模块，具体包括：
    未覆盖网格点获得单元，用于根据当前各节点位置的获得未覆盖网格点坐标；
    各节点位置虚拟移动单元，用于遍历各所述节点，通过虚拟移动各所述节点，对各节点的虚拟位置进行更新，迭代次数加1；
    返回单元，用于若迭代次数未达到预设迭代次数，则返回“根据当前各节点的虚拟位置获得未覆盖网格点坐标”；
    最终位置确定单元，用于若迭代次数达到预设迭代次数，则将当前各节点的虚拟位置确定为各节点的最终位置；
    各节点位置虚拟移动单元，具体包括：
    虚拟力计算子单元，用于计算第i个节点与各未覆盖网格点之间的虚拟力；
    合力计算子单元，用于计算第i个节点受到所有未覆盖网格点的合力，所述合力为各未覆盖网格点对第i个节点的虚拟力之和；
    虚拟移动子单元，用于根据第i个节点受到的合力虚拟移动第i个节点，获得更新后各节点的虚拟位置。
PCT/CN2021/098084 2021-04-20 2021-06-03 一种面向异构无线传感器网络的覆盖增强方法及系统 WO2022222236A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110425315.0A CN113242565A (zh) 2021-04-20 2021-04-20 一种面向异构无线传感器网络的覆盖增强方法及系统
CN202110425315.0 2021-04-20

Publications (1)

Publication Number Publication Date
WO2022222236A1 true WO2022222236A1 (zh) 2022-10-27

Family

ID=77128501

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/098084 WO2022222236A1 (zh) 2021-04-20 2021-06-03 一种面向异构无线传感器网络的覆盖增强方法及系统

Country Status (2)

Country Link
CN (1) CN113242565A (zh)
WO (1) WO2022222236A1 (zh)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101383736A (zh) * 2008-10-15 2009-03-11 中国科学院上海微系统与信息技术研究所 一种面向区域监测的无线传感器网络节点布设的优化方法
CN102740312A (zh) * 2012-07-12 2012-10-17 北京邮电大学 无线传感器网络的覆盖控制方法
CN103354642A (zh) * 2013-06-06 2013-10-16 东北大学 一种提高移动传感器网络覆盖率的方法
CN103997747A (zh) * 2014-05-14 2014-08-20 浪潮电子信息产业股份有限公司 一种基于虚拟力的空间网络节点均匀部署方法
CN105933915A (zh) * 2016-05-31 2016-09-07 昆明理工大学 一种基于虚拟势场的有向异构无线传感器网络覆盖优化方法
CN107071853A (zh) * 2017-03-23 2017-08-18 安徽师范大学 一种对事件进行监测的无线移动传感网络的构建方法
CN108271242A (zh) * 2017-12-14 2018-07-10 南京邮电大学 基于能量效率的d2d资源分配方法
CN110139286A (zh) * 2019-05-21 2019-08-16 西安邮电大学 面向三维环境的无线传感器网络覆盖增强方法及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102724681B (zh) * 2012-06-27 2014-09-03 东北大学 一种结合能量有效性的传感器网络覆盖空洞检测方法
CN106211190B (zh) * 2016-06-30 2018-02-06 广东工业大学 一种非规则区域的无线传感器网络节点部署方法
CN109640333B (zh) * 2018-12-13 2021-08-31 沈阳理工大学 基于集群划分的水下无线传感器网络覆盖漏洞修复算法
CN110401958B (zh) * 2019-08-05 2022-02-01 重庆邮电大学 一种基于虚拟力的节点动态覆盖增强方法
CN112637860B (zh) * 2020-12-21 2022-08-02 西安邮电大学 一种三维无线传感器网络覆盖方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101383736A (zh) * 2008-10-15 2009-03-11 中国科学院上海微系统与信息技术研究所 一种面向区域监测的无线传感器网络节点布设的优化方法
CN102740312A (zh) * 2012-07-12 2012-10-17 北京邮电大学 无线传感器网络的覆盖控制方法
CN103354642A (zh) * 2013-06-06 2013-10-16 东北大学 一种提高移动传感器网络覆盖率的方法
CN103997747A (zh) * 2014-05-14 2014-08-20 浪潮电子信息产业股份有限公司 一种基于虚拟力的空间网络节点均匀部署方法
CN105933915A (zh) * 2016-05-31 2016-09-07 昆明理工大学 一种基于虚拟势场的有向异构无线传感器网络覆盖优化方法
CN107071853A (zh) * 2017-03-23 2017-08-18 安徽师范大学 一种对事件进行监测的无线移动传感网络的构建方法
CN108271242A (zh) * 2017-12-14 2018-07-10 南京邮电大学 基于能量效率的d2d资源分配方法
CN110139286A (zh) * 2019-05-21 2019-08-16 西安邮电大学 面向三维环境的无线传感器网络覆盖增强方法及系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
DESIKAN K E SRINIVASA; KOTAGI VIJETH J; MURTHY C SIVA RAM: "Smart at Right Price: A Cost Efficient Topology Construction for Fog Computing Enabled IoT Networks in Smart Cities", 2018 IEEE 29TH ANNUAL INTERNATIONAL SYMPOSIUM ON PERSONAL, INDOOR AND MOBILE RADIO COMMUNICATIONS (PIMRC), IEEE, 9 September 2018 (2018-09-09), pages 1 - 7, XP033479565, DOI: 10.1109/PIMRC.2018.8580922 *
SONG ZHI-QIANG, FANG WU;LU AI-HONG: "Ｒesearch on Hybrid Sensor Nodes Coverage Deployment Based on Virtual Force", INSTRUMENT TECHNIQUE AND SENSOR, SHENYANG YIQI YIBIAO GONGYI YANJIUSUO, CHINA, no. 9, 15 September 2017 (2017-09-15), China , XP055978007, ISSN: 1002-1841 *

Also Published As

Publication number Publication date
CN113242565A (zh) 2021-08-10

Similar Documents

Publication Publication Date Title
CN104050390B (zh) 一种基于可变维粒子群膜算法的移动机器人路径规划方法
CN106647744B (zh) 一种机器人路径规划方法及装置
CN110401958B (zh) 一种基于虚拟力的节点动态覆盖增强方法
Li et al. Path planning of mobile robot based on genetic algorithm and gene rearrangement
CN113595619B (zh) 一种无人机群通联与覆盖组合优化方法
CN111474925A (zh) 一种非规则形状移动机器人的路径规划方法
CN113392539A (zh) 基于联邦强化学习的机器人通信控制方法、系统及设备
CN113115342A (zh) 一种虚拟力导向麻雀搜索算法的WSNs部署方法及系统
You et al. An improved ant colony system algorithm for robot path planning and performance analysis
CN110351735B (zh) 一种基于贪心算法的无线可充电传感网基站部署方法
WO2022222236A1 (zh) 一种面向异构无线传感器网络的覆盖增强方法及系统
Huang Multi-node topology location model of smart city based on Internet of Things
CN116709290A (zh) 一种基于无人机边缘计算的灾害地区应急通信方法及系统
Rana et al. Mobile ad-hoc clustering using inclusive particle swarm optimization algorithm
Pei et al. UAV swarm for connectivity enhancement of multiple isolated sensor networks for Internet of things application
CN110602718B (zh) 基于交替方向乘子法的异构蜂窝网络功率分配方法及系统
CN105072622B (zh) 一种二维入侵监测区域的可变k覆盖优化方法
CN109540022B (zh) 基于tof深度相机的平仓机器人路径规划与决策方法
Batsoyol et al. Constructing full-coverage 3D UAV ad-hoc networks through collaborative exploration in unknown urban environments
CN115297487A (zh) 一种异构无线传感网初始化部署方法
CN108614889A (zh) 基于混合高斯模型的移动对象连续k近邻查询方法及系统
CN114879726A (zh) 一种基于多无人机辅助数据收集的路径规划方法
CN115545106A (zh) 一种多无人机中的AoI敏感的数据收集方法及系统
CN114740849A (zh) 基于行人步行决策规则的移动机器人自主导航方法及装置
Wang Optimization of wireless network node deployment in smart city based on adaptive particle swarm optimization

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: 21937461

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: 21937461

Country of ref document: EP

Kind code of ref document: A1