[go: up one dir, main page]

CN115334612B - A wireless network topology inference method based on spatiotemporal features - Google Patents

A wireless network topology inference method based on spatiotemporal features Download PDF

Info

Publication number
CN115334612B
CN115334612B CN202210762307.XA CN202210762307A CN115334612B CN 115334612 B CN115334612 B CN 115334612B CN 202210762307 A CN202210762307 A CN 202210762307A CN 115334612 B CN115334612 B CN 115334612B
Authority
CN
China
Prior art keywords
node
cooperative
training
cooperative node
sensor
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN202210762307.XA
Other languages
Chinese (zh)
Other versions
CN115334612A (en
Inventor
陈睿
常莉莉
李浩铸
胡晓鹏
肖潇
杨俊�
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Infohand Technology Co ltd
Guangzhou Institute of Technology of Xidian University
Original Assignee
Guangzhou Infohand Technology Co ltd
Guangzhou Institute of Technology of Xidian University
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 Guangzhou Infohand Technology Co ltd, Guangzhou Institute of Technology of Xidian University filed Critical Guangzhou Infohand Technology Co ltd
Priority to CN202210762307.XA priority Critical patent/CN115334612B/en
Publication of CN115334612A publication Critical patent/CN115334612A/en
Application granted granted Critical
Publication of CN115334612B publication Critical patent/CN115334612B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning 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
    • H04L41/12—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
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

本发明涉及一种基于时空特征的无线网络拓扑推断方法，包括：在目标区域设置传感器网络；侦测非协作网络的功率；将目标区域划分成M个预设网格；将预设网格的功率输入至神经网络模型得到非协作节点的坐标；根据非协作节点的坐标得到两个非协作节点间的距离；根据非协作节点间交互的时间信息，得到各个非协作节点的发送信息的开始时间和结束时间；以得到各节点对的帧间隔的均值和帧间隔的方差；将节点对的帧间隔的均值、帧间隔的方差、距离信息输入训练好的分类模型中，得到各节点对的连通关系。本发明基于时空特征的无线网络拓扑推断方法将节点定位和拓扑推断结合为一个系统实现目标节点位置未知时的拓扑推断工作。

The present invention relates to a wireless network topology inference method based on spatiotemporal features, comprising: setting a sensor network in a target area; detecting the power of a non-cooperative network; dividing the target area into M preset grids; inputting the power of the preset grids into a neural network model to obtain the coordinates of non-cooperative nodes; obtaining the distance between two non-cooperative nodes according to the coordinates of the non-cooperative nodes; obtaining the start time and end time of sending information of each non-cooperative node according to the time information of the interaction between the non-cooperative nodes; obtaining the mean of the frame interval of each node pair and the variance of the frame interval; inputting the mean of the frame interval of the node pair, the variance of the frame interval, and the distance information into a trained classification model to obtain the connectivity relationship of each node pair. The wireless network topology inference method based on spatiotemporal features of the present invention combines node positioning and topology inference into a system to realize topology inference work when the position of the target node is unknown.

Description

Wireless network topology inference method based on space-time characteristics
Technical Field
The invention belongs to the technical field of positioning and topology inference, and relates to a wireless network topology inference method based on space-time characteristics.
Background
With the rapid development of information technology, in case of the entering of the interconnected era, the network scale is increasingly large, which makes network management difficult, so intelligent analysis of network behavior is increasingly important.
To understand the structure of a communication network, one promising technique is to identify the structure of the network by sensing and inference. The topology inference can help to identify the flow of the network, grasp the link degree of the user, optimize and improve the security of the network. For non-cooperative wireless networks, topology inference can mine information, obtaining information advantages.
However, in most of the existing topology estimation algorithms, topology is estimated based on the situation that the node position is known, and topology estimation unknown to the node becomes a problem to be solved.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a wireless network topology inference method based on space-time characteristics. The technical problems to be solved by the invention are realized by the following technical scheme:
The embodiment of the invention provides a wireless network topology inference method based on space-time characteristics, which comprises the following steps:
step 1, setting a sensor network in a target area, wherein the sensor network comprises M sensor nodes which are randomly distributed;
step 2, detecting the power of a non-cooperative network in the target area by using the sensor network, wherein the non-cooperative network comprises N non-cooperative nodes;
Step 3, dividing the target area into M preset grids with the same number as the sensor nodes, and determining the power of each preset grid by using the power detected by the sensor network;
step 4, inputting the power of the preset grid obtained in the step 3 into a trained neural network model to obtain the coordinate information of the non-cooperative node;
Step 5, obtaining the distance information between the two non-cooperative nodes according to the coordinate information of the non-cooperative nodes determined in the step 4;
step 6, obtaining the starting time and the ending time of the sending information of each non-cooperative node according to the time information of the interaction between the non-cooperative nodes in the target area detected by the sensor network;
Step 7, obtaining the mean value of the frame interval and the variance of the frame interval of each node pair according to the starting time and the ending time of the sending information of the non-cooperative nodes;
And 8, inputting the mean value of the frame interval, the variance of the frame interval and the distance information of the node pairs into a trained classification model to obtain the communication relation of each node pair.
In one embodiment of the present invention, the power detected by the ith sensor node is:
Wherein p i,j represents the power of the ith sensor node detecting the jth non-cooperative node, p i,j＝pjGtGr(λ/4πRi,j)2, wherein 1.ltoreq.i.ltoreq.M, 1.ltoreq.j.ltoreq.N, G t represents the antenna gain of the transmitting antenna, G r represents the antenna gain of the receiving antenna, λ represents the wavelength, and R i,j represents the distance between the ith sensor node detecting the jth non-cooperative node.
In one embodiment of the present invention, the power of the mth preset grid is:
Wherein, And B represents the number of sensors contained in the mth preset grid, wherein M is more than or equal to 1 and less than or equal to M.
In one embodiment of the present invention, the training method of the neural network model includes:
S1, acquiring a training set, wherein the training set comprises a plurality of non-cooperative training nodes;
S2, dividing the area where the non-cooperative training nodes are located into H training grids with the same size, wherein H is greater than M;
s3, calculating a distance R i,j' between the ith sensor training node and the jth non-cooperative training node;
S4, obtaining training power from the ith sensor training node to the jth non-cooperative training node according to the distance R i,j';
S5, inputting a plurality of input-output pairs (p (i),x(i)) into the neural network model to train the neural network model to obtain a trained neural network model, wherein p (i) is training power of the non-cooperative training nodes, and x (i) is a training grid where the real positions of the non-cooperative training nodes are located.
In one embodiment of the present invention, step 4 comprises:
And 4.1, inputting the power of each preset grid into the trained neural network model to obtain a positioning result of the non-cooperative node so as to determine the preset grid where the non-cooperative node is located.
And 4.2, taking the central position of the preset grid where the non-cooperative node is located as the coordinate position of the non-cooperative node.
In one embodiment of the present invention, the distance information between two non-cooperative nodes is:
Wherein, For the coordinates of the p-th non-cooperative node,For the coordinates of the q-th non-cooperative node, d p,q represents the distance between the p-th non-cooperative node and the q-th non-cooperative node.
In one embodiment of the present invention, the sensor network detects that the data acquired by the p-th non-cooperative node is:
where n represents the number of signals contained in F p, T represents the transpose of the matrix, Represents a set of real numbers,Representing n rows and 2 columns of real matrices;
The ith signal of the p-th non-cooperative node is:
Wherein, Indicating a start time of the ith signaling of the p-th non-cooperative node,Representing the duration of the ith signal of the p-th non-cooperative node.
In one embodiment of the present invention, the average value of the frame intervals is:
Wherein, Representing the average value of the frame intervals of the p-th non-cooperative node and the q-th non-cooperative node,Representing the frame interval of the p-th non-cooperative node and the q-th non-cooperative node,Indicating the start time of the ith signaling of the qth non-cooperative node.
In one embodiment of the present invention, the variance of the frame interval is:
Wherein, Representing the variance of the frame interval of the p-th non-cooperative node and the q-th non-cooperative node.
In one embodiment of the present invention, the connectivity of each node pair is represented by an adjacency matrix a, which is:
A=[aij]
the link connection situation a ij of the non-cooperative node v i and the non-cooperative node v j is as follows:
Wherein, 1 indicates that the non-cooperative node v i has a communication relationship with the non-cooperative node v j, and 0 indicates that the non-cooperative node v i does not have a communication relationship with the non-cooperative node v j.
Compared with the prior art, the invention has the beneficial effects that:
the wireless network topology inference method based on space-time characteristics combines node positioning and topology inference into a system to realize topology inference work when the position of a target node is unknown.
The positioning method adopts a machine learning method to realize the positioning of the target node.
The topology inference method of the invention provides space-time characteristics (mean value, variance of frame interval and distance between node pairs) as the basis for judging whether the communication relationship exists between the nodes, and adopts a machine learning method to rapidly obtain the topology inference result.
Other aspects and features of the present invention will become apparent from the following detailed description, which refers to the accompanying drawings. It is to be understood that the drawings are designed solely for purposes of illustration and not as a definition of the limits of the invention, for which reference should be made to the appended claims. It should be further understood that the drawings are not necessarily drawn to scale and that, unless otherwise indicated, they are merely intended to conceptually illustrate the structures and procedures described herein.
Drawings
Fig. 1 is a schematic flow chart of a wireless network topology inference method based on space-time features according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a target topology with a sensor network according to an embodiment of the present invention;
FIG. 3 is a system operation block diagram of a wireless network topology inference method based on space-time features according to an embodiment of the present invention;
fig. 4 is a space-time feature diagram of a wireless network topology inference method based on space-time features according to an embodiment of the present invention;
FIG. 5 is a CDF error accumulation map for target node positioning according to an embodiment of the present invention;
Fig. 6 is a schematic diagram of inference accuracy of a topology inference algorithm according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to specific examples, but embodiments of the present invention are not limited thereto.
Example 1
Referring to fig. 1, fig. 1 is a schematic flow chart of a wireless network topology inference method based on space-time features according to an embodiment of the present invention, and the present invention provides a wireless network topology inference method based on space-time features, where the wireless network topology inference method includes steps 1 to 8, where:
step 1, setting a sensor network in a target area, wherein the sensor network comprises M sensor nodes which are distributed randomly, and the positions of the sensor nodes are known.
And detecting the power of a non-cooperative network in the target area L by using a sensor network formed by M sensor nodes distributed randomly in the target area L, wherein the non-cooperative network comprises N non-cooperative nodes.
In a specific embodiment, step 1 may specifically include steps 1.1-1.2, wherein:
Step 1.1, randomly setting M sensor nodes in a target area L, wherein the positions of the sensor nodes are known
Specifically, referring to fig. 2, a schematic diagram of a non-cooperative network including sensor nodes is provided in an embodiment of the present invention, where the non-cooperative network is composed of N non-cooperative nodes, the sensor network is composed of M sensor nodes and a sink node (the sink node is used to collect data sent by all other nodes and perform related processing, etc.), the M sensor nodes are scattered randomly in a target area, and coordinates of the sensor nodes areT represents the transpose of the matrix,Represents a set of real numbers,Representing a real matrix of M rows and 2 columns, the coordinates of the ith sensor are
And 2, detecting the power of a non-cooperative network in the target area by using a sensor network, wherein the non-cooperative network comprises N non-cooperative nodes.
Specifically, according to the energy distribution condition of the M sensor nodes in the detection target area, each sensor node obtains the received power at the position of each sensor node.
In this embodiment, the sensor detects the power of the non-cooperative node to obtain p= [ p 1,p2,…,pi,…,pM ], and the power detected by the ith sensor node is:
Wherein p i,j represents the power of the ith sensor node detecting the jth non-cooperative node, p i,j＝pjGtGr(λ/4πRi,j)2, wherein 1.ltoreq.i.ltoreq.M, 1.ltoreq.j.ltoreq.N, N represents the total number of non-cooperative nodes in the non-cooperative network, G t represents the antenna gain of the transmitting antenna, G r represents the antenna gain of the receiving antenna, λ represents the wavelength, and R i,j represents the distance between the ith sensor node detecting the jth non-cooperative node.
And 3, dividing the target area into M preset grids with the same number as the sensor nodes, and determining the power of each preset grid by using the power detected by the sensor network.
Specifically, the target area is divided into M equal-sized preset grids according to the number of sensors, and then the power data of each preset grid is obtained according to the power data obtained by each sensor and the known sensor position information.
In this embodiment, the power distribution of M preset grids is obtained by obtaining the power data detected by each sensorWherein the power of the mth preset grid is expressed as: b represents the number of sensors contained in the M < th > preset grid, wherein M is more than or equal to 1 and less than or equal to M.
And 4, inputting the power of the preset grid obtained in the step 3 into a trained neural network model to obtain the coordinate information of the non-cooperative node.
In this embodiment, the training method of the neural network model includes:
s4.1, acquiring a training set, wherein the training set comprises a plurality of non-cooperative training nodes, and the non-cooperative training nodes are non-cooperative nodes for training.
Specifically, the training set is composed of data that can monitor a portion of non-cooperative nodes and can derive their locations.
S4.2, dividing the area where the non-cooperative training nodes are located into H training grids with the same size, wherein H > M.
S4.3, calculating the distance R i,j' between the ith sensor training node and the jth non-cooperative training node.
Specifically, the training set is composed of data from which part of non-cooperative nodes can be monitored and from which the positions thereof can be obtained, assuming that the coordinates of the j-th non-cooperative training node are z= (x j,yj), and then the distance between each non-cooperative training node and each sensor training node (i.e., the sensor node used for training) is calculated, and the distance R i,j' between the i-th sensor training node and the j-th non-cooperative training node is:
Wherein, The coordinates of the node are trained for the ith sensor.
And S4.4, obtaining the training power from the ith sensor training node to the jth non-cooperative training node according to the distance R i,j'.
Specifically, the distance R i,j' is introduced into the power formula of the step 2, so as to obtain the received power from the ith sensor training node to the jth uncooperative training node.
S4.5, inputting a plurality of input-output pairs (p (i),x(i)) into the neural network model to train the neural network model to obtain a trained neural network model, wherein p (i) is the training power of the non-cooperative training node, and x (i) is the training grid where the real position of the non-cooperative training node is located.
Specifically, a combination of non-cooperative training nodes is created, and assuming that there are n non-cooperative training nodes, the total number of combinations of non-cooperative training nodes that may occur in each training grid is:
The training set { (p, x) } = { (p (1),x(1)),(p(2),x(2)),…,(p(n),x(n)) } is input into the neural network model, where (p (i),x(i)) is the input-output pair, In order to obtain the power data, x (i)∈[0,1]H is a training grid where the real position of the non-cooperative node is located, and the weight parameters of the neural network model are trained to obtain a trained neural network model.
In a specific embodiment, step 4 may specifically include steps 4.1-4.2, wherein:
And 4.1, inputting the power of each preset grid into the trained neural network model to obtain a positioning result of the non-cooperative node so as to determine the preset grid where the non-cooperative node is located.
And 4.2, taking the central position of the preset grid where the non-cooperative node is located as the coordinate position of the non-cooperative node.
Specifically, when the sensor network detects the target area, power data p= [ p 1,p2,…,pi,…,pM ] of a scene to be detected is obtained, the data is used as input of a trained neural network model, the neural network model outputs a positioning result of the non-cooperative node, namely the non-cooperative node is positioned at a certain preset grid, and then the central position of the preset grid is positionedAs coordinates of a non-cooperative node, where r i＝(xi,yi) represents coordinates of an i-th node.
And 5, obtaining the distance information between the two non-cooperative nodes according to the coordinate information of the non-cooperative nodes determined in the step 4.
Specifically, according to the positioning result of the target node (the target node is a non-cooperative node)Distance information between each non-cooperative node can be obtained, taking the p-th non-cooperative node and the q-th non-cooperative node as examples, the distance between the two non-cooperative nodes is as follows:
Wherein, For the coordinates of the p-th non-cooperative node,For the coordinates of the q-th non-cooperative node, d p,q represents the distance between the p-th non-cooperative node and the q-th non-cooperative node.
And 6, obtaining the starting time and the ending time of the sending information of each non-cooperative node according to the time information of interaction between the non-cooperative nodes in the target area detected by the sensor network.
Specifically, after the position of the target node is obtained in step 4, the radio station fingerprint recognition means are applied in combination with the information such as the frequency and the interception time of the signal, so that the communication signals between the target nodes can be distinguished, and the detected signal can be determined to be sent by which non-cooperative node. The detecting non-cooperative node may effectively detect the start time and signal duration of the signal transmitted by the target node within the given area. The data detected by all the non-cooperative nodes is represented as f= { F 1,F2,…,Fp,…,FN }, and the data acquired by the p-th non-cooperative node is represented as:
Where n represents the number of signals contained in F p.
The ith signal of the p-th non-cooperative node is:
Wherein, Indicating a start time of the ith signaling of the p-th non-cooperative node,Representing the duration of the ith signal of the p-th non-cooperative node.
And 7, obtaining the average value of the frame interval and the variance of the frame interval of each node pair according to the starting time and the ending time of the sending information of the non-cooperative nodes, wherein the average value, the variance and the distance of the node pair are used as the characteristics for judging whether the node pairs are communicated or not, and the node pair consists of two non-cooperative nodes.
Specifically, taking the p-th non-cooperative node and the q-th non-cooperative node as examples, the i-th signal of the p-th non-cooperative nodeAnd the ith signal of the qth uncooperative nodeThe frame interval therebetween is expressed asThe calculation formula is as follows:
the average value of the frame interval between the p-th non-cooperative node and the q-th non-cooperative node is expressed as The calculation formula is as follows:
the variance of the frame interval between the p-th non-cooperative node and the q-th non-cooperative node is expressed as The calculation formula is as follows:
Wherein, Representing the variance of the frame interval of the p-th non-cooperative node and the q-th non-cooperative node.
Then, taking the mean value of the frame interval, the variance of the frame interval and the distance of the node pair obtained in the step 5 as the characteristic of judging whether the non-cooperative node pair is connectedWherein C p＝[c1→p,c2→p,…,cN→p]T represents the spatiotemporal characteristics of nodes other than the p-th non-cooperative node and the p-th non-cooperative node, the characteristics of the p-th non-cooperative node and the q-th non-cooperative node being
And 8, inputting the average value of the frame interval, the variance of the frame interval and the distance information of the node pairs into a trained classification model to obtain the communication relation of each node pair.
Specifically, a known training set is used for training a classification model, and then the mean value, the variance of the frame interval of the node pairs and the distance of the node pairs are used as features to be put into the trained classification model, so that the connection relation of the node pairs is obtained, and the topology inference result of the whole network is obtained.
In this embodiment, the training method of the classification model includes:
The training set is composed of data with known connection relations of node pairs, the frame interval mean value, variance and the distance of the node pairs of the training set are used as characteristic data, and the connection relations of the node pairs are used as labels, so that a trained classification model is obtained.
Specifically, a scene with known non-cooperative node connectivity is used as a training scene ,(C,A)＝((C1,A1),(C2,A2),…,(Ci,Ai),…,(CI,AI)) as a training set to be input into a classification model to be trained, wherein (C i,Ai) is a set of training data, C i is characteristic data of an ith scene, and A i is an adjacency matrix of network topology in the scene, so that a trained classification model is obtained.
In a specific embodiment, data of the non-cooperative network to be tested is fed into a trained classification model to obtain a communication relation of each node pair, namely, an adjacent matrix A of the whole network is obtained.
Specifically, the adjacency matrix of the network is denoted as a= [ a ij ], where the link connection situation a ij of the non-cooperative node v i and the non-cooperative node v j is:
Wherein, 1 indicates that the non-cooperative node v i has a communication relationship with the non-cooperative node v j, and 0 indicates that the non-cooperative node v i does not have a communication relationship with the non-cooperative node v j.
The characteristic data of the wireless network detected by the sensor is C pre, and the characteristic data is input into a trained classification model to obtain an adjacent matrix A of the whole network.
The effectiveness of the wireless network topology inference method based on the space-time features of the present embodiment is verified through experiments as follows.
Specifically, referring to fig. 3, fig. 3 is a system operation block diagram of a wireless network topology inference method based on space-time features according to an embodiment of the present invention. Firstly, detecting the energy distribution condition in a target area through a sensor, and coarsely positioning a target node. Specifically, the power data detected by each sensor are put into a trained neural network model to obtain a positioning result of a target node, then the starting time and the duration of information transmission of the target node are extracted through effective detection of the target node, the mean value, the variance of the frame interval of each node pair and the distance between the node pairs are calculated to obtain the characteristics of whether a communication relationship exists between the node pairs or not, a space-time characteristic diagram is shown in fig. 4, the data of the network are input into the trained classification model, and a topology inference result of the whole network is obtained.
In the simulation, the antenna type of the target node is set to be an omni-directional antenna, a given area is divided into grids with the same number as the sensors, the positioning error output by the training network is finally measured by taking the side length of the divided grids as a unit as shown in fig. 5, and the CDF positioning error of the node is obtained. Fig. 6 shows the inference accuracy of the target node topology inference algorithm.
The invention provides a wireless network topology deducing method based on space-time characteristics, under the condition of determining the position of a non-cooperative node, a sensor node collects the power distribution condition of a target area, feeds the power distribution condition into a trained neural network model to obtain a positioning result of the target node, takes the distance and frame interval information between the nodes as the characteristic for judging whether a communication relation exists between the nodes, and feeds the distance and frame interval information into a trained classification model to obtain the deduced topology.
The wireless network topology inference method based on space-time characteristics combines node positioning and topology inference into a system to realize topology inference work when the position of a target node is unknown.
The positioning method adopts a machine learning method to realize the positioning of the target node.
The topology inference method of the invention provides space-time characteristics (mean value, variance of frame interval and distance between node pairs) as the basis for judging whether the communication relationship exists between the nodes, and adopts a machine learning method to rapidly obtain the topology inference result.
Example two
Yet another aspect of the invention provides an electronic device comprising a memory having a computer program stored therein and a processor implementing the steps of the node location-based unknown uncooperative topology inference method of any of the embodiments above when the computer program in the memory is invoked by the processor.
In the description of the invention, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the description of the present specification, a description of the terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic data point described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristic data points described may be combined in any suitable manner in any one or more embodiments or examples. Further, one skilled in the art can engage and combine the different embodiments or examples described in this specification.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several simple deductions or substitutions may be made without departing from the spirit of the invention, and these should be considered to be within the scope of the invention.

Claims (10)

1.一种基于时空特征的无线网络拓扑推断方法，其特征在于，所述方法包括：1. A wireless network topology inference method based on spatiotemporal features, characterized in that the method comprises: 步骤1、在目标区域设置传感器网络，所述传感器网络包括随机分布的M个传感器节点；Step 1: Setting up a sensor network in a target area, wherein the sensor network includes M randomly distributed sensor nodes; 步骤2、利用所述传感器网络侦测所述目标区域内非协作网络的功率，所述非协作网络包括N个非协作节点；Step 2: Detecting the power of a non-cooperative network in the target area using the sensor network, where the non-cooperative network includes N non-cooperative nodes; 步骤3、将所述目标区域划分成与所述传感器节点的数量相等的M个预设网格，并利用所述传感器网络侦测到的功率确定每个所述预设网格的功率；Step 3: Divide the target area into M preset grids equal to the number of the sensor nodes, and determine the power of each preset grid using the power detected by the sensor network; 步骤4、将步骤3得到的所述预设网格的功率输入至训练好的神经网络模型中，得到所述非协作节点的坐标信息；Step 4: input the power of the preset grid obtained in step 3 into the trained neural network model to obtain the coordinate information of the non-cooperative node; 步骤5、根据步骤4所确定得所述非协作节点的坐标信息得到两个所述非协作节点之间的距离信息；Step 5: obtaining the distance information between two non-cooperative nodes according to the coordinate information of the non-cooperative nodes determined in step 4; 步骤6、根据所述传感器网络侦测的目标区域内所述非协作节点之间交互的时间信息，得到各个所述非协作节点的发送信息的开始时间和结束时间；Step 6: According to the time information of the interaction between the non-cooperative nodes in the target area detected by the sensor network, the start time and the end time of the information sent by each non-cooperative node are obtained; 步骤7、根据所述非协作节点的发送信息的开始时间和结束时间得到各节点对的帧间隔的均值和帧间隔的方差；Step 7: Obtain the mean value and variance of the frame interval of each node pair according to the start time and end time of the information sent by the non-cooperative node; 步骤8、将所述节点对的帧间隔的均值、帧间隔的方差、距离信息输入训练好的分类模型中，得到各节点对的连通关系。Step 8: Input the mean of the frame interval, the variance of the frame interval, and the distance information of the node pairs into the trained classification model to obtain the connectivity relationship of each node pair. 2.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，第i个传感器节点侦测到的功率为：2. The wireless network topology inference method based on spatiotemporal features according to claim 1 is characterized in that the power detected by the i-th sensor node is: 其中，pi,j表示第i个传感器节点侦测到第j个非协作节点的功率，pi,j＝pjGtGr(λ4πRi,j)2，其中，1≤i≤M，1≤j≤N，Gt表示发射天线的天线增益，Gr表示接收天线的天线增益，λ表示波长，Ri,j表示第i个传感器节点侦测到第j个非协作节点之间的距离。Wherein, p i,j represents the power detected by the i-th sensor node to the j-th non-cooperative node, p i,j =p j G t G r (λ4πR i,j ) 2 , where 1≤i≤M, 1≤j≤N, G t represents the antenna gain of the transmitting antenna, Gr represents the antenna gain of the receiving antenna, λ represents the wavelength, and R i,j represents the distance between the i-th sensor node and the j-th non-cooperative node. 3.根据权利要求2所述的基于时空特征的无线网络拓扑推断方法，其特征在于，第m个预设网格的功率为：3. The wireless network topology inference method based on spatiotemporal features according to claim 2, wherein the power of the mth preset grid is: 其中，表示第m个预设网格的功率，B表示第m个预设网格中包含的传感器数目，其中，1≤m≤M。in, represents the power of the mth preset grid, and B represents the number of sensors included in the mth preset grid, where 1≤m≤M. 4.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，所述神经网络模型的训练方法包括：4. The method for inferring wireless network topology based on spatiotemporal features according to claim 1, wherein the training method of the neural network model comprises: S1、获取训练集，所述训练集包括若干非协作训练节点；S1. Obtain a training set, where the training set includes a number of non-cooperative training nodes; S2、将所述非协作训练节点所在的区域划分成H个大小相同的训练网格，其中，H＞M；S2, dividing the area where the non-cooperative training node is located into H training grids of the same size, where H>M; S3、计算第i个传感器训练节点和第j个非协作训练节点之间的距离Ri,j'；S3, calculate the distance R i,j ' between the i-th sensor training node and the j-th non-cooperative training node; S4、根据所述距离Ri,j'得到第i个传感器训练节点到第j个非协作训练节点的训练功率；S4. Obtaining the training power from the i-th sensor training node to the j-th non-cooperative training node according to the distance R i,j ′; S5、将若干输入输出对(p(i),x(i))输入至所述神经网络模型，以训练所述神经网络模型，得到训练好的神经网络模型，其中，p(i)为所述非协作训练节点的训练功率，x(i)为非协作训练节点的真实位置所在训练网格。S5. Input several input-output pairs (p (i) , x (i) ) into the neural network model to train the neural network model and obtain a trained neural network model, wherein p (i) is the training power of the non-cooperative training node, and x (i) is the training grid where the actual position of the non-cooperative training node is located. 5.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，步骤4包括：5. The method for inferring wireless network topology based on spatiotemporal features according to claim 1, wherein step 4 comprises: 步骤4.1、将各个预设网格的功率输入至训练好的神经网络模型中，得到非协作节点的定位结果，以确定非协作节点所在的预设网格；Step 4.1, input the power of each preset grid into the trained neural network model to obtain the positioning result of the non-cooperative node to determine the preset grid where the non-cooperative node is located; 步骤4.2、将非协作节点所在的预设网格的中心位置作为该非协作节点的坐标位置。Step 4.2: The center position of the preset grid where the non-cooperative node is located is used as the coordinate position of the non-cooperative node. 6.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，两个所述非协作节点之间的距离信息为：6. The wireless network topology inference method based on spatiotemporal features according to claim 1, wherein the distance information between the two non-cooperative nodes is: 其中，为第p个非协作节点的坐标，为第q个非协作节点的坐标，dp,q表示第p个非协作节点和第q个非协作节点之间的距离。in, is the coordinate of the pth non-cooperative node, is the coordinate of the qth non-cooperative node, and d p,q represents the distance between the pth non-cooperative node and the qth non-cooperative node. 7.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，所述传感器网络侦测第p个非协作节点获取的数据为：7. The wireless network topology inference method based on spatiotemporal features according to claim 1 is characterized in that the data obtained by the sensor network detecting the pth non-cooperative node is: 其中，n表示Fp中包含的信号数量，T表示矩阵的转置，表示实数集，表示n行2列的实数矩阵；Where n represents the number of signals contained in Fp , T represents the transpose of the matrix, represents the set of real numbers, Represents a real matrix with n rows and 2 columns; 第p个非协作节点的第i个信号为：The i-th signal of the p-th non-cooperative node is: 其中，表示第p个非协作节点的第i个信号发送的开始时间，表示第p个非协作节点的第i个信号的持续时间。in, represents the start time of the i-th signal sent by the p-th non-cooperative node, represents the duration of the i-th signal of the p-th non-cooperative node. 8.根据权利要求7所述的基于时空特征的无线网络拓扑推断方法，其特征在于，所述帧间隔的均值为：8. The wireless network topology inference method based on spatiotemporal features according to claim 7, wherein the mean value of the frame interval is: 其中，表示第p个非协作节点与第q个非协作节点的帧间隔的均值，表示第p个非协作节点与第q个非协作节点的帧间隔，表示第q个非协作节点的第i个信号发送的开始时间。in, represents the mean of the frame intervals between the pth non-cooperative node and the qth non-cooperative node, represents the frame interval between the pth non-cooperative node and the qth non-cooperative node, It indicates the start time of sending the i-th signal of the q-th non-cooperative node. 9.根据权利要求8所述的基于时空特征的无线网络拓扑推断方法，其特征在于，所述帧间隔的方差为：9. The wireless network topology inference method based on spatiotemporal features according to claim 8, wherein the variance of the frame interval is: 其中，表示第p个非协作节点与第q个非协作节点的帧间隔的方差。in, represents the variance of the frame interval between the p-th non-cooperative node and the q-th non-cooperative node. 10.根据权利要求1所述的基于时空特征的无线网络拓扑推断方法，其特征在于，各节点对的连通关系通过邻接矩阵A表示，所述邻接矩阵A为：10. The wireless network topology inference method based on spatiotemporal features according to claim 1, wherein the connectivity relationship of each node pair is represented by an adjacency matrix A, and the adjacency matrix A is: A＝[aij]A＝[a ij ] 其中，非协作节点vi与非协作节点vj的链路连接情况aij为：Among them, the link connection status aij between the non-cooperative node vi and the non-cooperative node vj is: 其中，1表示非协作节点vi与非协作节点vj存在连通关系，0表示非协作节点vi与非协作节点vj不存在连通关系。Among them, 1 indicates that the non-cooperative node vi and the non-cooperative node v j are connected, and 0 indicates that the non-cooperative node vi and the non-cooperative node v j are not connected.
CN202210762307.XA 2022-06-30 2022-06-30 A wireless network topology inference method based on spatiotemporal features Active CN115334612B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210762307.XA CN115334612B (en) 2022-06-30 2022-06-30 A wireless network topology inference method based on spatiotemporal features

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210762307.XA CN115334612B (en) 2022-06-30 2022-06-30 A wireless network topology inference method based on spatiotemporal features

Publications (2)

Publication Number Publication Date
CN115334612A CN115334612A (en) 2022-11-11
CN115334612B true CN115334612B (en) 2025-06-03

Family

ID=83917798

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210762307.XA Active CN115334612B (en) 2022-06-30 2022-06-30 A wireless network topology inference method based on spatiotemporal features

Country Status (1)

Country Link
CN (1) CN115334612B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112272364A (en) * 2020-10-19 2021-01-26 广州大学 Non-cooperative game power control method based on rechargeable wireless sensor network
CN112637952A (en) * 2021-03-03 2021-04-09 南京天创电子技术有限公司 Method for distributing power of wireless cooperative positioning network

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7535911B2 (en) * 2003-05-27 2009-05-19 Alcatel-Lucent Usa Inc. System and method for determining the physical topology of a network having multiple subnets
US7567523B2 (en) * 2004-01-29 2009-07-28 Microsoft Corporation System and method for network topology discovery
US20060256733A1 (en) * 2005-05-12 2006-11-16 Yigal Bejerano Methods and devices for discovering the topology of large multi-subnet LANs
US20090185507A1 (en) * 2006-05-30 2009-07-23 Insightix Ltd. Method and system for determining topology of a dynamic network
EP2903207A1 (en) * 2014-01-30 2015-08-05 Proximetry Poland Sp. z o.o. Systems and methods for topology aware data multicast
US9942717B1 (en) * 2017-06-08 2018-04-10 Rapid Focus Security, Llc Method and apparatus for direction-aware device location
CN113411197A (en) * 2021-04-28 2021-09-17 西安电子科技大学 Non-cooperative key node identification method without prior topological information
CN113328881B (en) * 2021-05-26 2022-05-03 南京航空航天大学 Topology sensing method, device and system for non-cooperative wireless network
CN113489619B (en) * 2021-09-06 2021-11-19 中国人民解放军国防科技大学 A network topology inference method and device based on time series analysis
CN114666227B (en) * 2022-03-18 2024-02-06 中国人民解放军战略支援部队信息工程大学 A method for network topology detection under non-cooperative conditions

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112272364A (en) * 2020-10-19 2021-01-26 广州大学 Non-cooperative game power control method based on rechargeable wireless sensor network
CN112637952A (en) * 2021-03-03 2021-04-09 南京天创电子技术有限公司 Method for distributing power of wireless cooperative positioning network

Also Published As

Publication number Publication date
CN115334612A (en) 2022-11-11

Similar Documents

Publication Publication Date Title
CN112153736A (en) A Person Action Recognition and Location Estimation Method Based on Channel State Information
CN113110207A (en) Insect pest remote monitoring method and system based on sensor of Internet of things and storage medium
Feiyang et al. Monitoring behavior of poultry based on RFID radio frequency network
CN101610516B (en) Intrusion detection method and device thereof in self-organized network
CN111461187B (en) An intelligent detection system for building settlement
CN108038419A (en) Wi-Fi-based indoor personnel passive detection method
Moshtaghi et al. Anomaly detection by clustering ellipsoids in wireless sensor networks
CN102752784B (en) Detection method of distribution type event domain based on graph theory in wireless sensor network
CN105813020A (en) RSSI corrected wireless sensor network positioning algorithm of self-adaptive environment
CN107276999B (en) An event detection method in wireless sensor network
Zhao et al. An accurate approach of device-free localization with attention empowered residual network
CN102608571A (en) Target location method for wireless sensor network for wild animals
CN109547929B (en) Distributed sensor node localization method based on conjugate gradient method
US20220291326A1 (en) A system and method for event recognition
CN101751018B (en) Distributing data monitoring and prealarming system under test environment and method therefor
Chen et al. Noncooperative topology inference of wireless networks with monitoring sensors
CN108834043A (en) Priori knowledge-based compressed sensing multi-target passive positioning method
CN115334612B (en) A wireless network topology inference method based on spatiotemporal features
CN116709242A (en) Side slope displacement multipoint sensor monitoring method
CN115643534A (en) Livestock and poultry positioning management method and system, electronic equipment and storage medium
CN112183624A (en) An anomaly detection method for dam monitoring data based on ensemble learning
Sung Multi-sensors data fusion system for wireless sensors networks of factory monitoring via BPN technology
CN113328881A (en) Topology sensing method, device and system for non-cooperative wireless network
CN119691670A (en) Communication data fusion method and system based on edge calculation
CN114397521A (en) A kind of fault diagnosis method and system for electronic equipment

Legal Events

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
GR01 Patent grant
GR01 Patent grant