CN112423303A - Data fusion method based on trust and gray model - Google Patents
Data fusion method based on trust and gray model Download PDFInfo
- Publication number
- CN112423303A CN112423303A CN202011189176.8A CN202011189176A CN112423303A CN 112423303 A CN112423303 A CN 112423303A CN 202011189176 A CN202011189176 A CN 202011189176A CN 112423303 A CN112423303 A CN 112423303A
- Authority
- CN
- China
- Prior art keywords
- data
- fusion
- cluster
- nodes
- cluster head
- 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.)
- Pending
Links
- 238000007500 overflow downdraw method Methods 0.000 title claims abstract description 11
- 230000004927 fusion Effects 0.000 claims abstract description 69
- 241000854291 Dianthus carthusianorum Species 0.000 claims abstract description 53
- 230000002159 abnormal effect Effects 0.000 claims description 36
- 238000000034 method Methods 0.000 claims description 8
- 238000007499 fusion processing Methods 0.000 claims description 5
- 230000000977 initiatory effect Effects 0.000 claims description 4
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 238000005265 energy consumption Methods 0.000 abstract description 8
- 230000005540 biological transmission Effects 0.000 abstract description 5
- 230000008569 process Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 239000013585 weight reducing agent Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
- H04W12/12—Detection or prevention of fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Security & Cryptography (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A data fusion method based on trust and grey model relates to a model data fusion method, which comprises a node module and a cluster head module; clustering and electing cluster heads in the node module, and after the sensor nodes are deployed, clustering the sensor network by adopting a classic LEACH protocol; initializing and updating the weight in the node module, broadcasting a fusion result in a cluster by a cluster head after data fusion is completed, comparing the stored data with the fusion result by each node in the cluster, and introducing a deviation threshold value
For controlling accuracy; in clusterThe invention aims to improve the reliability of data fusion and reduce the energy consumption of nodes, ensures that the fused data is credible data, reduces the data transmission quantity and simultaneously obtains more reliable results which are closer to the actual situation.
Description
Technical Field
The invention relates to a model data fusion method, in particular to a data fusion method based on a trust model and a gray model.
Background
The transmission energy consumption of data in the wireless sensor network is far greater than the calculation energy consumption, in the sensor network, a large amount of redundant data can be generated by adjacent sensor nodes, and the data prediction and data fusion technology is applied to the network, so that the transmission quantity of the data in the network can be effectively reduced, and the energy consumption is reduced. On the other hand, the reliability of data needs to be considered, default data in a data fusion algorithm is reliable, however, sensor nodes are difficult to avoid being influenced by various external factors (such as environmental mutation, malicious attack and the like) to generate false and unreliable data, so that a trust mechanism is introduced to ensure the reliability of the data and maintain the safety of a network. The data fusion algorithm based on the trust and grey prediction model is provided by researching the energy consumption and the safety of the wireless sensor network.
Disclosure of Invention
The invention aims to provide a data fusion method based on trust and gray models, and designs a data fusion algorithm TGDA based on the trust and gray models from the viewpoint of improving the reliability of data fusion and reducing the energy consumption of nodes. The fused data is guaranteed to be credible data, the data transmission quantity is reduced, and meanwhile, the obtained result is more reliable and closer to the actual situation.
The purpose of the invention is realized by the following technical scheme:
a data fusion method based on trust and grey model comprises a node module and a cluster head module; clustering and cluster head election in the node module are adopted after the sensor nodes are deployedClustering a sensor network by a classical LEACH protocol; initializing and updating the weight in the node module, broadcasting a fusion result in a cluster by a cluster head after data fusion is completed, comparing the stored data with the fusion result by each node in the cluster, and introducing a deviation threshold value
For controlling accuracy; detecting abnormal nodes in a cluster head module, wherein in order to ensure the accuracy and reliability of a fusion result, the cluster head detects the abnormal nodes before initiating data fusion so as to ensure that the data which finally participates in the fusion process is credible data; data prediction based on a gray model in the cluster head module, data loss is caused because data collected by abnormal nodes cannot be used as credible data for data fusion, accuracy of a final fusion result is affected, data collected by the nodes have strong time correlation, and the gray prediction model is introduced to predict the data of the abnormal nodes; credible data in the cluster head module are fused, the cluster head also screens credible data while detecting abnormal nodes, and data predicted by a grey model is adopted to replace data detected by the abnormal nodes; uploading and feeding back the fusion result in the cluster head module, and after the data fusion initiated by the cluster head is completed, respectively uploading the fusion result to a base station and broadcasting in a cluster; the base station can make corresponding decision and judgment by synthesizing the fusion results uploaded by each cluster head; and each node in the cluster compares the stored data with the fed-back fusion result and updates the weight according to the comparison result.
In the data fusion method based on the trust and gray model, the data fusion functions, such as COUNT, AVERAGE, MAX, MIN, are simplified to SUM functions, so that the SUM function defines the fusion function as:
the SUM fusion function defined is as follows:
The invention has the advantages and effects that:
from the perspective of improving the reliability of Data fusion and reducing the energy consumption of nodes, the invention designs a Data fusion algorithm TGDA (trust and Grey Model based Data aggregation) based on a trust and gray Model. The algorithm combines a trust mechanism, data prediction and data fusion with the aim of reducing the data volume in the network and improving the fusion accuracy on the premise of ensuring the reliability of the data fusion and the safety of the network. The cluster head detects abnormal nodes before initiating data fusion, the abnormal nodes with trust values lower than a certain value are pulled into a blacklist, and a grey model is adopted to predict missing data of the abnormal nodes. The algorithm ensures that the fused data is credible data, reduces the data transmission quantity, and simultaneously obtains more reliable results which are closer to the actual situation.
Drawings
FIG. 1 is a TGDA framework of the present invention;
FIG. 2 is a function of the present invention
The variation curve of (d);
FIG. 3 is a flow chart of gray model prediction according to the present invention.
Detailed Description
The present invention will be described in detail with reference to the embodiments shown in the drawings.
Description of the framework
As shown in fig. 1, the TGDA framework is composed of two parts, a node module and a cluster head module. In the node module, firstly, an efficient clustering method is adopted to cluster all nodes and select cluster heads, and all nodes are endowed with an initial weight. In the cluster head module, cluster head nodes collect the weight of each node to divide a credible node and an abnormal node, a blacklist, namely information of the abnormal node, is broadcasted in a cluster, meanwhile, data collected by the abnormal node is discarded, and a grey prediction method is adopted to predict missing data of the abnormal node. And finally, fusing the credible data by the cluster head node, and sending a fusion result to the base station and each node in the cluster. The base station integrates all the fusion results, and the nodes in the cluster compare the stored data with the feedback fusion results and update the weight according to the comparison results.
Second, TGDA algorithm design
1. Node module
(1) Clustering and cluster head election
After the sensor nodes are deployed, clustering is carried out on the sensor network by adopting a classic LEACH protocol. The LEACH protocol randomly selects cluster head nodes, so that the probability that each node is selected as a cluster head is the same (except for abnormal nodes on a blacklist), and the premature death of some nodes can not be caused, thereby realizing the energy consumption load balance. Before selecting cluster head, each node will generate a random number
If, if
Less than a given threshold
Then the node is selected as a cluster head in the current round.
Calculated by formula (1).
Wherein the content of the first and second substances,
indicating the proportion of cluster head nodes required in a particular network;
indicating the proportion of cluster head nodes required in a particular network;
is the current round;
is front
The wheel is not selected as the set of nodes of the cluster head.
(2) Initialization and updating of weights
The selected cluster head is generally secure and reliable since the sensor node just deployed is generally not damaged or attacked within a short period of time. The cluster head assigns an initial weight to each node in the cluster according to the residual energy of the node
And stored in the corresponding node. Typically, the initial weight of each node is equal.
After the data fusion is completed, the cluster head broadcasts a fusion result in the cluster, each node in the cluster compares the stored data with the fusion result,and introducing a deviation threshold
For controlling the accuracy. The process is as follows:
1) calculating a deviation of the stored data and the fused result
2）
3) Inaccurate information collection, weight reduction:
4）else
5) accurate information acquisition and unchanged weight
Under the condition of inaccurate collected information, the weight is reduced, and
is defined as follows:
wherein the content of the first and second substances,
is the number of abnormal nodes in the fusion process;
is the number of nodes in a cluster; parameter(s)
Is used to control the rate of change. Suppose that
Taking out the raw materials of 20 percent,
get 4, then function
The variation curve of (2) is shown in fig. 2. When the number of abnormal nodes in the cluster is small,
the change is gentle, the weight is slowly reduced, and the influence on data fusion is small; but when the number of abnormal nodes is large, as shown in the figure, when
When the temperature of the water is higher than the set temperature,
and the weight is rapidly reduced until the weight is 0, so that the data fusion process cannot be continued, and the safety and the accuracy of the data fusion are ensured. As shown in fig. 2.
2. Cluster head module
(1) Abnormal node detection
In order to ensure the accuracy and reliability of the fusion result, the cluster head can detect abnormal nodes before initiating data fusion so as to ensure that the data participating in the fusion process finally is credible data. Therefore, a weight threshold needs to be set
To distinguish trusted nodes from anomalous nodes. The process is as follows:
1) cluster head collects weight of each node in cluster
2）if
3) This node is an abnormal node, pulled into the blacklist
4）if
5) The node is a credible node and can perform data fusion
After all nodes in the cluster are divided, the cluster head broadcasts the blacklist in the cluster. Abnormal nodes are marked and cannot participate in the process of cluster head election, and the weight of the abnormal nodes is set as the initial weight
And remain unchanged. Therefore, once some nodes are abnormal, the nodes are considered as incapable of being used as cluster heads for data fusion, and collected data cannot be used as credible data.
(2) Data prediction based on gray model
Data acquired by the abnormal nodes cannot be used as credible data to perform data fusion, so that data loss is caused, and the accuracy of a final fusion result is influenced. In consideration of strong time correlation among data collected by nodes, a gray prediction model GM (1,1) is introduced to predict the data of abnormal nodes. Since the nodes which are just deployed can not be attacked or damaged in a short time, all the nodes can work normally firstly
And (4) wheels. Suppose that when a node finishes transmitting
After the data are round, the data are detected as abnormal nodes, and the cluster head nodes call the gray model to predict the abnormal nodes
The wheel data. Since the accuracy of predictions using recent data can be high, an abnormal node is employed
Wheel data, i.e. recent
The round data constitutes the original time series.
Let the original time series of GM (1,1) model
：
Because the original sequence has a certain randomness, a first-order accumulation sequence needs to be constructed to obtain a generation sequence with strong regularity
：
Wherein the content of the first and second substances,
the expression is as follows:
fitting and approximating the data generated by accumulation by using a dynamic linear model to obtain a differential equation corresponding to the GM (1,1) model:
in the formula (I), the compound is shown in the specification,
and
for variables to be solved, note
. Solving by a least square method to obtain:
wherein
The prediction model can be obtained by solving the differential equation
：
From equation (8) by subtraction
FIG. 3 is a flow chart of prediction using a gray model:
(3) trusted data fusion
The cluster head also screens out credible data while detecting abnormal nodes, and data predicted by a grey model is adopted to replace data detected by the abnormal nodes. Compared with data loss caused by other methods, the data fusion method based on the gray model enables all the data fused by the cluster heads to be credible data, and therefore the obtained result is more reliable and closer to the real situation. Since many classical data fusion functions, such as COUNT, AVERAGE, MAX, MIN, etc., can be simplified to SUM function, SUM function is used as a research object herein, and the fusion function is defined as:
the SUM fusion function defined is as follows:
(4) Fusion result uploading and feedback
After the data fusion initiated by the cluster head is completed, the fusion result is respectively uploaded to the base station and broadcasted in the cluster. The base station can make corresponding decision and judgment by synthesizing the fusion results uploaded by each cluster head; and each node in the cluster compares the stored data with the fed-back fusion result and updates the weight according to the comparison result.
Claims (2)
1. A data fusion method based on trust and grey models is characterized by comprising a node module and a cluster head module; clustering and electing cluster heads in the node module, and after the sensor nodes are deployed, clustering the sensor network by adopting a classic LEACH protocol; initializing and updating the weight in the node module, and after the data fusion is completed, the cluster head canBroadcasting the fusion result in the cluster, comparing the stored data with the fusion result by each node in the cluster, and introducing a deviation threshold value
For controlling accuracy; detecting abnormal nodes in a cluster head module, wherein in order to ensure the accuracy and reliability of a fusion result, the cluster head detects the abnormal nodes before initiating data fusion so as to ensure that the data which finally participates in the fusion process is credible data; data prediction based on a gray model in the cluster head module, data loss is caused because data collected by abnormal nodes cannot be used as credible data for data fusion, accuracy of a final fusion result is affected, data collected by the nodes have strong time correlation, and the gray prediction model is introduced to predict the data of the abnormal nodes; credible data in the cluster head module are fused, the cluster head also screens credible data while detecting abnormal nodes, and data predicted by a grey model is adopted to replace data detected by the abnormal nodes; uploading and feeding back the fusion result in the cluster head module, and after the data fusion initiated by the cluster head is completed, respectively uploading the fusion result to a base station and broadcasting in a cluster; the base station can make corresponding decision and judgment by synthesizing the fusion results uploaded by each cluster head; and each node in the cluster compares the stored data with the fed-back fusion result and updates the weight according to the comparison result.
2. A method for fusion of data based on trust and grey model according to claim 1, characterized in that the data fusion functions, such as COUNT, AVERAGE, MAX, MIN are reduced to SUM function, so that with SUM function, the fusion function is defined as:
the SUM fusion function defined is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011189176.8A CN112423303A (en) | 2020-10-30 | 2020-10-30 | Data fusion method based on trust and gray model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011189176.8A CN112423303A (en) | 2020-10-30 | 2020-10-30 | Data fusion method based on trust and gray model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112423303A true CN112423303A (en) | 2021-02-26 |
Family
ID=74827831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011189176.8A Pending CN112423303A (en) | 2020-10-30 | 2020-10-30 | Data fusion method based on trust and gray model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112423303A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117593A (en) * | 2015-08-18 | 2015-12-02 | 河海大学 | Wavelet transform and particle swarm optimized grey model-based short-term wind speed forecasting method |
| CN105636081A (en) * | 2016-01-26 | 2016-06-01 | 大连理工大学 | Improved EAST (Efficient Data Collection Aware of Spatio-Temporal Correlation) algorithm based on grey model prediction |
| CN106231617A (en) * | 2016-07-18 | 2016-12-14 | 南京航空航天大学 | Wireless sensor network multi-Sensor Information Fusion Approach based on fuzzy logic |
| CN108337685A (en) * | 2018-01-26 | 2018-07-27 | 重庆邮电大学 | Wireless sensor network data fusion method based on sub-clustering DGM |
-
2020
- 2020-10-30 CN CN202011189176.8A patent/CN112423303A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117593A (en) * | 2015-08-18 | 2015-12-02 | 河海大学 | Wavelet transform and particle swarm optimized grey model-based short-term wind speed forecasting method |
| CN105636081A (en) * | 2016-01-26 | 2016-06-01 | 大连理工大学 | Improved EAST (Efficient Data Collection Aware of Spatio-Temporal Correlation) algorithm based on grey model prediction |
| CN106231617A (en) * | 2016-07-18 | 2016-12-14 | 南京航空航天大学 | Wireless sensor network multi-Sensor Information Fusion Approach based on fuzzy logic |
| CN108337685A (en) * | 2018-01-26 | 2018-07-27 | 重庆邮电大学 | Wireless sensor network data fusion method based on sub-clustering DGM |
Non-Patent Citations (4)
| Title |
|---|
| 余修武等: "基于无偏灰色马尔科夫预测WSN数据融合算法", 《传感技术学报》 * |
| 余修武等: "无线传感器网络自适应预测加权数据融合算法", 《传感技术学报》 * |
| 梁青等: "基于灰色预测的无线传感器网络数据融合算法", 《西安邮电大学学报》 * |
| 王汝言等: "WSN中基于GM-LSSVM的数据融合方法", 《计算机工程与设计》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108092826B (en) | Wireless sensor network security model based on backbone node security role hierarchy | |
| CN112019373B (en) | Intelligent home security data acquisition method based on dynamic trust evaluation model | |
| CN110166454B (en) | Mixed feature selection intrusion detection method based on adaptive genetic algorithm | |
| CN108124261B (en) | Wireless sensor network security clustering method integrating credit assessment and patrol mechanism | |
| CN113242259B (en) | Network abnormal flow detection method and device | |
| CN107733877B (en) | Management method and system for wireless communication architecture of Internet of things | |
| CN103237308B (en) | Distributed intrusion detection method of vehicle ad hoc network | |
| CN108848072A (en) | A kind of vehicle-mounted CAN bus method for detecting abnormality based on relative entropy | |
| CN111970229B (en) | CAN bus data anomaly detection method aiming at multiple attack modes | |
| CN112380081A (en) | Commercial vehicle-mounted CAN bus fuzzy test method based on field weight | |
| CN103747003A (en) | Peer-to-peer botnet core node detection method and detection device | |
| Zhang et al. | An energy-efficient multilevel secure routing protocol in IoT networks | |
| Bhargava et al. | Kate: Kalman trust estimator for internet of drones | |
| CN112423303A (en) | Data fusion method based on trust and gray model | |
| CN110213770A (en) | The anti-collusion method and system of wireless sensor network | |
| CN114553458A (en) | Method for establishing and dynamically maintaining credible group in power Internet of things environment | |
| CN105246173A (en) | Node clustering method suitable for RSN (Radar Sensor Network) and fusion decision method of node clustering method | |
| CN115208604B (en) | AMI network intrusion detection method, device and medium | |
| CN107371125B (en) | Wireless sensor network fault repairing method and device based on particle swarm optimization | |
| Ding et al. | In-vehicle network intrusion detection system based on Bi-LSTM | |
| CN111489317B (en) | Intelligent cinerary casket storage system | |
| Murugan et al. | Ensemble of ADA booster with SVM classifier for anomaly intrusion detection in wireless ad hoc network | |
| CN114553606A (en) | Industrial control network intrusion detection method and system | |
| CN110971471B (en) | Power communication backbone network fault recovery method and device based on state perception | |
| CN112437407A (en) | Method and system for calculating initialized safety state of power internet of things wireless terminal equipment based on mixed information entropy |
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 |