CN113015195A - Wireless sensor network data acquisition method and system - Google Patents
Wireless sensor network data acquisition method and system Download PDFInfo
- Publication number
- CN113015195A CN113015195A CN202110186912.2A CN202110186912A CN113015195A CN 113015195 A CN113015195 A CN 113015195A CN 202110186912 A CN202110186912 A CN 202110186912A CN 113015195 A CN113015195 A CN 113015195A
- Authority
- CN
- China
- Prior art keywords
- data
- node
- abnormal
- nodes
- emergency
- 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/04—Arrangements for maintaining operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0268—Traffic management, e.g. flow control or congestion control using specific QoS parameters for wireless networks, e.g. QoS class identifier [QCI] or guaranteed bit rate [GBR]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
- H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
-
- 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
Abstract
The invention discloses a method and a system for acquiring data of a wireless sensor network, wherein the method comprises the following steps: screening nodes with larger data value changes acquired at the current moment, and judging whether the nodes are abnormal nodes or not; acquiring an abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low reliability and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data; the invention has the advantages that: the node reliability and the data reliability are accurately judged, meanwhile, the rapid collection of emergency data is guaranteed, and the accuracy of a monitoring result is improved, so that the service quality of the wireless sensor network can better meet the application requirement.
Description
Technical Field
The invention relates to the field of data acquisition, in particular to a method and a system for acquiring data of a wireless sensor network.
Background
The wireless Sensor network wsn (wireless Sensor network) has a wide application in many fields such as military, medical treatment and industry, and is intensively embodied in the aspects of data collection and event monitoring. However, wireless sensor nodes are mostly battery powered, have limited computing power and bandwidth capacity, and are often deployed in some harsh environments. These factors easily cause the nodes to be abnormal and interfered, and further the reliability of the monitoring data cannot be guaranteed. Wireless sensor networks are data-centric networks, where rapid response to emergency events and reliable data acquisition are essential.
Data exception caused by node self fault or external environment interference is usually caused on one hand by abnormal data in the WSN; on the other hand, the data change caused by an emergency event exceeds the normal amplitude. The former is an important factor causing the reliability of the network to be reduced, and the latter is a main monitoring task of the wireless sensor network. Abnormal data detection in wireless sensor networks typically relies on statistical methods or exploits spatial correlation between neighboring nodes. For example, in the Hidden Markov model-based data anomaly detection method in the literature, "Wang Chen, Lin Hong Zhi, Jiang Hong Bo, project-based multi-dimensional outer detection in Wireless sensor Networks using high Markov Models [ J ]. Wireless Networks,2014,20(8): 2409-; the document "Wazid M, Das A K.An Efficient analysis Detection Scheme Using K-Means Clustering for Wireless sensory networks [ J ]. Wireless Personal Communications,2016,90(4): 1971-. Data acquisition is one of the main functions of the wireless sensor network, and the reliability of the acquired data directly determines the safe operation of the network, but the requirement on the data reliability of the wireless sensor network cannot be completely met only by screening abnormal data. Data in the wireless sensor network come from a plurality of sensor nodes, so that the reliability of the sensor nodes is one of important inducements influencing the reliability of data acquisition, the problems of routing selection, data authentication and the like of the sensor nodes are solved due to the occurrence of a trust model, and theoretical support is provided for judging the reliability of the nodes. Currently, trust models are widely applied in the field of wireless sensor networks, wherein the documents "ISHMANOV F, MALIK A S, KIM S W, et al. Trust management system in wireless sensor networks: design contacts and research channels [ J ]. transformation on electronic computing communication Technologies,2015,26(2): 107) 130" help to find reliable nodes through the trust models, thereby improving the cooperation among the nodes and the performance of the system; the document ' Zhao Zhi Guo, Tan Ming Sheng, Xishiyingshu, Wangshu ' wireless sensing network trust model [ J ] based on time factors computer engineering and design, 2017,38(4):883 + 887 ' proposes a WSN trust model based on time factors, utilizes a dynamic feedback mechanism to calculate the node trust degree, and simultaneously introduces a time attenuation model to reduce the network delay and increase the model reliability; in the document, an improved WSNs trust evaluation model [ J ] based on Bayes, a sensing technology academic report, 2016,29(06):927 plus 933, uses abnormal attenuation to correct a Bayes equation during direct trust calculation, uses entropy as reference to serve as node trust degree to assign weights, and well overcomes the limitation of subjective assignment. At present, WSN data acquisition mainly aims at removing redundancy and reducing energy consumption, for example, in the document' Yikezhen, Zhouxinshen, Mao technology, Chengqing, a WSN data acquisition algorithm research [ J ] of an improved region growing method, a small-sized microcomputer system, 2019(3): 567-; the document Zhou Wei, Jing Bo, Huang Piao, Jade, Hu Jia xing, Liangwei, the CS-based airborne clustering-type WSN data acquisition method [ J ] Communications, 2015,36(5): 130-. However, as the application requirements of the internet of things are continuously improved, such as internet of vehicles, automatic driving and the like, the reliability of the sensing layer data becomes more important.
In practical applications, the sensor nodes generally collect and transmit data at a constant frequency, and in most cases, the data changes slowly, such as temperature, humidity, pressure, and the like. In this case, the existence of the unreliable node directly affects the accuracy of the monitoring result, so that the accurate judgment of the reliability of the node and the reliability of the data is related to the quality of service (quality of service) of the network.
Disclosure of Invention
The technical problem to be solved by the invention is that the data acquisition method of the wireless sensor network in the prior art lacks accurate judgment on the node reliability and the data reliability, and directly influences the accuracy of the monitoring result, so that the service quality of the wireless sensor network cannot well meet the requirement.
The invention solves the technical problems through the following technical means: a wireless sensor network data acquisition method, the method comprising:
the method comprises the following steps: screening nodes with larger data value changes acquired at the current moment, and judging whether the nodes are abnormal nodes or not;
step two: acquiring an abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low reliability and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
Based on the fact that the data abnormity of most nodes is caused by an emergency and the data abnormity of few nodes is caused by node abnormity or interference with a certain probability, the invention firstly judges whether the data nodes are abnormal or not and counts the abnormal data node proportion. For a few 'data abnormity', namely when the proportion of the abnormal data nodes is lower than a scale threshold value, filtering abnormal nodes with low reliability, and then starting an emergency data acquisition mechanism to acquire data; when data abnormity occurs in most nodes, namely the occupation ratio of the abnormal data nodes is higher than a scale threshold value, an emergency data acquisition mechanism is started to acquire data, the reliability of the nodes and the reliability of the data are accurately judged, meanwhile, the rapid collection of the emergency data is guaranteed, the accuracy of monitoring results is improved, and therefore the service quality of the wireless sensor network can well meet the requirements.
Further, the first step comprises: setting a data window with the size of M based on the time correlation of the monitoring data, and regarding the node i, the data acquired at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3Is the upper quartile value of M-1 data, and IQR is equal to Q3-Q1。
Further, the second step comprises:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Representing the appearance of a small part of abnormal values, and adopting a trust model-based acquisition mode to acquire data; wherein the content of the first and second substances,
represents a scale threshold;
step 203: when in use
And (4) indicating that data abnormity occurs in most nodes, and starting an emergency data acquisition mechanism to acquire data.
Still further, the step 202 includes:
acquiring the credibility of the cluster head to the abnormal node i;
acquiring the data offset of the abnormal node i;
acquiring the trust degree of the abnormal node i according to the trust degree of the cluster head to the abnormal node i and the data migration degree of the abnormal node i, and filtering the node data when the trust degree of the data is lower than a threshold lambda; and when the trust degree of the abnormal node i is greater than the threshold lambda, collecting the data of the node, and starting an emergency data collection mechanism to collect the data.
Still further, the obtaining the reliability of the cluster head on the abnormal node i includes:
by the formula
Obtaining the credibility D of the cluster head to the abnormal node iiWherein E () represents a probability distribution operator, αijIndicating a successful interaction alphaijSecond, betaijIndicating a failed interaction betaijNext, the process of the present invention,
represents the probability density distribution function of abnormal node i relative to cluster head node j and
for the probability of successful interaction, alpha represents the historical interaction success times of the abnormal node i, beta represents the historical interaction failure times of the cluster head node j, alpha is greater than 0, beta is greater than 0, and f () represents a gamma function.
Still further, the obtaining the data offset degree of the abnormal node i includes:
for the abnormal node i, k neighbor nodes in the same cluster communication radius are selected, and the current data collected by the k neighbor nodes is set as ai1，ai2....aikAnd then the average value of the data of the neighboring nodes is:
by the formula Ti＝ω*|ai-Ei|+(1-ω)*|ai-EkI, acquiring the data offset of the abnormal node i, wherein omega is a first balance coefficient and belongs to [0, 1 ]]，aiIndicating that the abnormal node i is currently collecting data, EiAnd collecting historical mean values of data for the abnormal nodes i.
Still further, the obtaining of the trust level of the abnormal node i according to the trust level of the cluster head on the abnormal node i and the data migration level of the abnormal node i includes:
by the formula
Obtaining the Trust degree of the abnormal node i, wherein TrustiIndicating the degree of trust of the abnormal node i,
is a second equilibrium coefficient and
the invention also provides a wireless sensor network data acquisition system, which comprises:
the abnormal node judgment module is used for screening the nodes with larger data value change acquired at the current moment and judging whether the nodes are abnormal nodes or not;
the data acquisition module is used for acquiring the abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low credibility and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
Further, the abnormal node determining module is further configured to: setting a data window with the size of M based on the time correlation of the monitoring data, and regarding the node i, the data acquired at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3Is the upper quartile value of M-1 data, and IQR is equal to Q3-Q1。
Further, the data acquisition module is further configured to:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Representing the appearance of a small part of abnormal values, and adopting a trust model-based acquisition mode to acquire data; wherein the content of the first and second substances,
represents a scale threshold;
step 203: when in use
And (4) indicating that data abnormity occurs in most nodes, and starting an emergency data acquisition mechanism to acquire data.
Still further, the step 202 includes:
acquiring the credibility of the cluster head to the abnormal node i;
acquiring the data offset of the abnormal node i;
acquiring the trust degree of the abnormal node i according to the trust degree of the cluster head to the abnormal node i and the data migration degree of the abnormal node i, and filtering the node data when the trust degree of the data is lower than a threshold lambda; and when the trust degree of the abnormal node i is greater than the threshold lambda, collecting the data of the node, and starting an emergency data collection mechanism to collect the data.
Still further, the obtaining the reliability of the cluster head on the abnormal node i includes:
by the formula
Obtaining the credibility D of the cluster head to the abnormal node iiWherein E () represents a probability distribution operator, αijTo representSuccessful interaction alphaijSecond, betaijIndicating a failed interaction betaijNext, the process of the present invention,
represents the probability density distribution function of abnormal node i relative to cluster head node j and
for the probability of successful interaction, alpha represents the historical interaction success times of the abnormal node i, beta represents the historical interaction failure times of the cluster head node j, alpha is greater than 0, beta is greater than 0, and f () represents a gamma function.
Still further, the obtaining the data offset degree of the abnormal node i includes:
for the abnormal node i, k neighbor nodes in the same cluster communication radius are selected, and the current data collected by the k neighbor nodes is set as ai1，ai2....aikAnd then the average value of the data of the neighboring nodes is:
by the formula Ti＝ω*|ai-Ei|+(1-ω)*|ai-EkI, acquiring the data offset of the abnormal node i, wherein omega is a first balance coefficient and belongs to [0, 1 ]]，aiIndicating that the abnormal node i is currently collecting data, EiAnd collecting historical mean values of data for the abnormal nodes i.
Still further, the obtaining of the trust level of the abnormal node i according to the trust level of the cluster head on the abnormal node i and the data migration level of the abnormal node i includes:
by the formula
Obtaining the Trust degree of the abnormal node i, wherein TrustiIndicating the degree of trust of the abnormal node i,
is a second equilibrium coefficient and
the invention has the advantages that: based on the fact that the data abnormity of most nodes is caused by an emergency and the data abnormity of few nodes is caused by node abnormity or interference with a certain probability, the invention firstly judges whether the data nodes are abnormal or not and counts the abnormal data node proportion. For a few 'data abnormity', namely when the proportion of the abnormal data nodes is lower than a scale threshold value, filtering abnormal nodes with low reliability, and then starting an emergency data acquisition mechanism to acquire data; when data abnormity occurs in most nodes, namely the occupation ratio of the abnormal data nodes is higher than a scale threshold value, an emergency data acquisition mechanism is started to acquire data, the reliability of the nodes and the reliability of the data are accurately judged, meanwhile, the rapid collection of the emergency data is guaranteed, the accuracy of monitoring results is improved, and therefore the service quality of the wireless sensor network can well meet the requirements.
Drawings
Fig. 1 is a clustered network topology in a data acquisition method of a wireless sensor network according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating data anomaly determination based on a box model in a wireless sensor network data acquisition method according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an emergency data transmission mechanism in a data acquisition method of a wireless sensor network according to an embodiment of the present invention;
fig. 4 is a diagram illustrating a comparison between a data acquisition method of a wireless sensor network provided by an embodiment of the present invention and a detection rate of a classical trust model RFSN, where AT-DG represents an algorithm of the present invention;
fig. 5 is a graph illustrating a mean value change trend of the trust levels of normal nodes and abnormal nodes in a cluster, which is randomly selected when the percentage of the abnormal nodes is 15% in the data acquisition method for the wireless sensor network according to the embodiment of the present invention;
fig. 6 is a comparison diagram of data acquisition accuracy of a wireless sensor network data acquisition method and a native HEED algorithm according to an embodiment of the present invention;
fig. 7 is a comparison diagram of a wireless sensor network data acquisition method and a native HEED algorithm in terms of total network energy consumption according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
1. Network model
The layered WSN network topology is widely applied to various monitoring activities due to its simplicity, high efficiency and convenient management. The adjacent sensor nodes are combined according to a certain rule to form a cluster, as shown in fig. 1. Each cluster is composed of a cluster head and a plurality of member nodes. Because member nodes in the same cluster have certain correlation between communication behaviors and spatial characteristics, the cluster head can better distinguish unreliable data by establishing a trust model. Meanwhile, the occurrence of the event is also shown as that the data change difference in a part of clusters is large, and an abnormality detection mechanism taking the clusters as units is easier to realize. The invention carries out hierarchical division on the network according to the HEED protocol, and the sensor nodes in the network meet the following conditions: firstly, once the position of a node in a network is determined, the node does not move, and each node has a unique identifier; only a single Sink node exists in the network, and the data route in the network is directly or indirectly sent to the Sink node by the cluster head; each member node in the cluster is adjacent in position and can only communicate with the cluster head; and fourthly, the communication radius and the routing energy consumption of each node are the same, and the energy is not renewable. The invention provides a wireless sensor network data acquisition method based on a layered WSN network topological structure, which has the general idea that:
the method comprises the following steps: screening nodes with larger data value changes acquired at the current moment, and judging whether the nodes are abnormal nodes or not;
step two: acquiring an abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low reliability and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
In order to facilitate understanding of the solution of the present invention, the whole design process of the present invention is described in detail in the following sections:
2 data acquisition scheme
2.1 data anomaly discrimination and Collection rules
Generally, if most node monitoring data in the cluster have mutation, the monitoring event should be a large probability, and the mutation of the few node monitoring data is a data collection error to a large extent. Therefore, whether the abnormal data is the monitoring event occurrence or the error data can be distinguished by setting a threshold value of the abnormal data ratio. For the judgment of abnormal data, the abnormal data can be defined by a distance-based method in statistics, namely, the difference value of the front and back changes of the data. In order to highlight the variation of the abnormal data and the proportion of the abnormal data, the threshold is set according to the definition of the variation of the data, and then the proportion of the nodes is judged, and further whether an event occurs or the data is wrong is judged. The data level anomaly detection usually utilizes the characteristics of timeliness, spatiality and the like of node data flow, and the node trust degree is calculated by comparing the difference between the current anomalous data and the historical data and the data of the adjacent nodes. The classification-based method in the data anomaly detection method is not suitable for being realized at the nodes of the wireless sensor network with limited resources because the detection accuracy of the classification-based method is influenced by the scale of the data set, and the anomaly detection based on statistics better meets the requirements of the wireless sensor network, wherein the quartile spreading method has small calculated amount and high convergence speed, so that a box-type model of the quartile spreading method is selected as the basis of anomaly detection.
In step one, a data window of size M is set based on the time correlation of the monitored data, as shown in fig. 2. For node i, the data collected at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3For the upper four quantile values in M-1 data, adopting a downward rounding strategy to determine Q when calculating the quantile1And Q3A value of, wherein IQR ═ Q3-Q1. And carrying out interval division on each node in the cluster through historical data, and taking the interval as a judgment basis for data abnormity at the current time.
The occurrence of the monitoring event is mainly represented by local data change, and the size of the event scale is positively correlated with the number of data change nodes. Therefore, the data anomaly percentage can be defined as the ratio of the number of data anomaly nodes to the number of summary nodes in the cluster, and therefore, the second step comprises:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Representing that a small part of abnormal values are generated and possibly caused by node abnormality, and performing data acquisition by adopting an acquisition mode based on a trust model; the algorithm is based on the size of the abnormal scale of the network as the acquisition basis, and the occurrence of the monitoring event has low probability and locality, so that the method is convenient to research
Is a size threshold. In practical application, the system can be adjusted according to the performance of the monitoring system
Taking values; the specific process of step 202 is:
acquiring the credibility of the cluster head to the abnormal node i;
acquiring the data offset of the abnormal node i;
acquiring the trust degree of the abnormal node i according to the trust degree of the cluster head to the abnormal node i and the data migration degree of the abnormal node i, and filtering the node data when the trust degree of the data is lower than a threshold lambda; and when the trust degree of the abnormal node i is greater than the threshold lambda, collecting the data of the node, and starting an emergency data collection mechanism to collect the data.
Step 203: when in use
The data abnormality of most nodes is shown, which may be caused by the occurrence of an event, and an emergency data acquisition mechanism is started for data acquisition.
2.2 node screening based on Trust model
The data exception of a few nodes is possibly caused by node failure, but the data exception caused by the occurrence of an event is not excluded. Therefore, the nodes with data abnormity need to be considered from two aspects of nodes and data, and theoretical basis is provided for the reliability data acquisition scheme through calculation of the trust degree of the nodes and the data offset.
2.2.1 calculation of node-level Trust
The communication quality of the nodes is an important guarantee for data reliability, and the node-level trust can be calculated by utilizing the communication behavior of the nodes in the cluster and the data correlation. WSN security model [ J ] based on skeleton node security role hierarchy]Chongqing post and telecommunications university newspaper: science edition, 2019, 31 (5): 722 + 728 and improved WSNs trust evaluation model based on Bayes]Technical report of sensing, 2016,29(06): 927-. The decision can be provided for the posterior event by combining a Bayes formula through the known conditional probability density parameter expression and the prior probability. Then node i and node j reputation distribution reputationsijCan be expressed as
reputationij～Beta(α+1，β+1) (3)
Wherein alpha represents the historical interaction success times of the abnormal node i, beta represents the historical interaction failure times of the cluster head node j, alpha is larger than 0, and beta is larger than 0. The similarity degree of the node interaction history and the adjacent points is used as the main basis of the node fault, and the probability density distribution function of the node i relative to the node j obtained by the Bayesian evaluation method is as follows:
wherein E () represents a probability distribution operator, αijIndicating a successful interaction alphaijSecond, betaijIndicating a failed interaction betaijNext, the process is carried out.
2.2.2 Trust calculation of data offset
The credibility calculation of the node is a precondition for data reliability acquisition and is also a judgment basis for whether the node is normal or not. Because the invention uses the node reliability of the data layer to collect, the data deviation degree is introduced to improve the accuracy of the collection model. On the basis of node-level reliability calculation, abnormal node data migration degree is calculated by using self node historical data and adjacent node data. For the calculation of the data migration degree of the abnormal node i, k neighbor nodes in the same cluster communication radius are selected at first, and the current data collected by the k neighbor nodes is set as ai1，ai2....aik. The mean value of the data of the neighboring nodes is:
meanwhile, M-1 historical data before the current moment are taken from the window for the abnormal node i, and the historical mean value is EiTherefore, the data offset of the abnormal node i is:
Ti＝ω*|ai-Ei|+(1-ω)*|ai-Ek| (7)
wherein, ω is a first balance coefficient and ω ∈ [0, 1 ]]，aiIndicating that the abnormal node i is currently collecting data, EiAs abnormal nodesi historical mean of the collected data.
2.2.3 abnormal data node Trust calculation
For judging the data abnormal node, the quality and the data of the node are comprehensively considered, and the definitions of the node quality and the data deviation are given by formula (5) and formula (7). The higher the trust degree of the node is and the smaller the data deviation from the surrounding nodes is, the higher the reliability of the abnormal node of the data is. Trust of abnormal node iiCan be expressed as:
wherein, TrustiIndicating the degree of trust of the abnormal node i,
is a second equilibrium coefficient and
trust when abnormal node iiWhen the data is larger than the threshold lambda, the data of the node is collected and sent, and an emergency data transmission mechanism is started; and when the data reliability is lower than the threshold lambda, filtering the node data.
2.3 Emergency data acquisition mechanism
When data flow caused by occurrence of monitoring events such as forest fires, gas leakage and the like is abnormal, the sensor nodes need to be capable of accurately sensing and quickly transmitting to the sink nodes. And aiming at different monitoring data types, corresponding data acquisition modes are adopted. Normal common data is transmitted in a conventional data transmission mode; and determining the collection mode of the abnormal data according to whether the proportion of the abnormal data exceeds a set threshold value. If the proportion of the abnormal data exceeds a set threshold value, an emergency data acquisition mechanism is adopted, otherwise, the acquisition mode of the abnormal data is determined according to the credibility of the abnormal data, the abnormal data with low credibility is filtered in time, and the abnormal data with high credibility is adopted the emergency data acquisition mechanism.
At present, research on a reliable data transmission mechanism mainly starts from two aspects of congestion control and reliability guarantee. The burst event can cause data to increase rapidly, the buffer space of the cluster head is limited, and the flooding of a large amount of data can cause queue overflow, so that the transmission delay and the packet loss rate of a data packet are increased rapidly. Therefore, for the burst data flow in the emergency, the entire network resources need to be fully utilized to ensure the reliable transmission of the emergency data. In addition, each cluster head in the cluster structure also has the problems of link competition and data collision in the data transmission process, and also has the relation to the problem of reliable transmission of emergency data. Based on the characteristics of the burst event data stream, a Redundant Assistance transmission Mechanism (RAM) is provided. The RAM fully utilizes cluster head resources in a normal state, and based on a routing idea of emergency data priority, the method ensures accurate perception and timely early warning of emergency events. The specific RAM work flow is as follows:
step 1: when the cluster head detects that the trust degree of the abnormal node is high or the abnormal node exceeds a scale threshold, an emergency data acquisition mechanism is started immediately;
step 2: the cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises the cluster position information;
step 3: after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster itself, reduces the data forwarding of other cluster heads, and improves the priority of requesting the data forwarding of the cluster head to ensure the emergency data transmission of the cluster head;
step 4: and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
The RAM meets the urgent data transmission needs by increasing the urgent cluster head priority and reducing the downstream cluster head acquisition frequency, as shown in fig. 3. Although the RAM mechanism sacrifices partial hierarchical data acquisition, the timely reporting and processing of the network to the emergency can be guaranteed to the greatest extent. In practice the frequency of the emergency is low and the impact of the setting of the mechanism on the overall performance of the network is not significant.
2.4 data acquisition method
The occurrence of "abnormal data" may be error data generated by a node failure, or may be data mutation caused by an event. As most of monitoring data of the sensor network are redundant data with low importance degree, the difference between the data is small, abnormal data is mainly represented as data abnormality of a few nodes, and extremely valuable abnormal data caused by event occurrence is represented as data abnormality of a plurality of nodes. Judging the type of the abnormal data by dividing the abnormal data proportion threshold, adopting a trust model-based acquisition mode for the former data abnormality, and starting an emergency data acquisition scheme for the latter data abnormality. Aiming at different data, the method disclosed by the invention is designed into different acquisition methods shown in an algorithm 1 to realize logic process coding, and the method comprises the following specific steps:
3 simulation and Performance analysis
In order to verify the accuracy and timeliness of algorithm data acquisition provided by the invention, MATLAB is used as a simulation tool to verify the relevant performance of the algorithm. Sensor nodes are randomly distributed in a 100 x 100 rectangular area, and the network is hierarchically divided based on the HEED protocol. The relevant simulation experiment parameter configuration is shown in table 1.
TABLE 1 simulation experiment parameters
3.1 Trust model Performance analysis
The judgment of the data abnormal type based on the trust degree is the basis of the algorithm of the invention, the trust degree model of the invention carries out modeling based on two dimensions of node interaction and data deviation, and the trust degree is utilized to distinguish whether the data abnormal node is normal or not. Therefore, whether abnormal nodes can be effectively detected is an important index for measuring the model. In combination with the data anomaly-oriented trust evaluation of the invention, the detection rate of the algorithm is defined as the proportion of the detected abnormal nodes in the total abnormal nodes. In order to verify the effectiveness of the trust model, 150 sensor nodes are set in the experiment, the network is hierarchically divided according to the HEED protocol, the detection rates of the algorithm of the invention and the classical trust model rfsn (routing based frame for sensor networks) are compared under different abnormal node occupation ratios, and the result is shown in fig. 4. With the increase of the proportion of abnormal nodes in the network, the detection rates of the algorithm and the RESN both show a descending trend, but the average detection rate of the algorithm is about 9.14 percent higher than that of the RFSN algorithm based on the data and the node quality.
In addition, when the percentage of the abnormal nodes is 15%, a cluster is randomly selected, and the mean value change trend of the credibility of the normal nodes and the credibility of the abnormal nodes in the cluster is detected. In the experimental process, abnormal nodes are randomly distributed in the network, normal nodes can accurately report data in the environment, all nodes in the cluster have the same attribute, and the initial trust level is set to be 0.5. The mean change of the confidence level of the nodes in the cluster in the first 50 sampling periods is selected in the experiment, and is shown in fig. 5. In the previous 15 times of sampling, the difference value of the reported data of the normal node and the abnormal node is large, so that the trust degree distinguishing effect is not obvious; along with the increase of the sampling times and the accumulation of trust punishment, the trust value of the abnormal node is gradually reduced, and the normal node always performs well. Therefore, the types of the abnormal data nodes can be well distinguished based on the trust model, and decision support is provided for data reliability collection.
3.2 network Performance analysis
In order to verify the data acquisition performance of the algorithm, the data acquisition accuracy and the network energy consumption are compared with the original HEED algorithm in a test mode. FIG. 6 is a comparison of data acquisition accuracy rates, a cluster simulation event is randomly selected during a test, normal node and abnormal node data packets are marked, and the percentage of the normal node data volume received by a Sink node in the total data volume is counted as a test result. Results are adopted for comparison in the first 20 times of test comparison, and the average accuracy of the HEED protocol data packet is about 71.75% under the influence of abnormal nodes. The algorithm of the invention adds an abnormal data detection mechanism to filter the abnormal node data packets, and the average accuracy rate is maintained at about 91.73% along with the increase of simulation rounds, and the data acquisition reliability is higher.
FIG. 7 is a graph of the total energy consumption of the algorithm nodes of the present invention. It can be seen from fig. 7 that the algorithm of the present invention has a lower total energy consumption than the native HEED algorithm in the same round robin cycle. In which, the HEED protocol reaches the maximum total network energy consumption (all nodes die) in 2200 rounds, while the network life lasts to 3200 rounds under the algorithm of the present invention. In addition, the algorithm of the invention introduces an emergency data transmission mechanism, avoids large-scale data transmission caused by emergency, balances network load, and causes the phenomenon of node death about 200 rounds later than the HEED protocol. The method has the advantages of higher energy utilization rate and longer network service life.
Through the technical scheme, the data reliability collection algorithm is designed by taking mutation data as a starting point based on the clustered WSN network topology structure and combining the event monitoring characteristics. And judging the abnormal type of the node according to the trust degree of the abnormal data node, and further adopting a corresponding acquisition mechanism. Simulation results show that the algorithm can accurately and timely process data in event monitoring.
Example 2
Corresponding to embodiment 1 of the present invention, embodiment 2 of the present invention further provides a wireless sensor network data acquisition system, where the system includes:
the abnormal node judgment module is used for screening the nodes with larger data value change acquired at the current moment and judging whether the nodes are abnormal nodes or not;
the data acquisition module is used for acquiring the abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low credibility and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
Specifically, the abnormal node judgment module is further configured to: setting a data window with the size of M based on the time correlation of the monitoring data, and regarding the node i, the data acquired at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3Is the upper quartile value of M-1 data, and IQR is equal to Q3-Q1。
Specifically, the data acquisition module is further configured to:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Indicating that a small proportion of outliers are present,data acquisition is carried out by adopting an acquisition mode based on a trust model; wherein the content of the first and second substances,
the scale threshold is represented, the value is 0.25 in the embodiment, and the size of the scale threshold is adjusted according to the actual situation in the practical application;
step 203: when in use
And (4) indicating that data abnormity occurs in most nodes, and starting an emergency data acquisition mechanism to acquire data.
More specifically, the step 202 includes:
acquiring the credibility of the cluster head to the abnormal node i;
acquiring the data offset of the abnormal node i;
acquiring the trust degree of the abnormal node i according to the trust degree of the cluster head to the abnormal node i and the data migration degree of the abnormal node i, and filtering the node data when the trust degree of the data is lower than a threshold lambda; and when the trust degree of the abnormal node i is greater than the threshold lambda, collecting the data of the node, and starting an emergency data collection mechanism to collect the data.
More specifically, the obtaining of the reliability of the cluster head to the abnormal node i includes:
by the formula
Obtaining the credibility D of the cluster head to the abnormal node iiWherein E () represents a probability distribution operator, αijIndicating a successful interaction alphaijSecond, betaijIndicating a failed interaction betaijNext, the process of the present invention,
represents the probability density distribution function of abnormal node i relative to cluster head node j and
for the probability of successful interaction, alpha represents the historical interaction success times of the abnormal node i, beta represents the historical interaction failure times of the cluster head node j, alpha is greater than 0, beta is greater than 0, and f () represents a gamma function.
More specifically, the obtaining the data offset degree of the abnormal node i includes:
for the abnormal node i, k neighbor nodes in the same cluster communication radius are selected, and the current data collected by the k neighbor nodes is set as ai1，ai2....aikAnd then the average value of the data of the neighboring nodes is:
by the formula Ti＝ω*|ai-Ei|+(1-ω)*|ai-EkI, acquiring the data offset of the abnormal node i, wherein omega is a first balance coefficient and belongs to [0, 1 ]]，aiIndicating that the abnormal node i is currently collecting data, EiAnd collecting historical mean values of data for the abnormal nodes i.
More specifically, the obtaining of the trust level of the abnormal node i according to the trust level of the cluster head on the abnormal node i and the data migration level of the abnormal node i includes:
by the formula
Obtaining the Trust degree of the abnormal node i, wherein TrustiIndicating the degree of trust of the abnormal node i,
is a second equilibrium coefficient and
the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for acquiring data of a wireless sensor network is characterized by comprising the following steps:
the method comprises the following steps: screening nodes with larger data value changes acquired at the current moment, and judging whether the nodes are abnormal nodes or not;
step two: acquiring an abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low reliability and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
2. The method for acquiring data of a wireless sensor network according to claim 1, wherein the first step comprises: setting a data window with the size of M based on the time correlation of the monitoring data, and regarding the node i, the data acquired at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3Is the upper quartile value of M-1 data, and IQR is equal to Q3-Q1。
3. The method for collecting data in a wireless sensor network according to claim 2, wherein the second step comprises:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Representing the appearance of a small part of abnormal values, and adopting a trust model-based acquisition mode to acquire data; wherein the content of the first and second substances,
represents a scale threshold;
step 203: when in use
And (4) indicating that data abnormity occurs in most nodes, and starting an emergency data acquisition mechanism to acquire data.
4. The method of claim 3, wherein the step 202 comprises:
acquiring the credibility of the cluster head to the abnormal node i;
acquiring the data offset of the abnormal node i;
acquiring the trust degree of the abnormal node i according to the trust degree of the cluster head to the abnormal node i and the data migration degree of the abnormal node i, and filtering the node data when the trust degree of the data is lower than a threshold lambda; and when the trust degree of the abnormal node i is greater than the threshold lambda, collecting the data of the node, and starting an emergency data collection mechanism to collect the data.
5. The method according to claim 4, wherein the acquiring the credibility of the cluster head to the abnormal node i comprises:
by the formula
Obtaining the credibility D of the cluster head to the abnormal node iiWherein E () represents a probability distribution operator, αijIndicating a successful interaction alphaijSecond, betaijIndicating a failed interaction betaijNext, the process of the present invention,
represents the probability density distribution function of abnormal node i relative to cluster head node j and
for the probability of successful interaction, alpha represents the historical interaction success times of the abnormal node i, beta represents the historical interaction failure times of the cluster head node j, alpha is greater than 0, beta is greater than 0, and f () represents a gamma function.
6. The method according to claim 4, wherein the obtaining the data offset of the abnormal node i comprises:
for the abnormal node i, k neighbors in the same cluster communication radius are selectedA living node, wherein the current data collected by k adjacent nodes is set as ai1，ai2....aikAnd then the average value of the data of the neighboring nodes is:
by the formula Ti＝ω*|ai-Ei|+(1-ω)*|ai-EkI, acquiring the data offset of the abnormal node i, wherein omega is a first balance coefficient and belongs to [0, 1 ]]，aiIndicating that the abnormal node i is currently collecting data, EiAnd collecting historical mean values of data for the abnormal nodes i.
7. The method for acquiring data of a wireless sensor network according to claim 4, wherein the obtaining of the credibility of the abnormal node i according to the credibility of the cluster head to the abnormal node i and the data migration degree of the abnormal node i comprises:
by the formula
Obtaining the Trust degree of the abnormal node i, wherein TrustiIndicating the degree of trust of the abnormal node i,
is a second equilibrium coefficient and
8. a wireless sensor network data acquisition system, the system comprising:
the abnormal node judgment module is used for screening the nodes with larger data value change acquired at the current moment and judging whether the nodes are abnormal nodes or not;
the data acquisition module is used for acquiring the abnormal data node proportion, and when the abnormal data node proportion is lower than a scale threshold value, filtering out node data with low credibility and then starting an emergency data acquisition mechanism to acquire data; when the data abnormal node occupation ratio is higher than a scale threshold value, starting an emergency data acquisition mechanism to acquire data;
the starting of the emergency data acquisition mechanism for data acquisition comprises: the current emergency cluster head node adopts a multilink transmission strategy and sends an emergency request to the downstream, and the request information comprises current cluster position information; after receiving the request message, the downstream cluster head node reduces the data acquisition frequency in the cluster, reduces the data forwarding of other cluster heads, and improves the data forwarding priority of the current emergency cluster head to ensure the data transmission of the current emergency cluster head; and after the abnormal data is acquired, each cluster head recovers a normal acquisition mechanism.
9. The system according to claim 8, wherein the abnormal node determining module is further configured to: setting a data window with the size of M based on the time correlation of the monitoring data, and regarding the node i, the data acquired at the current moment t is ai(t) and t is n × Δ t, Δ t is the sampling interval, n is the total number of samples, and the data acquired at the previous time, i.e., t-1 is (n-1) × Δ t, is ai(t-1), Cluster head utilizing a box model
Performing box type calculation on M-1 data in front of the node i and dividing a credible interval [ epsilon ]a，εb]If a isi(t) in [ epsilon ]a，εb]Judging the data to be abnormal data when the data deviation is too large outside the interval, and judging the node i to be an abnormal node; q1For the lower quartile value, Q, of M-1 data3Is the upper quartile value of M-1 data, and IQR is equal to Q3-Q1。
10. The wireless sensor network data acquisition system of claim 9, wherein the data acquisition module is further configured to:
step 201: by the formula
Acquiring an abnormal data node proportion theta, wherein N is the number of abnormal data nodes in the cluster counted by the cluster head at the current moment based on a box model, and NCHThe total number of member nodes in the cluster;
step 202: when in use
Representing the appearance of a small part of abnormal values, and adopting a trust model-based acquisition mode to acquire data; wherein the content of the first and second substances,
represents a scale threshold;
step 203: when in use
And (4) indicating that data abnormity occurs in most nodes, and starting an emergency data acquisition mechanism to acquire data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110186912.2A CN113015195B (en) | 2021-02-08 | 2021-02-08 | Wireless sensor network data acquisition method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110186912.2A CN113015195B (en) | 2021-02-08 | 2021-02-08 | Wireless sensor network data acquisition method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113015195A true CN113015195A (en) | 2021-06-22 |
| CN113015195B CN113015195B (en) | 2022-04-19 |
Family
ID=76402467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110186912.2A Active CN113015195B (en) | 2021-02-08 | 2021-02-08 | Wireless sensor network data acquisition method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113015195B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113709598A (en) * | 2021-08-31 | 2021-11-26 | 宁夏百川电力股份有限公司 | Carbon emission monitoring and management system based on Internet of things |
| CN114465911A (en) * | 2022-02-10 | 2022-05-10 | 成都阿普奇科技股份有限公司 | Internet of things sensing equipment resource unified description method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2096513A1 (en) * | 2008-02-28 | 2009-09-02 | Sap Ag | Trustworthiness assessment of sensor data from wireless sensor networks to business applications |
| US20120137367A1 (en) * | 2009-11-06 | 2012-05-31 | Cataphora, Inc. | Continuous anomaly detection based on behavior modeling and heterogeneous information analysis |
| CN104994535A (en) * | 2015-06-04 | 2015-10-21 | 浙江农林大学 | Sensor data flow abnormality detection method based on multidimensional data model |
| CN110198549A (en) * | 2019-06-10 | 2019-09-03 | 安徽理工大学 | Event driving type wireless sensor network cluster routing method and system |
| CN111083733A (en) * | 2020-02-10 | 2020-04-28 | 安徽理工大学 | Congestion control method and system for wireless sensor network |
| CN112019373A (en) * | 2020-07-10 | 2020-12-01 | 浙江工业大学 | Smart home security data acquisition method based on dynamic trust evaluation model |
| CN112165694A (en) * | 2020-11-02 | 2021-01-01 | 上海第二工业大学 | Method for establishing trust model of wireless sensor network |
-
2021
- 2021-02-08 CN CN202110186912.2A patent/CN113015195B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2096513A1 (en) * | 2008-02-28 | 2009-09-02 | Sap Ag | Trustworthiness assessment of sensor data from wireless sensor networks to business applications |
| US20120137367A1 (en) * | 2009-11-06 | 2012-05-31 | Cataphora, Inc. | Continuous anomaly detection based on behavior modeling and heterogeneous information analysis |
| CN104994535A (en) * | 2015-06-04 | 2015-10-21 | 浙江农林大学 | Sensor data flow abnormality detection method based on multidimensional data model |
| CN110198549A (en) * | 2019-06-10 | 2019-09-03 | 安徽理工大学 | Event driving type wireless sensor network cluster routing method and system |
| CN111083733A (en) * | 2020-02-10 | 2020-04-28 | 安徽理工大学 | Congestion control method and system for wireless sensor network |
| CN112019373A (en) * | 2020-07-10 | 2020-12-01 | 浙江工业大学 | Smart home security data acquisition method based on dynamic trust evaluation model |
| CN112165694A (en) * | 2020-11-02 | 2021-01-01 | 上海第二工业大学 | Method for establishing trust model of wireless sensor network |
Non-Patent Citations (6)
| Title |
|---|
| DYLAN COHEN等: "Trustability based on beta distribution detecting abnormal behaviour nodes in WSN", 《 2013 19TH ASIA-PACIFIC CONFERENCE ON COMMUNICATIONS (APCC)》 * |
| 叶正旺等: "基于信任模型的WSNs安全数据融合算法", 《东北大学学报( 自然科学版)》 * |
| 姚红燕等: "一种基于节点行为的无线传感器网络信誉计算模型", 《南通大学学报(自然科学版)》 * |
| 谭德坤等: "基于异常数据驱动的WSN簇内数据融合方法", 《传感技术学报》 * |
| 陈辉等: "一种基于卡尔曼滤波的分簇WSNs拥塞检测与控制方案", 《传感技术学报》 * |
| 黄力等: "无线传感器网络基于簇的数据传输可靠性研究", 《计算机应用与软件》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113709598A (en) * | 2021-08-31 | 2021-11-26 | 宁夏百川电力股份有限公司 | Carbon emission monitoring and management system based on Internet of things |
| CN114465911A (en) * | 2022-02-10 | 2022-05-10 | 成都阿普奇科技股份有限公司 | Internet of things sensing equipment resource unified description method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113015195B (en) | 2022-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111885012B (en) | Network situation perception method and system based on information acquisition of various network devices | |
| CN113015195B (en) | Wireless sensor network data acquisition method and system | |
| CN106844161B (en) | Abnormity monitoring and predicting method and system in calculation system with state flow | |
| CN107707431A (en) | The data safety monitoring method and system of a kind of facing cloud platform | |
| Rajasegarar et al. | Elliptical anomalies in wireless sensor networks | |
| CN110475224B (en) | Sensor data processing and collaborative prediction method based on edge calculation | |
| CN106878995B (en) | Method for identifying abnormal type of wireless sensor network based on perception data | |
| US9203857B2 (en) | Method and system for detecting anomaly of user behavior in a network | |
| Norouzi et al. | A new clustering protocol for wireless sensor networks using genetic algorithm approach | |
| Verma et al. | Data redundancy implications in wireless sensor networks | |
| Zhe et al. | DoS attack detection model of smart grid based on machine learning method | |
| Zawaideh et al. | An efficient weighted trust‐based malicious node detection scheme for wireless sensor networks | |
| Sabitha et al. | A Fuzzy Trust Evaluation of Cloud Collaboration Outlier Detection in Wireless Sensor Networks. | |
| CN108289285B (en) | Method for recovering and reconstructing lost data of ocean wireless sensor network | |
| CN110825545A (en) | Cloud service platform anomaly detection method and system | |
| Gharib et al. | Enhanced multiband multiuser cooperative spectrum sensing for distributed CRNs | |
| CN111093201A (en) | Wireless sensor network and clustering method thereof | |
| CN112149967A (en) | Power communication network vulnerability assessment method and system based on complex system theory | |
| Huang et al. | Hierarchical outlier detection approach for online distributed structural identification | |
| CN117527479B (en) | Soft bus networking connection method, device, equipment and storage medium | |
| CN107590008B (en) | A kind of method and system judging distributed type assemblies reliability by weighted entropy | |
| CN113765896A (en) | Internet of things implementation system and method based on artificial intelligence | |
| CN112437440A (en) | Malicious collusion attack resisting method based on correlation theory in wireless sensor network | |
| WO2023004899A1 (en) | Method and apparatus for detecting abnormal data of satellite and wireless communication convergence network performance | |
| Yao | Group-ordered SPRT for decentralized detection |
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 |