A digital tag deviation correction method and system based on electronic tags

Abstract

The present invention discloses a digital tag deviation correction method and system based on electronic tags, including real-time collection of three-dimensional position information of each electronic tag and comparison with a preset safe position, judging whether the tag drifts, and if the tag drifts, reconstructing its real motion path by using a Kalman filter algorithm, and constructing a feature vector in combination with environmental sensor data, using a support vector machine algorithm to analyze the feature vector, identifying potential causes of position drift, using a K-Means clustering algorithm to group abnormal tags, determining areas where deployment strategies need to be adjusted, and automatically adjusting tag density and position according to the frequency and amplitude of tag drift in the area, thereby improving positioning accuracy and reliability of drift detection. The present invention realizes accurate detection and cause identification of object position drift, and can adaptively adjust tag deployment strategies, thereby improving the efficiency of object positioning and management.

Description

Technical Field

[0001] The present invention relates to the technical field of data reading, and in particular to a method and system for correcting digital marking deviation based on electronic tags. Background Art

[0002] RFID UHF is a technology that uses radio waves for data transmission and identification. RFID UHF tags use radio waves for communication. The working principle is to attach the electronic tag to the target object and use the radio waves emitted by the reader for identification and tracking. When the reader emits radio waves, the electronic tag receives the radio waves and decodes them, and then sends the information back to the reader. After receiving the information, the reader performs further processing and application. RFID UHF technology provides efficient and convenient solutions for item management and tracking with its advantages of long-distance identification, high-speed data transmission, penetration, multi-tag reading and easy integration.

[0003] In traditional RFID UHF systems, due to factors such as the distance between the electronic tag and the reader and environmental interference, multipath effects, electromagnetic interference and reduced penetration occur, causing the location information of the electronic tag to drift, resulting in low object positioning accuracy and difficulty in identifying the cause of the drift.

[0004] The present invention solves the above-mentioned technical shortcomings of the prior art by:

[0005] By collecting the three-dimensional position information of each electronic tag in real time, comparing it with the three-dimensional information of the preset safe position, calculating the position offset, and judging whether the tag drifts through the threshold, the Kalman filter algorithm is used to reconstruct the real movement path of the tag that has drifted. Combined with the environmental information, the support vector machine algorithm is used to analyze the feature vector to identify the potential cause of the tag position offset. The K-Means clustering algorithm is used to group abnormal tags, determine the area where the deployment strategy needs to be adjusted, and automatically adjust the density and position of the tag in the determined area according to the frequency f and amplitude A of the tag, so as to achieve accurate detection and cause identification of the object position drift, and can adaptively adjust the tag deployment strategy, thereby improving the efficiency of object positioning and management. Summary of the invention

[0006] In order to solve the above-mentioned technical problems, the present invention provides a digital marking deviation correction method based on electronic tags.

[0007] The technical solution of the present invention is implemented as follows: a method for correcting digital marking deviation based on electronic tags, comprising:

[0008] S101, collecting three-dimensional position information of each tag in real time through a high-density electronic tag network to build a tag real-time position information database;

[0009] S102, reading the real-time position information of each tag from the tag real-time position information database, and reading the preset safety position three-dimensional coordinates of the corresponding tag from the pre-established tag preset safety position information database, and calculating the position offset;

[0010] S103, triggering an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment, and recording the real-time position information of the tag into an abnormal position information database;

[0011] S104, reading the drift trajectory data of the abnormal tag from the abnormal position information database, constructing the state vector X of the Kalman filter, and reconstructing the real motion path of the tag using the Kalman filter algorithm to obtain the reconstructed tag position information and speed information;

[0012] S105, combining the reconstructed tag position information, calculating the degree of tag position drift, and obtaining environmental information from environmental sensors, including temperature, humidity, and electromagnetic interference intensity;

[0013] S106, using a support vector machine algorithm to analyze the feature vector V to identify potential causes of label position drift;

[0014] S107, obtaining the spatial distribution information of abnormal labels from the abnormal position information database, combining the position drift reasons obtained by the support vector machine algorithm analysis, using the K-Means clustering algorithm to group the abnormal labels, using the silhouette coefficient to determine the optimal number of cluster centers K, clustering according to the optimal K value, and determining the area where the deployment strategy needs to be adjusted according to the number of labels in each cluster, the frequency and amplitude of drift;

[0015] S108. For the determined area, according to the frequency f and amplitude A of the label drift in the area, the density ρ of the label in the area is automatically adjusted, and the drift frequency threshold and the drift amplitude threshold are defined. If the position of the existing label needs to be adjusted, the new label position is calculated according to the frequency f and amplitude A of the label drift.

[0016] Furthermore, in step S101, receivers are deployed in the high-density electronic tag network area to receive electronic tag signals, and data is transmitted through the wireless communication network. The receivers collect the three-dimensional position information of each electronic tag in real time, and create a tag real-time position information database table in the database to store the three-dimensional information of each electronic tag collected in real time.

[0017] Further, in step S102, by pre-establishing a database table of tag preset safety position information, the preset safety position information is imported into the established database table of tag preset safety position information, and for each electronic tag, the position offset is calculated based on reading the real-time position three-dimensional coordinates and the preset safety position three-dimensional coordinates in the tag real-time database table and the tag preset safety position information database table at the same time.

[0018] Furthermore, in step S103, the calculated tag position offset is compared with a preset safety threshold. If the calculated tag position offset is greater than the preset safety threshold, an alarm is triggered and the real-time three-dimensional coordinate information of the tag is recorded in the abnormal position information database table.

[0019] Further, in step S104, the drift trajectory data of the abnormal tag is read from the abnormal position information database to construct the state vector of the Kalman filter. , the Kalman filter algorithm is used to reconstruct the true path of each tag, including continuous iterative prediction and update of each observation of the tag. The obtained state vector is the reconstructed tag position information and speed information.

[0020] Further, in step S105, the reconstructed tag position information is compared with the real-time tag position information to calculate the Euclidean distance between the two as the tag position offset degree, and the environmental temperature, humidity and electromagnetic interference intensity parameters are obtained from the environmental sensor.

[0021] Furthermore, in step S106, a feature vector V is constructed based on the tag abnormal position information data and the environmental sensor data, and the degree of tag position deviation is defined. The feature vector V is analyzed using a support vector machine algorithm to identify the potential causes of each tag position drift as environmental interference or human movement.

[0022] Furthermore, in step S107, the spatial distribution information of abnormal labels is obtained from the abnormal position information database, and the K-Means clustering algorithm is used to group the abnormal labels in combination with the position drift causes obtained by the support vector machine algorithm. The silhouette coefficient is used to determine the number of cluster centers K, and clustering is performed according to the optimal K value. According to the clustering results, the space is divided into different areas, and each area corresponds to a cluster. The area where the deployment strategy needs to be adjusted is determined according to the number of labels in each cluster, the frequency and amplitude of drift, and it is determined whether the strategy for adjusting the deployment in the area needs to increase the label density or reduce the label density.

[0023] Further, in step S108, for the determined area, according to the frequency f and amplitude A of the label drift in the area, the label drift frequency threshold and drift amplitude threshold ,like , increase the label density, add new labels in this area, if , reduce the label density, remove some labels, and if the position of the existing label needs to be adjusted, calculate the label offset vector d according to the frequency f and amplitude A of the label drift, and obtain the new label position through calculation to adjust the position of the existing label.

[0024] The system solution of the present invention is implemented as follows: a digital tag deviation correction system based on electronic tags, comprising:

[0025] The acquisition module is connected to the storage module and the transmission module, and is used to collect the three-dimensional position information of each electronic tag in real time, and collect the ambient temperature, humidity and electromagnetic interference intensity through the environmental sensor;

[0026] The storage module is connected to the acquisition module, the calculation and judgment module, the reconstruction calculation module, the identification module, the classification module, the correction module, and the transmission module, and is used to store the three-dimensional position information of each electronic tag acquired in real time, the preset safety position information, the abnormal position information, the reconstructed tag position information and speed information, and the drift frequency threshold and drift amplitude threshold of the tag;

[0027] The calculation and judgment module is connected to the storage module and the transmission module, and is used to calculate the tag position offset, and trigger an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment;

[0028] A reconstruction calculation module is connected to the storage module and the transmission module, and is used to reconstruct the tag position information and speed information of the abnormal tag position information data according to the real movement path of the tag, and calculate the degree of tag position drift in combination with the reconstructed tag position information;

[0029] The recognition module is connected to the storage module and the transmission module and is used to analyze the feature vector through the support vector machine algorithm to identify the potential causes of the tag position drift;

[0030] The classification module is connected to the storage module and the transmission module. It is used to combine the spatial distribution information of the abnormal labels and the causes of position drift, use the K-Means clustering algorithm to group the abnormal labels, and determine the areas where the deployment strategy needs to be adjusted;

[0031] The correction module is connected to the storage module and the transmission module, and is used to automatically adjust the density of the labels in the determined area according to the frequency f and amplitude A of the label drift;

[0032] The transmission module is connected to the acquisition module, the storage module, the calculation and judgment module, the reconstruction calculation module, the identification module, the classification module, and the correction module, and is used to transmit the three-dimensional position information of each tag collected in real time, the preset safety position information, the abnormal position information, the reconstructed position information and speed information, the drift frequency threshold and the drift amplitude threshold of the tag through the wireless communication network.

[0033] The present invention aims to solve the problem that the existing technology cannot solve the low positioning accuracy of objects in complex environments and the difficulty in identifying the drift causes. The present invention collects the three-dimensional position information of each electronic tag in real time through a high-density electronic tag network, compares it with the three-dimensional coordinates of the preset safe position of the tag, calculates the position offset, and triggers an alarm by comparing the calculated tag position offset with the preset safety threshold through threshold judgment, records the real-time position information of the tag that triggers the alarm due to the position offset as abnormal position information data, constructs the state vector of the Kalman filter using the drift trajectory data of the abnormal tag, uses the Kalman filter algorithm to reconstruct the real movement path of the tag, obtains the reconstructed tag position information and speed information, and combines the reconstructed tag position information with the The degree of tag position drift is calculated, and a feature vector is constructed by obtaining environmental information and abnormal position information. The support vector machine algorithm is used to analyze the feature vector to identify potential causes of tag position drift, including environmental interference and human movement. The K-Means clustering algorithm is used to group abnormal tags according to the cause of position drift, and the area where the deployment strategy needs to be adjusted is determined according to the number of tags in each cluster, the frequency and amplitude of drift. The label density and position are automatically adjusted according to the frequency and amplitude of regional label drift, which improves the positioning accuracy and the reliability of drift detection, realizes accurate detection and cause identification of object position drift, and can adaptively adjust the label deployment strategy, thereby improving the efficiency of object positioning and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a structural block diagram of a digital marking deviation correction method based on an electronic tag in an embodiment of the present invention;

[0035] Figure 2 This is a flowchart of a method for correcting a digital tag deviation based on an electronic tag in an embodiment of the present invention;

[0036] Figure 3 The present invention is a structural block diagram of a digital tag deviation correction system based on electronic tags in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0038] The preferred implementation methods of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these implementation methods are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0039] When used herein, the singular forms "a", "an", and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" and the like specify the presence of stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not exclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0040] See also Figure 1-Figure 2 As shown, a method for correcting digital marking deviation based on electronic tags includes:

[0041] S101, collecting three-dimensional position information of each tag in real time through a high-density electronic tag network to build a tag real-time position information database;

[0042] S102, reading the real-time position information of each tag from the tag real-time position information database, and reading the preset safety position three-dimensional coordinates of the corresponding tag from the pre-established tag preset safety position information database, and calculating the position offset;

[0043] S103, triggering an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment, and recording the real-time position information of the tag into an abnormal position information database;

[0044] S104, reading the drift trajectory data of the abnormal tag from the abnormal position information database, constructing the state vector X of the Kalman filter, and reconstructing the real motion path of the tag using the Kalman filter algorithm to obtain the reconstructed tag position information and speed information;

[0045] S105, combining the reconstructed tag position information, calculating the degree of tag position drift, and obtaining environmental information from environmental sensors, including temperature, humidity, and electromagnetic interference intensity;

[0046] S106, using a support vector machine algorithm to analyze the feature vector V to identify potential causes of label position drift;

[0047] S107, obtaining the spatial distribution information of abnormal labels from the abnormal position information database, combining the position drift reasons obtained by the support vector machine algorithm analysis, using the K-Means clustering algorithm to group the abnormal labels, using the silhouette coefficient to determine the optimal number of cluster centers K, clustering according to the optimal K value, and determining the area where the deployment strategy needs to be adjusted according to the number of labels in each cluster, the frequency and amplitude of drift;

[0048] S108. For the determined area, according to the frequency f and amplitude A of the label drift in the area, the density ρ of the label in the area is automatically adjusted, and the drift frequency threshold and the drift amplitude threshold are defined. If the position of the existing label needs to be adjusted, the new label position is calculated according to the frequency f and amplitude A of the label drift.

[0049] like Figure 1-Figure 2 As shown, in step S101, the three-dimensional position information of each tag is collected in real time through a high-density electronic tag network to build a tag real-time position information database;

[0050] Specifically, receivers are deployed in high-density electronic tag network areas to receive electronic tag signals and establish a wireless communication network for data transmission. The receivers collect the three-dimensional location information of each electronic tag in real time;

[0051] Create a tag real-time location information database table and create a unique primary key id for each record. Use the unique identifier id of the electronic tag as the unique primary key id in the tag real-time location information database table. After cleaning, add the three-dimensional location information data of each tag collected in real time to the tag real-time location information database table.

[0052] like Figure 1-Figure 2 As shown, in step S102, the real-time position information of each tag is read from the tag real-time position information database, and the preset safety position three-dimensional coordinates of the corresponding tag are read from the pre-established tag preset safety position information database to calculate the position offset;

[0053] Specifically, by pre-establishing a tag preset safety position information database, the preset safety position information data is imported into the established tag preset safety position information database table. For the same electronic tag, the unique identifier ID of the electronic tag is used to query the three-dimensional coordinates of the preset safety position of the tag in the tag preset safety position information database table through a subquery method, and at the same time, the real-time position three-dimensional coordinates of the tag are queried in the tag real-time position information database table through a subquery method.

[0054] The position offset is calculated by comparing the three-dimensional coordinates of the real-time position of the tag with the three-dimensional coordinates of the preset safety position of the corresponding tag, and calculating the Euclidean distance. Specifically, the three-dimensional coordinates of the real-time position of the tag and the three-dimensional coordinates of the preset safety position are subtracted in each coordinate dimension, and then the squares of the differences in each coordinate dimension are added and the square root is taken to obtain the distance from the real-time position point of the tag to the safety preset position point, which is used as the position offset s of the tag.

[0055] like Figure 1-Figure 2 As shown, in step S103, the calculated tag position offset and the preset safety threshold are judged by threshold value to trigger an alarm, and the real-time position information of the tag is recorded in the abnormal position information database;

[0056] Specifically, create an abnormal location information database table, and create a unique primary key id for each record. The unique identifier id of the electronic tag is used as the unique primary key id in the database table. The preset safety threshold distance is , the calculated label position offset s and the preset safety threshold Compare, if the calculated label offset s is greater than the safety threshold , an alarm is triggered, and the real-time location information of the tag is recorded in the abnormal location information database.

[0057] like Figure 1-Figure 2 As shown, in step S104, the drift trajectory data of the abnormal tag is read from the abnormal position information database, the state vector X of the Kalman filter is constructed, and the real motion path of the tag is reconstructed using the Kalman filter algorithm to obtain the reconstructed tag position information and speed information;

[0058] Specifically, the Kalman filter can predict the accurate 3D position and 3D velocity of a tag from a limited set of noisy data by observing the 3D position of the tag;

[0059] By reading the drift trajectory data of the abnormal label in the abnormal position information database, the state vector of the Kalman filter is constructed. ,in is the three-dimensional position of the label, is the three-dimensional velocity of the tag. According to the motion model of the tag, the state transfer matrix B, the observation matrix H, the process noise Q(k) and the measurement noise U(k) are defined. The state transfer matrix of the tag describes the law of the tag changing over time. The observation matrix describes the part directly observed from the state vector X. The process noise reflects the uncertainty in the state transfer process and is a diagonal matrix. The measurement noise is the uncertainty in the observation process.

[0060] The state transfer equation is constructed as:

[0061] ;

[0062] Among them, X(k) is the state vector at time k, B is the state transfer matrix, X(k-1) is the state vector at time k-1, and Q(k) is the process noise;

[0063] The observation equation is constructed as:

[0064] ;

[0065] Among them, Z(k) is the observed quantity, H is the observation matrix, X(k) is the state vector at time k, and U(k) is the measurement noise;

[0066] The Kalman filter algorithm is used to reconstruct the real motion path of the tag:

[0067] Use the state transfer matrix B and the state estimate at the previous moment to predict the state at the current moment;

[0068] Update the estimated value of the tag position at the current moment by using the observation matrix H, the observation value and the measurement noise U(k);

[0069] By continuously iterating the prediction and updating steps for each observed value of the tag until all the observed data are processed, the obtained state vector X is the reconstructed tag position information and speed information.

[0070] like Figure 1-Figure 2 As shown, in step S105, the tag position drift degree is calculated in combination with the reconstructed tag position information, and environmental information including temperature, humidity and electromagnetic interference intensity is obtained from the environmental sensor;

[0071] Specifically, the degree of drift of the tag position is calculated by combining the reconstructed tag position information, comparing the reconstructed tag position coordinate information with the corresponding real-time position three-dimensional coordinates, and calculating the Euclidean distance to measure the degree of drift of the tag position. Specifically, the reconstructed three-dimensional coordinates of the tag position are subtracted from each coordinate dimension of the three-dimensional coordinates of the real-time position of the tag, and then the squares of the differences in each coordinate dimension are added and the square root is taken to obtain the distance from the position point of the reconstructed tag to the real-time position point of the tag, which is the degree of drift of the tag position.

[0072] Temperature sensors, humidity sensors and electromagnetic interference sensors are deployed in the environment where the tag is located to obtain the temperature, humidity and electromagnetic interference intensity parameters of the environment.

[0073] like Figure 1-Figure 2 As shown, in step S106, the support vector machine algorithm is used to analyze the feature vector V to identify potential causes of label position drift;

[0074] Specifically, based on the abnormal position information data and environmental sensor data in the abnormal position information database, a feature vector V is constructed and the degree of tag drift is defined. The feature vector V is constructed by the temperature, humidity, and electromagnetic interference intensity of the tag environment. The degree of tag position drift is composed of the drift of the tag position and the non-drift of the tag position. The potential causes of each tag position drift include environmental interference or human movement.

[0075] The support vector machine algorithm is used to analyze the feature vector V, and a support vector machine model is constructed, a radial basis function kernel (RBF) is selected as the kernel function, and the parameters of the support vector machine model are determined by parameter setting, and a regularization parameter C and a maximum number of iterations are used, and a k value of a k-fold cross validation is selected, and k=10 is set;

[0076] Train the support vector machine model. Divide the data of feature vector V into k equal-sized subsets through k-fold cross validation. Repeat k times. Each time, use one subset as the test set and the remaining k-1 subsets as the training set. In each iteration, use the training set to train the support vector machine model. The accuracy is calculated by dividing the number of correctly identified samples by the total number of samples. In cross validation, divide the training set into k groups, train and test them alternately, and calculate the accuracy of each test.

[0077] The trained support vector machine model is used to analyze the feature vector V, analyze the weight of each feature, identify the feature that has the greatest impact on the result, and analyze the potential causes of position drift. The feature weight thresholds of temperature, humidity, and electromagnetic interference are preset as follows: , , If the feature weight of any one of the environmental factors, temperature, humidity or electromagnetic interference, is greater than the threshold of the preset feature weight, the potential cause of the position drift is identified as environmental interference; if the feature weights of the environmental factors, temperature, humidity and electromagnetic interference, are all lower than the threshold of the preset feature weight, the cause of the position drift is identified as human movement.

[0078] like Figure 1-Figure 2 As shown, in step S107, the spatial distribution information of abnormal labels is obtained from the abnormal position information database, and the position drift causes obtained by the support vector machine algorithm analysis are combined, and the K-Means clustering algorithm is used to group the abnormal labels, and the silhouette coefficient is used to determine the optimal number of cluster centers K, and clustering is performed according to the optimal K value, and the area where the deployment strategy needs to be adjusted is determined according to the number of labels in each cluster, the frequency and amplitude of drift;

[0079] Specifically, the spatial distribution information of abnormal labels is obtained from the abnormal location information database, and the position drift reasons obtained by the support vector machine algorithm are combined. The K-Means clustering algorithm is used to group the abnormal labels. Specifically, K abnormal label data are read from the abnormal location information database table as the cluster centers of K classes, and other data are classified into the class closest to themselves to obtain K cluster groups, and the center point of each cluster is recalculated, and then the classification is continued according to the center point, and the steps of iterative cluster grouping and recalculating the center point of each cluster are repeated until the center point no longer changes or changes very little;

[0080] Use the silhouette coefficient to determine the number of cluster centers K, and perform clustering based on the optimal K value. For each data point, calculate the average distance a from the point to other points in the same cluster, which is the intra-class distance. The smaller a is, the more likely the data point should be clustered into the cluster. Then calculate the minimum value b of the average distance from the point to all points in other clusters, which is the inter-class distance.

[0081] Calculate the silhouette coefficient L using the following formula:

[0082] L = (ba) / max(a,b);

[0083] Among them, a is the intra-class distance, b is the inter-class distance, max(a,b) is the larger number of a and b, and the larger the silhouette coefficient L, the better the clustering effect;

[0084] The closer L is to 1, the more reasonable the clustering is; the closer L is to -1, the more likely it is to be classified into another cluster; and the closer L is to 0, the more likely it is to be classified into the boundary between two clusters.

[0085] According to the clustering results, the space is divided into different areas. Each area corresponds to a cluster. The area where the deployment strategy needs to be adjusted is determined based on the number of labels in each cluster, the frequency and amplitude of drift, and whether the strategy for adjusting the deployment in the area needs to increase or reduce the label density.

[0086] like Figure 1-Figure 2 As shown, in step S108, for the determined area, according to the frequency f and amplitude A of the label drift in the area, the density ρ of the label in the area is automatically adjusted, and the drift frequency threshold and the drift amplitude threshold are defined. If the position of the existing label needs to be adjusted, the new label position is calculated according to the frequency f and amplitude A of the label drift;

[0087] Specifically, for a determined area, the density ρ of the label in the area is automatically adjusted according to the frequency f and amplitude A of the label drift in the area, where the frequency f of the label drift is the number of times the label position deviation exceeds the preset threshold per unit time, and the amplitude A of the label drift is the maximum value of the label position deviation;

[0088] The density ρ of the label in the region is automatically adjusted to define the drift frequency threshold and drift amplitude threshold , if the drift frequency is greater than the drift frequency threshold, that is Or the drift amplitude is greater than the drift amplitude threshold, that is, , indicating that the label drift is serious, then increase the label density ρ, add new labels in this area, if the drift frequency is less than half of the drift frequency threshold and the drift amplitude is less than half of the drift amplitude threshold, that is, , indicating that the label drift is stable, then reduce the label density ρ and remove some labels;

[0089] If the position of the existing tag needs to be adjusted, the drift vector d of the tag is calculated according to the frequency f and amplitude A of the existing tag drift. The direction of the drift vector d indicates the direction of the tag movement, and the magnitude indicates the distance the tag moves. The position of the existing tag is preset to p, and the new tag position is calculated as:

[0090]

[0091] Where p' represents the new tag position, p represents the existing tag position, f represents the frequency of existing tag drift, and A represents the amplitude of existing tag drift;

[0092] By calculating the new label position, the position of the existing label is adjusted.

[0093] See also Figure 3 As shown, a digital marking deviation correction system based on electronic tags includes:

[0094] The acquisition module is connected to the storage module and the transmission module, and is used to collect the three-dimensional position information of each electronic tag in real time, and collect the ambient temperature, humidity and electromagnetic interference intensity through the environmental sensor;

[0095] The storage module is connected to the acquisition module, the calculation and judgment module, the reconstruction calculation module, the identification module, the classification module, the correction module, and the transmission module, and is used to store the three-dimensional position information of each electronic tag acquired in real time, the preset safety position information, the abnormal position information, the reconstructed tag position information and speed information, and the drift frequency threshold and drift amplitude threshold of the tag;

[0096] The calculation and judgment module is connected to the storage module and the transmission module, and is used to calculate the tag position offset, and trigger an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment;

[0097] A reconstruction calculation module is connected to the storage module and the transmission module, and is used to reconstruct the tag position information and speed information of the abnormal tag position information data according to the real movement path of the tag, and calculate the degree of tag position drift in combination with the reconstructed tag position information;

[0098] The recognition module is connected to the storage module and the transmission module and is used to analyze the feature vector through the support vector machine algorithm to identify the potential causes of the tag position drift;

[0099] The classification module is connected to the storage module and the transmission module. It is used to combine the spatial distribution information of the abnormal labels and the causes of position drift, use the K-Means clustering algorithm to group the abnormal labels, and determine the areas where the deployment strategy needs to be adjusted;

[0100] The correction module is connected to the storage module and the transmission module, and is used to automatically adjust the density of the labels in the determined area according to the frequency f and amplitude A of the label drift;

[0101] The transmission module is connected to the acquisition module, the storage module, the calculation and judgment module, the reconstruction calculation module, the identification module, the classification module, and the correction module, and is used to transmit the three-dimensional position information of each tag collected in real time, the preset safety position information, the abnormal position information, the reconstructed position information and speed information, the drift frequency threshold and the drift amplitude threshold of the tag through the wireless communication network.

[0102] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and rules of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for correcting digital marking deviation based on electronic tags, characterized in that: The specific steps include: S101, collecting three-dimensional position information of each tag in real time through a high-density electronic tag network to build a tag real-time position information database; S102, reading the real-time position information of each tag from the tag real-time position information database, and reading the preset safety position three-dimensional coordinates of the corresponding tag from the pre-established tag preset safety position information database, and calculating the position offset; S103, triggering an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment, and recording the real-time position information of the tag into an abnormal position information database; S104, reading the drift trajectory data of the abnormal tag from the abnormal position information database, constructing the state vector X of the Kalman filter, and reconstructing the real motion path of the tag using the Kalman filter algorithm to obtain the reconstructed tag position information and speed information; S105, combining the reconstructed tag position information, calculating the degree of tag position drift, and obtaining environmental information from environmental sensors, including temperature, humidity, and electromagnetic interference intensity; S106, using a support vector machine algorithm to analyze the feature vector V to identify potential causes of label position drift; S107, obtaining the spatial distribution information of abnormal labels from the abnormal position information database, combining the position drift reasons obtained by the support vector machine algorithm analysis, using the K-Means clustering algorithm to group the abnormal labels, using the silhouette coefficient to determine the optimal number of cluster centers K, clustering according to the optimal K value, and determining the area for adjusting the deployment strategy according to the number of labels in each cluster, the frequency and amplitude of drift; S108. For the determined area, according to the frequency f and amplitude A of the label drift in the area, the density ρ of the label in the area is automatically adjusted, and the drift frequency threshold and the drift amplitude threshold are defined. If the position of the existing label needs to be adjusted, the new label position is calculated according to the frequency f and amplitude A of the label drift.

2. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S101, the three-dimensional position information of each tag is collected in real time through the high-density electronic tag network to build a real-time position information database of the tag, and a receiver is deployed in the high-density electronic tag network area to receive the electronic tag signal and establish a wireless communication network for data transmission; Create a tag real-time location information database, use the unique identifier id of the electronic tag as the unique primary key id in the tag real-time location information database table, and add the three-dimensional location information data of each tag collected in real time to the tag real-time location information database.

3. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S102, the position offset is calculated. For the same electronic tag, the Euclidean distance between the real-time position three-dimensional coordinates and the corresponding preset safety position three-dimensional coordinates is calculated by reading the real-time position information from the tag real-time position information database and reading the corresponding preset safety position three-dimensional coordinates from the tag preset safety position information database, and the distance is recorded as the position offset of the tag.

4. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S103, the calculated tag position offset and the preset safety threshold are compared through threshold judgment to trigger an alarm. The calculated tag position offset is compared with the preset safety threshold. If the tag position offset is greater than the safety threshold, an alarm is triggered, an abnormal position information database is created, and the real-time position information of the tag is recorded in the abnormal position information database.

5. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S104, the Kalman filter algorithm is used to reconstruct the real motion path of the tag, and the state vector of the Kalman filter is constructed by reading the drift trajectory data of the abnormal tag in the abnormal position information database. , where (x, y, z) is the three-dimensional position of the label, is the three-dimensional velocity of the tag. According to the tag's motion model, the state transfer matrix B, the observation matrix H, the process noise Q(k), and the measurement noise U(k) are defined; Use the state transfer matrix B and the state estimate at the previous moment to predict the state at the current moment; Use the observation matrix H, observation values ​​and measurement noise U(k) to update the estimated value of the tag position at the current moment; By continuously iterating the prediction and updating steps for each observed value of the tag until all the observed data are processed, the obtained state vector X is the reconstructed tag position information and speed information.

6. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S105, the tag position drift degree is calculated by combining the reconstructed tag position information, comparing the reconstructed tag position coordinate information with the corresponding real-time position three-dimensional coordinates, and calculating the Euclidean distance as the tag position drift degree.

7. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S106, the potential causes of tag position drift are identified, and a feature vector V is constructed and a tag drift degree is defined based on the abnormal position information data and the environmental sensor data in the abnormal position information database. The feature vector V is constructed by the temperature, humidity, and electromagnetic interference intensity of the tag environment, and the tag position drift degree is composed of the tag position drifting and the tag position not drifting. Use the support vector machine model to analyze the feature vector V, analyze the weight of each feature, analyze the feature that has the greatest impact on position drift, and identify the potential causes of position drift; If the feature weight of any one of the environmental factors, temperature, humidity or electromagnetic interference, is greater than a preset threshold, the potential cause of the position drift is identified as environmental interference; If the feature weights of temperature, humidity and electromagnetic interference in the environmental factors are all lower than the preset threshold, it is identified that the cause of the position drift is human movement.

8. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S107, the position drift cause obtained by combining the support vector machine algorithm analysis is used, and the K-Means clustering algorithm is used to group the abnormal labels, and the spatial distribution information of the abnormal labels is obtained from the abnormal position information database. The position drift cause obtained by the support vector machine algorithm is combined, and the K-Means clustering algorithm is used to group the abnormal labels, and the silhouette coefficient is used to determine the number of cluster centers K, and clustering is performed according to the optimal K value. According to the clustering results, the space is divided into different areas, each area corresponds to a cluster, and the area where the deployment strategy needs to be adjusted is determined according to the number of labels in each cluster, the frequency and amplitude of drift, and it is determined whether the strategy for adjusting the deployment in the area needs to increase the label density or reduce the label density.

9. The method for correcting digital marking deviation based on electronic tags according to claim 1, characterized in that: In step S108, the density ρ of the region label is automatically adjusted. For the determined region, the drift frequency threshold is defined according to the frequency f and amplitude A of the region label drift. and drift amplitude threshold ; If the drift frequency is greater than the drift frequency threshold, Or the drift amplitude is greater than the drift amplitude threshold, that is, , then increase the label density ρ and add new labels in this area; If the drift frequency is less than half of the drift frequency threshold and the drift amplitude is less than half of the drift amplitude threshold, that is, , then reduce the label density ρ and remove some labels; If the position of an existing tag needs to be adjusted, the new tag position is calculated based on the frequency f and amplitude A of the tag drift.

10. A digital tag deviation correction system based on electronic tags, characterized in that: The acquisition module is connected to the storage module and the transmission module, and is used to collect the three-dimensional position information of each electronic tag in real time, and collect the ambient temperature, humidity and electromagnetic interference intensity through the environmental sensor; The storage module is connected to the acquisition module, the calculation and judgment module, the reconstruction module, the recognition module, the classification module, the correction module, and the transmission module, and is used to store the three-dimensional position information of each electronic tag acquired in real time, the preset safety position information, the abnormal position information, the reconstructed tag position information and speed information, and the drift frequency threshold and drift amplitude threshold of the tag; The calculation and judgment module is connected to the storage module and the transmission module, and is used to calculate the tag position offset, and trigger an alarm by comparing the calculated tag position offset with a preset safety threshold through a threshold judgment; A reconstruction calculation module is connected to the storage module and the transmission module, and is used to reconstruct the tag position information and speed information of the abnormal tag position information data according to the real movement path of the tag, and calculate the degree of tag position drift in combination with the reconstructed tag position information; The recognition module is connected to the storage module and the transmission module and is used to analyze the feature vector through the support vector machine algorithm to identify the potential causes of the tag position drift; The classification module is connected to the storage module and the transmission module. It is used to combine the spatial distribution information of the abnormal labels and the causes of position drift, use the K-Means clustering algorithm to group the abnormal labels, and determine the areas where the deployment strategy needs to be adjusted; The correction module is connected to the storage module and the transmission module, and is used to automatically adjust the density of the labels in the determined area according to the frequency f and amplitude A of the label drift; The transmission module is connected to the acquisition module, the storage module, the calculation and judgment module, the reconstruction calculation module, the identification module, the classification module, and the correction module, and is used to transmit the three-dimensional position information of each tag collected in real time, the preset safety position information, the abnormal position information, the reconstructed position information and speed information, the drift frequency threshold and the drift amplitude threshold of the tag through the wireless communication network.

Images