Anti-counterfeiting verification tracking method and system based on RFID technology

Abstract

The invention discloses an anti-counterfeiting verification tracking method and system based on an RFID technology, and relates to the technical field of anti-counterfeiting tracking. All unexpected nodes in a second sequence are identified, historical data of the unexpected nodes are analyzed, then a confidence score of each unexpected node is obtained, an influence weight of each unexpected node is obtained according to a correlation index, and the confidence score of each unexpected node is obtained; calculating confidence scores and influence weights of all unexpected nodes through a fusion algorithm, generating a verification coefficient for a product, generating a corresponding management strategy for the product according to a comparison result of the verification coefficient and a verification threshold value, and regularly obtaining an unexpected node identification number of all products in the same series. And analyzing whether the same series of products need to be managed in advance. The tracking system recognizes all unexpected nodes through sequence comparison and performs historical data analysis on the unexpected nodes, so that path abnormity can be effectively found, the recognition capability of the system on abnormal circulation is enhanced, and the tracking precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of anti-counterfeiting tracking technology, and in particular to an anti-counterfeiting verification and tracking method and system based on RFID technology. Background Art

[0002] The globalization trend of modern supply chains has exacerbated the emergence of counterfeit and shoddy goods, especially in the fields of medicine, food, electronic products and high-end luxury goods. The harm of counterfeit products is particularly prominent, which not only harms the interests of consumers, but also damages the reputation of brand owners. The tracking system aims to use radio frequency identification (RFID) technology to realize the identity recognition, authenticity verification and full tracking of goods, prevent the circulation of counterfeit and shoddy goods, and improve the transparency and efficiency of supply chain management.

[0003] The prior art has the following defects:

[0004] Traditional tracking systems are mainly based on comparisons of predefined standard paths to identify whether products deviate from normal circulation paths. However, tracking systems can often only detect whether products have reached unexpected nodes. They lack in-depth path analysis and are unable to identify the real risks of these unexpected nodes. It is also difficult to determine which nodes may represent serious supply chain anomalies or counterfeit risks. This rough path anomaly detection mechanism may lead to frequent false alarms or fail to issue timely warnings in actual risk situations.

[0005] Based on this, the present invention proposes an anti-counterfeiting verification and tracking method and system based on RFID technology, which can effectively discover path anomalies by identifying all unexpected nodes through sequence comparison and performing historical data analysis on them, thereby enhancing the system's ability to identify abnormal circulation and improve tracking accuracy. Summary of the invention

[0006] The purpose of the present invention is to provide an anti-counterfeiting verification and tracking method and system based on RFID technology to solve the shortcomings of the background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: an anti-counterfeiting verification and tracking method based on RFID technology, the tracking method comprising the following steps:

[0008] The tracking system scans the RFID tag of the tracked product, marks the standard path of the product as a first sequence, obtains the actual circulation path of the product, marks the actual circulation path as a second sequence, performs sequence comparison on the first sequence and the second sequence, and identifies all unexpected nodes in the second sequence;

[0009] After analyzing the historical data of unexpected nodes, the confidence score of each unexpected node is obtained. If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is marked on the RFID tag of the product. Otherwise, the tracking system obtains the impact weight of each unexpected node based on the correlation index;

[0010] After the confidence scores and impact weights of all unexpected nodes are calculated through a fusion algorithm, a verification coefficient is generated for the product. Based on the comparison result between the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product. After regularly obtaining the number of unexpected nodes identified for all products in the same series, it is analyzed whether the same series of products needs to be managed in advance, and the analysis results are sent to the administrator.

[0011] In a preferred embodiment, after analyzing the historical data of the unexpected nodes, obtaining the confidence score of each unexpected node includes the following steps:

[0012] Obtain historical data of unexpected nodes, including the occurrence frequency, normal marking ratio, abnormal marking ratio and position deviation index of the unexpected nodes;

[0013] The confidence score of the unexpected node is obtained by comprehensively calculating the occurrence frequency, normal mark ratio, abnormal mark ratio and position deviation index. The expression is: , where is the confidence score, is the frequency of occurrence, is the normal marking ratio, is the abnormal labeling ratio, is the position deviation index, is the adjustment coefficient, , are the weights of normal label ratio and abnormal label ratio respectively, and .

[0014] In a preferred embodiment, the confidence scores and influence weights of all unexpected nodes are calculated by a fusion algorithm to generate a verification coefficient for the product, including the following steps:

[0015] After calculating the confidence scores and impact weights of all unexpected nodes through the fusion algorithm, a verification coefficient is generated for the product. The expression is: , where is the verification coefficient, For the The influence weight of unexpected nodes, For the The confidence scores of the unexpected nodes, is the number of unexpected nodes in the product material path;

[0016] The obtained verification coefficient is compared with the preset verification threshold. The verification threshold is used to determine whether the product has a risk of counterfeiting. If the verification coefficient is less than or equal to the verification threshold, it is determined that the product has no risk of counterfeiting. If the verification coefficient is greater than the verification threshold, it is determined that the product has a risk of counterfeiting.

[0017] In a preferred embodiment, after periodically obtaining the number of unexpected node identifications for all products in the same series, analyzing whether it is necessary to manage the products in the same series in advance includes the following steps:

[0018] Create a product set for the same series of products, obtain the number of unexpected nodes identified by the RFID tag of each product in the product set, and calculate the mean and standard deviation of the number of unexpected nodes identified by all products. The expression is: , where is the quantity mean, is the quantity standard deviation, is the number of products in the same series, For the Each product identifies the unexpected number of nodes, and the management value is obtained by dividing the mean of the number by the standard deviation of the number;

[0019] The management value of the same series of products is compared with the preset management threshold. The management threshold is used to determine whether it is necessary to manage the same series of products in advance. If the management value is greater than the management threshold, it is determined that the same series of products needs to be managed in advance. If the management value is less than or equal to the management threshold, it is determined that it is not necessary to manage the same series of products in advance.

[0020] In a preferred embodiment, the tracking system obtains the influence weight of each unexpected node according to the correlation index, including the following steps:

[0021] The logic for obtaining the correlation index is as follows: obtain the number of subsequent nodes of the unexpected node, and use the number of subsequent nodes of the unexpected node as the correlation index. The larger the correlation index of the unexpected node, the greater the relative importance of the current unexpected node in the logistics path. The total correlation index is obtained by summing the correlation indexes of all unexpected nodes, and the correlation index of the unexpected node is divided by the total correlation index to obtain the influence weight of the unexpected node.

[0022] In a preferred embodiment, if the confidence score of any unexpected node of a product is lower than a score threshold, a warning mark is marked on the RFID tag of the product, including the following steps:

[0023] The confidence scores of all unexpected nodes in the product are compared with the preset score threshold. The score threshold is used to determine whether there is an abnormality in the unexpected node. If the confidence score of the unexpected node is less than the score threshold, it is determined that the unexpected node is abnormal, and a warning mark is placed on the RFID tag of the product. If the confidence score of the unexpected node is greater than or equal to the score threshold, it is determined that there is no abnormality in the unexpected node. The tracking system obtains the influence weight of each unexpected node based on the correlation index.

[0024] In a preferred embodiment, the calculation logic of the position deviation index is: obtaining the geographical coordinates of the unexpected node, the geographical coordinates of the preceding node and the coordinates of the succeeding node, and calculating the Euclidean distance between the unexpected node and the preceding node, and the expression is: , where is the Euclidean distance between the unexpected node and the preceding node, are the geographic coordinates of the unexpected nodes, is the geographic coordinate of the preceding node, and the Euclidean distance between the unexpected node and the succeeding node is calculated. The expression is: , where is the Euclidean distance between the unexpected node and the subsequent node, are the geographic coordinates of the unexpected nodes, is the geographic coordinates of the post-node, then the position deviation index .

[0025] In a preferred embodiment, the calculation logic of the normal marking ratio is: obtain the historical normal marking times and abnormal marking times of unexpected nodes, sum the normal marking times and the abnormal marking times to obtain the total marking times, and divide the normal marking times by the total marking times to obtain the normal marking ratio;

[0026] The calculation logic of the abnormal marking ratio is as follows: obtaining the historical normal marking times and abnormal marking times of unexpected nodes, summing the normal marking times and the abnormal marking times to obtain the total marking times, and dividing the abnormal marking times by the total marking times to obtain the abnormal marking ratio;

[0027] The calculation logic of the occurrence frequency is: obtain the total number of times unexpected nodes appear in historical data, obtain the total number of times all nodes appear in historical data, and divide the total number of times unexpected nodes appear in historical data by the total number of times all nodes appear in historical data to obtain the occurrence frequency.

[0028] In a preferred embodiment, performing sequence alignment on the first sequence and the second sequence to identify all unexpected nodes in the second sequence comprises the following steps:

[0029] Select the dynamic programming algorithm and initialize the comparison matrix. Start from the starting nodes of the standard path and the actual path, and compare each pair of nodes one by one. If the nodes in the two paths match, continue to compare the next pair of nodes. If not, mark the node as an unexpected node.

[0030] For all the unmatched nodes that appear in the comparison, these nodes are recorded as unexpected nodes. Unexpected nodes refer to nodes in the actual path that do not appear in the standard path, that is, abnormal points in the circulation path;

[0031] The position of each unexpected node in the second sequence is determined and recorded in the comparison result. The position information is used for subsequent risk analysis and path adjustment. A sequence comparison result report is generated, which contains a list of all unexpected nodes and their positions in the actual path.

[0032] The anti-counterfeiting verification and tracking system based on RFID technology includes a marking module, a node identification module, a calculation module, and a management module;

[0033] Marking module: scans the RFID tag of the tracked product, marks the standard path of the product as the first sequence, obtains the actual circulation path of the product, and marks the actual circulation path as the second sequence;

[0034] Node identification module: perform sequence comparison on the first sequence and the second sequence, identify all unexpected nodes in the second sequence, analyze the historical data of the unexpected nodes, and obtain the confidence score of each unexpected node;

[0035] Calculation module: If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is placed on the RFID tag of the product. Otherwise, the influence weight of each unexpected node is obtained according to the correlation index, and the confidence scores and influence weights of all unexpected nodes are calculated through a fusion algorithm to generate a verification coefficient for the product.

[0036] Management module: Generates corresponding management strategies for products based on the comparison results of verification coefficient and verification threshold, regularly obtains the number of unexpected node identifications for all products in the same series, and analyzes whether it is necessary to manage the same series of products in advance.

[0037] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0038] 1. The present invention compares the first sequence and the second sequence, identifies all unexpected nodes in the second sequence, analyzes the historical data of the unexpected nodes, obtains the confidence score of each unexpected node, obtains the influence weight of each unexpected node based on the correlation index, and calculates the confidence scores and influence weights of all unexpected nodes through a fusion algorithm to generate a verification coefficient for the product. According to the comparison result of the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product. After regularly obtaining the number of unexpected nodes identified for all products in the same series, it analyzes whether it is necessary to manage the products in the same series in advance. The tracking system identifies all unexpected nodes through sequence comparison and analyzes their historical data, which can effectively discover path anomalies, thereby enhancing the system's ability to identify abnormal circulation and improve tracking accuracy;

[0039] 2. The present invention establishes a product set based on the same series of products, obtains the number of unexpected nodes identified by the RFID tag of each product in the product set (regardless of whether the confidence score of the unexpected node is greater than or equal to the score threshold or less than the score threshold, it is recorded), calculates the number mean and the number standard deviation based on the number of unexpected nodes identified by all products (reflecting the degree of fluctuation in the number of unexpected nodes identified between different products), and obtains the management value by dividing the number mean by the number standard deviation. The larger the management value, the more unexpected nodes are identified by the whole series of products. Based on the comparison result of the management value and the preset management threshold, it is determined whether it is necessary to manage the same series of products in advance, thereby effectively improving the management efficiency of the same series of products by the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1: Please refer to Figure 1As shown, the anti-counterfeiting verification and tracking method based on RFID technology described in this embodiment includes the following steps:

[0044] The tracking system scans the RFID tag of the tracked product, marks the standard path of the product as the first sequence, obtains the actual circulation path of the product, marks the actual circulation path as the second sequence, compares the first sequence and the second sequence, identifies all unexpected nodes in the second sequence, and obtains the confidence score of each unexpected node after analyzing the historical data of the unexpected nodes. If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is placed on the RFID tag of the product. Otherwise, the tracking system obtains the influence weight of each unexpected node based on the association index, and calculates the confidence scores and influence weights of all unexpected nodes through a fusion algorithm to generate a verification coefficient for the product. According to the comparison result of the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product. After regularly obtaining the number of unexpected node identifications for all products in the same series, it is analyzed whether the same series of products needs to be managed in advance.

[0045] This application compares the first sequence and the second sequence, identifies all unexpected nodes in the second sequence, analyzes the historical data of the unexpected nodes, obtains the confidence score of each unexpected node, obtains the influence weight of each unexpected node based on the correlation index, and calculates the confidence scores and influence weights of all unexpected nodes through a fusion algorithm to generate a verification coefficient for the product. According to the comparison result of the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product. After regularly obtaining the number of unexpected node identifications for all products in the same series, it analyzes whether it is necessary to manage the same series of products in advance. The tracking system identifies all unexpected nodes through sequence comparison and analyzes their historical data, which can effectively discover path anomalies, thereby enhancing the system's ability to identify abnormal circulation and improve tracking accuracy.

[0046] Embodiment 2: The tracking system scans the RFID tag of the tracked product, marks the standard path of the product as a first sequence, obtains the actual circulation path of the product, and marks the actual circulation path as a second sequence, including the following steps:

[0047] The tracking system scans the RFID tag of the tracked product, reads the standard path of the product, and marks it as the first sequence. The standard path is the ideal circulation path preset by the system, which usually includes all expected logistics nodes (such as factories, warehouses, distribution centers, retail stores, etc.).

[0048] Actual path acquisition: The tracking system also obtains the actual circulation path of the product through continuous scanning and marks the actual circulation path as the second sequence. The actual path is the real path that the product goes through during the logistics process, including all time and location data.

[0049] Assume that a high-end watch of a certain brand is affixed with a unique RFID tag when it leaves the factory. The system presets a standard circulation path for this watch. This path starts from the manufacturing factory, passes through the regional warehouse, the brand retail store, and finally reaches the consumer. This standard path is marked as the first sequence:

[0050] The first sequence (standard path): manufacturing plant → regional warehouse → brand retail store.

[0051] In actual circulation, the tracking system obtains the circulation path of the watch through RFID scanning and records it as the second sequence:

[0052] The second sequence (actual path): manufacturing plant → regional warehouse → temporary storage center → brand retail store.

[0053] Performing a sequence alignment on the first sequence and the second sequence to identify all unexpected nodes in the second sequence includes the following steps:

[0054] Select a dynamic programming algorithm and initialize an alignment matrix to compare each node in the first sequence (standard path) with the second sequence (actual path).

[0055] Starting from the starting nodes of the standard path and the actual path, each pair of nodes is compared one by one. If the nodes in the two paths match, the next pair of nodes is compared; if not, the node is marked as a potential unexpected node.

[0056] For all the mismatched nodes that appear in the alignment, these nodes are recorded as “unexpected nodes”. These nodes represent nodes in the actual path that do not appear in the standard path, that is, abnormal points in the circulation path.

[0057] The position of each unexpected node in the second sequence is determined and recorded in the comparison result. The position information is used for subsequent risk analysis and path adjustment.

[0058] Generate a sequence alignment report containing a list of all unexpected nodes and their locations in the actual path. This report provides a basis for further risk analysis and is used to determine the product's anti-counterfeiting status or for circulation management.

[0059] Assume that the high-end watches of this brand are affixed with a unique RFID tag when they leave the factory, and the standard circulation path (first sequence) of the watch is:

[0060] First sequence (standard path):

[0061] Manufacturing plants;

[0062] Regional warehouses;

[0063] Branded retail stores;

[0064] consumer;

[0065] During the actual circulation process, the tracking system records the actual path of the watch (second sequence) as follows:

[0066] Second sequence (actual path):

[0067] Manufacturing plants;

[0068] Regional warehouses;

[0069] temporary storage centres;

[0070] Branded retail stores;

[0071] consumer;

[0072] Initialize the alignment algorithm: The system selects the dynamic programming algorithm as the alignment tool and establishes an alignment matrix to gradually compare the nodes of the first sequence and the second sequence.

[0073] Compare nodes step by step:

[0074] Compare the first node: manufacturing plant, match.

[0075] Compare the second node: regional warehouse, match.

[0076] Compare the third node: the node in the first sequence is a brand retail store, while the node in the second sequence is a temporary storage center, so there is a mismatch.

[0077] Identification of unexpected nodes: The temporary storage center is identified as an unexpected node because it is not in the standard path.

[0078] Locating the unexpected node position: The system records the position of the temporary storage center in the second sequence, which is the third node.

[0079] Output comparison result: The comparison result indicates that the unexpected node is a temporary storage center, located at the third position of the actual path.

[0080] Result: Through comparison, the system found that the watch passed through an unexpected node temporary storage center during the circulation process. This anomaly may indicate a non-standard storage process. The tracking system will further analyze the historical data of the node to assess the risk and, if necessary, mark the RFID tag with a warning mark to remind subsequent processes to pay attention to the circulation status of the watch.

[0081] After analyzing the historical data of unexpected nodes, the confidence score of each unexpected node is obtained, including the following steps:

[0082] Obtain historical data of unexpected nodes, including the occurrence frequency, normal marking ratio, abnormal marking ratio and position deviation index of the unexpected nodes;

[0083] The confidence score of the unexpected node is obtained by comprehensively calculating the occurrence frequency, normal mark ratio, abnormal mark ratio and position deviation index. The expression is: , where is the confidence score, is the frequency of occurrence, is the normal marking ratio, is the abnormal labeling ratio, is the position deviation index, is the adjustment coefficient, which is used to adjust the influence of the position deviation index on the confidence score. , are the weights of normal label ratio and abnormal label ratio respectively, and .

[0084] The larger the confidence score of the unexpected node, the more normal the unexpected node is in the current logistics path. If the unexpected node appears for the first time, the confidence score is 0, indicating that the unexpected node is unreliable.

[0085] The calculation logic of the occurrence frequency is: obtain the total number of times the unexpected node appears in the historical data, obtain the total number of times all nodes appear in the historical data, and divide the total number of times the unexpected node appears in the historical data by the total number of times all nodes appear in the historical data to obtain the occurrence frequency. The higher the occurrence frequency, the more frequently the unexpected node appears in the product's logistics path.

[0086] The frequency of unexpected nodes is used to correct the normal label ratio or abnormal label ratio. When the calculated value is positive, it indicates that the normal label ratio is greater than the abnormal label ratio. At this time, the higher the frequency of unexpected nodes, the greater the confidence score. When the calculated value is negative, it indicates that the normal mark ratio is smaller than the abnormal mark ratio. At this time, the higher the frequency of occurrence of unexpected nodes, the smaller the confidence score. In addition, the normal mark ratio and the abnormal mark ratio of unexpected nodes may be equal. Generally speaking, in order to improve the anti-counterfeiting verification of products, the weight of the abnormal mark ratio is greater than the weight of the normal mark ratio, that is, when the normal mark ratio and the abnormal mark ratio are equal, The calculated value is negative.

[0087] The calculation logic of the normal marking ratio is as follows: obtain the historical normal marking times and abnormal marking times of the unexpected node, sum the normal marking times and the abnormal marking times to obtain the total marking times, and divide the normal marking times by the total marking times to obtain the normal marking ratio. The larger the normal marking ratio, the more likely it is that the unexpected node was legal in history.

[0088] The calculation logic of the abnormal marking ratio is as follows: obtain the historical normal marking times and abnormal marking times of the unexpected node, sum the normal marking times and the abnormal marking times to obtain the total marking times, and divide the abnormal marking times by the total marking times to obtain the abnormal marking ratio. The larger the abnormal marking ratio, the greater the probability that the unexpected node was considered abnormal in history.

[0089] The calculation logic of the position deviation index is: obtain the geographical coordinates of the unexpected node, the geographical coordinates of the preceding node, and the coordinates of the following node (the preceding node is the previous node of the unexpected node, that is, the product is transported to the unexpected node through the preceding node, and the following node is the next node of the unexpected node, that is, the product is transported to the following node through the unexpected node), and calculate the Euclidean distance between the unexpected node and the preceding node. The expression is: , where is the Euclidean distance between the unexpected node and the preceding node, are the geographic coordinates of the unexpected nodes, is the geographic coordinate of the preceding node, and the Euclidean distance between the unexpected node and the succeeding node is calculated. The expression is: , where is the Euclidean distance between the unexpected node and the subsequent node, are the geographic coordinates of the unexpected nodes, is the geographic coordinates of the post-node, then the position deviation index .

[0090] Normal marking ratio:

[0091] High ratio (large value): A high normal mark ratio indicates that the node has been considered normal in history, has a relatively high credibility, and therefore has a high confidence score.

[0092] Low ratio (small value): A low normal label ratio indicates that the occurrence of this node is often regarded as abnormal, thereby reducing the confidence score.

[0093] Relationship: The higher the normal label ratio, the higher the confidence score, which can be expressed as a proportional relationship.

[0094] Abnormal marking ratio:

[0095] High ratio (large value): A high anomaly labeling ratio indicates that the node is often marked as anomaly in historical data, which means that the node has a higher risk, thereby reducing the confidence score.

[0096] Low ratio (small value): A low anomaly label ratio indicates that the node is rarely considered an anomaly and may be a deviation from a legitimate node, so the confidence score is high.

[0097] Relationship: The higher the abnormal label ratio, the lower the confidence score, which can be expressed as an inverse relationship.

[0098] Position deviation index:

[0099] High deviation index (large value): When the position deviation index is high, it means that the unexpected node is far away from both the previous node and the next node, which may mean that the node is abnormally located or that the node moves beyond the normal path range. Therefore, the confidence score is usually low, indicating that the node is highly abnormal.

[0100] Low deviation index (small value): When the position deviation index is low, it means that the unexpected node is close to the preceding node and the succeeding node, indicating that the node is near the expected path. The deviation may be due to temporary reasons. The confidence score is high, indicating that the node is relatively reliable.

[0101] If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is placed on the RFID tag of the product. Otherwise, the tracking system obtains the impact weight of each unexpected node based on the correlation index, including the following steps:

[0102] The confidence scores of all unexpected nodes in the product are compared with the preset score threshold. The score threshold is used to determine whether there is an abnormality in the unexpected node. If the confidence score of the unexpected node is less than the score threshold, it is determined that the unexpected node is abnormal, and a warning mark is placed on the RFID tag of the product. If the confidence score of the unexpected node is greater than or equal to the score threshold, it is determined that there is no abnormality in the unexpected node. The tracking system obtains the influence weight of each unexpected node based on the correlation index.

[0103] The tracking system obtains the influence weight of each unexpected node based on the correlation index, including the following steps:

[0104] The logic for obtaining the correlation index is as follows: obtain the number of subsequent nodes of the unexpected node, and use the number of subsequent nodes of the unexpected node as the correlation index. The larger the correlation index of the unexpected node, the greater the relative importance of the current unexpected node in the logistics path. The total correlation index is obtained by summing the correlation indexes of all unexpected nodes, and the correlation index of the unexpected node is divided by the total correlation index to obtain the influence weight of the unexpected node.

[0105] After the confidence scores and impact weights of all unexpected nodes are calculated through the fusion algorithm, a verification coefficient is generated for the product. Based on the comparison result between the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product, including the following steps:

[0106] After calculating the confidence scores and impact weights of all unexpected nodes through the fusion algorithm, a verification coefficient is generated for the product. The expression is: , where is the verification coefficient, For the The influence weight of unexpected nodes, For the The confidence scores of the unexpected nodes, It is the number of unexpected nodes in the product material path. The larger the verification coefficient is, the greater the risk of product counterfeiting is.

[0107] The obtained verification coefficient is compared with the preset verification threshold. The verification threshold is used to determine whether the product has a risk of counterfeiting. If the verification coefficient is less than or equal to the verification threshold, it is determined that the product has no risk of counterfeiting. If the verification coefficient is greater than the verification threshold, it is determined that the product has a risk of counterfeiting.

[0108] 1) Management strategy when judging that there is no risk of counterfeiting of products:

[0109] Release strategy: Allow the product to continue to circulate in the supply chain and update the tracking information and node status in the system.

[0110] Path optimization: The current node of the product is recorded in the path information as a normal path node to provide a reference for subsequent circulation path optimization of similar products.

[0111] Trust Update: Improve the trust of the product and its current circulation node, and mark it as "normal circulation" in the history record.

[0112] Regular inspection and adjustment: Set a longer inspection cycle for the product in the system to reduce the frequency of repeated inspections and reduce the system burden.

[0113] 2) Management strategies when judging that products have counterfeit risks:

[0114] Risk marking and isolation: Mark the product as "counterfeit risk" in the system, isolate or lock the product, and stop its circulation in the supply chain.

[0115] Further investigation: Initiate a detailed manual or automated investigation process to analyze the abnormal nodes in the product's distribution path to confirm the cause of the counterfeit or abnormality.

[0116] Notify relevant parties: Send early warning notifications to key stakeholders in the supply chain (such as warehouses, retailers or manufacturers) to alert them to the potential risks of the product.

[0117] Trace the source: Trace back based on abnormal node information to determine whether it is necessary to review the upstream links involved to prevent the spread of counterfeit products.

[0118] Customized Tracking Solution: Generate a high-frequency tracking and verification plan for the product to ensure that any anomalies can be identified and responded to in a timely manner during further circulation.

[0119] After regularly obtaining the number of unexpected node identifications for all products in the same series, analyze whether it is necessary to manage the same series of products in advance, including the following steps:

[0120] Generally speaking, the entire logistics path of a product is generally pre-set through the product management platform. If the number of unexpected node identifications for all products in the same series is greater, it not only indicates that the product may be counterfeit risk, but also indicates that the product logistics path management may have poor performance, for example:

[0121] Unclear or unstable route planning: The logistics route is not clearly preset or changes frequently, resulting in multiple unexpected nodes. This may be due to frequent changes in transporters and storage warehouses in different regions, or the lack of standardization of product distribution processes in different regions.

[0122] Imperfect supervision process: Multiple nodes in the logistics chain are not strictly supervised, and no reasonable inspection mechanism has been established, resulting in the inability to timely discover and respond to abnormal circulation nodes. For example, the lack of real-time upload of node information or the lack of verification process increases the probability of unexpected nodes appearing.

[0123] System update lag: The path data of the product management platform is not updated in a timely manner, causing the actual logistics nodes to be inconsistent with the paths set in the system. Unexpected nodes may occur due to failure to enter the latest paths into the system.

[0124] Chaotic multi-channel distribution management: Lack of coordination in multi-channel sales management may lead to the generation of unexpected nodes in different distribution channels. In particular, unauthorized distribution paths are prone to occur when irregular third-party logistics companies or authorized dealers are improperly managed.

[0125] Large differences in regional distribution standards: Distribution standards are not uniform in different regions, and non-standardized transportation or storage operations may be adopted in some areas, resulting in unplanned transfer nodes or temporary storage nodes, deviating from the preset path.

[0126] Missing or delayed information transmission: The communication between the logistics node and the platform is unstable, and the information transmission is not timely, resulting in a lag in path status information, or the passing information of a specific node is not uploaded to the management platform in a timely manner, affecting the timely verification of the path node.

[0127] Therefore, in order to help enterprises further understand the actual situation of the logistics routes of the same series of products, and to make corresponding management in response to the actual situation of the logistics routes, improve management efficiency and avoid waste of resources, we propose the following solutions:

[0128] Create a product set for the same series of products, obtain the number of unexpected nodes identified by the RFID tag of each product in the product set (regardless of whether the confidence score of the unexpected node is greater than or equal to the score threshold or less than the score threshold, it is recorded), and calculate the mean number and standard deviation of the number of unexpected nodes identified by all products (reflecting the degree of fluctuation in the number of unexpected nodes identified between different products). The expression is: , where is the quantity mean, is the quantity standard deviation, is the number of products in the same series, For the Products identified an unexpected number of nodes;

[0129] The management value is obtained by dividing the mean of the quantity by the standard deviation of the quantity. The larger the management value, the more unexpected nodes are identified by the overall product series.

[0130] The management value of the same series of products is compared with the preset management threshold. The management threshold is used to determine whether it is necessary to manage the same series of products in advance. If the management value is greater than the management threshold, it is determined that the same series of products needs to be managed in advance. If the management value is less than or equal to the management threshold, it is determined that it is not necessary to manage the same series of products in advance.

[0131] This application establishes a product set for the same series of products, obtains the number of unexpected nodes identified by the RFID tag of each product in the product set (regardless of whether the confidence score of the unexpected node is greater than or equal to the score threshold or less than the score threshold, it is recorded), calculates the number mean and the number standard deviation based on the number of unexpected nodes identified by all products (reflecting the degree of fluctuation in the number of unexpected nodes identified between different products), and obtains the management value by dividing the number mean by the number standard deviation. The larger the management value, the more unexpected nodes are identified by the whole series of products. Based on the comparison result of the management value and the preset management threshold, it is judged whether it is necessary to manage the same series of products in advance, thereby effectively improving the management efficiency of the company for the same series of products.

[0132] Example 3: Please refer to Figure 1As shown, the anti-counterfeiting verification and tracking system based on RFID technology described in this embodiment includes a marking module, a node identification module, a calculation module, and a management module;

[0133] Marking module: scans the RFID tag of the tracked product, marks the standard path of the product as the first sequence, obtains the actual circulation path of the product, marks the actual circulation path as the second sequence, and sends the first sequence and second sequence information to the node identification module;

[0134] Node identification module: performs sequence comparison on the first sequence and the second sequence, identifies all unexpected nodes in the second sequence, analyzes the historical data of the unexpected nodes, obtains the confidence score of each unexpected node, and sends the confidence score to the calculation module;

[0135] Calculation module: If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is placed on the RFID tag of the product. Otherwise, the influence weight of each unexpected node is obtained according to the correlation index, and the confidence scores and influence weights of all unexpected nodes are calculated through a fusion algorithm to generate a verification coefficient for the product, which is sent to the management module.

[0136] Management module: Generates corresponding management strategies for products based on the comparison results of verification coefficient and verification threshold, regularly obtains the number of unexpected node identifications for all products in the same series, and analyzes whether it is necessary to manage the same series of products in advance.

[0137] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0138] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0139] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. Anti-counterfeiting verification and tracking method based on RFID technology, characterized by: The tracking method comprises the following steps: The tracking system scans the RFID tag of the tracked product, marks the standard path of the product as a first sequence, obtains the actual circulation path of the product, marks the actual circulation path as a second sequence, performs sequence comparison on the first sequence and the second sequence, and identifies all unexpected nodes in the second sequence; After analyzing the historical data of unexpected nodes, the confidence score of each unexpected node is obtained. If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is marked on the RFID tag of the product. Otherwise, the tracking system obtains the impact weight of each unexpected node based on the correlation index; After the confidence scores and impact weights of all unexpected nodes are calculated through a fusion algorithm, a verification coefficient is generated for the product. Based on the comparison result between the verification coefficient and the verification threshold, a corresponding management strategy is generated for the product. After regularly obtaining the number of unexpected nodes identified for all products in the same series, it is analyzed whether the same series of products needs to be managed in advance, and the analysis results are sent to the administrator.

2. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 1 is characterized in that: After analyzing the historical data of unexpected nodes, the confidence score of each unexpected node is obtained, including the following steps: Obtain historical data of unexpected nodes, including the occurrence frequency, normal marking ratio, abnormal marking ratio and position deviation index of the unexpected nodes; The confidence score of the unexpected node is obtained by comprehensively calculating the occurrence frequency, normal mark ratio, abnormal mark ratio and position deviation index. The expression is: , where is the confidence score, is the frequency of occurrence, is the normal marking ratio, is the abnormal labeling ratio, is the position deviation index, is the adjustment coefficient, , are the weights of normal label ratio and abnormal label ratio respectively, and .

3. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 2 is characterized in that: After calculating the confidence scores and impact weights of all unexpected nodes through a fusion algorithm, a verification coefficient is generated for the product, including the following steps: After calculating the confidence scores and impact weights of all unexpected nodes through the fusion algorithm, a verification coefficient is generated for the product. The expression is: , where is the verification coefficient, For the The influence weight of unexpected nodes, For the The confidence scores of the unexpected nodes, is the number of unexpected nodes in the product material path; The obtained verification coefficient is compared with the preset verification threshold. The verification threshold is used to determine whether the product has a risk of counterfeiting. If the verification coefficient is less than or equal to the verification threshold, it is determined that the product has no risk of counterfeiting. If the verification coefficient is greater than the verification threshold, it is determined that the product has a risk of counterfeiting.

4. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 3 is characterized in that: After regularly obtaining the number of unexpected node identifications for all products in the same series, analyze whether it is necessary to manage the same series of products in advance, including the following steps: Create a product set for the same series of products, obtain the number of unexpected nodes identified by the RFID tag of each product in the product set, and calculate the mean and standard deviation of the number of unexpected nodes identified by all products. The expression is: , where is the quantity mean, is the quantity standard deviation, is the number of products in the same series, For the Each product identifies the unexpected number of nodes, and the management value is obtained by dividing the mean of the number by the standard deviation of the number; The management value of the same series of products is compared with the preset management threshold. The management threshold is used to determine whether it is necessary to manage the same series of products in advance. If the management value is greater than the management threshold, it is determined that the same series of products needs to be managed in advance. If the management value is less than or equal to the management threshold, it is determined that it is not necessary to manage the same series of products in advance.

5. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 4 is characterized in that: The tracking system obtains the influence weight of each unexpected node based on the correlation index, including the following steps: The logic for obtaining the correlation index is as follows: obtain the number of subsequent nodes of the unexpected node, and use the number of subsequent nodes of the unexpected node as the correlation index. The larger the correlation index of the unexpected node, the greater the relative importance of the current unexpected node in the logistics path. The total correlation index is obtained by summing the correlation indexes of all unexpected nodes, and the correlation index of the unexpected node is divided by the total correlation index to obtain the influence weight of the unexpected node.

6. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 5 is characterized in that: If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is marked on the RFID tag of the product, including the following steps: The confidence scores of all unexpected nodes in the product are compared with the preset score threshold. The score threshold is used to determine whether there is an abnormality in the unexpected node. If the confidence score of the unexpected node is less than the score threshold, it is determined that the unexpected node is abnormal, and a warning mark is placed on the RFID tag of the product. If the confidence score of the unexpected node is greater than or equal to the score threshold, it is determined that there is no abnormality in the unexpected node. The tracking system obtains the influence weight of each unexpected node based on the correlation index.

7. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 6 is characterized in that: The calculation logic of the position deviation index is: obtain the geographical coordinates of the unexpected node, the geographical coordinates of the preceding node, and the coordinates of the succeeding node, and calculate the Euclidean distance between the unexpected node and the preceding node. The expression is: , where is the Euclidean distance between the unexpected node and the preceding node, are the geographic coordinates of the unexpected nodes, is the geographic coordinate of the preceding node, and the Euclidean distance between the unexpected node and the succeeding node is calculated. The expression is: , where is the Euclidean distance between the unexpected node and the subsequent node, are the geographic coordinates of the unexpected nodes, is the geographic coordinates of the post-node, then the position deviation index .

8. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 7 is characterized in that: The calculation logic of the normal marking ratio is as follows: obtaining the historical normal marking times and abnormal marking times of unexpected nodes, summing the normal marking times and the abnormal marking times to obtain the total marking times, and dividing the normal marking times by the total marking times to obtain the normal marking ratio; The calculation logic of the abnormal marking ratio is as follows: obtaining the historical normal marking times and abnormal marking times of unexpected nodes, summing the normal marking times and the abnormal marking times to obtain the total marking times, and dividing the abnormal marking times by the total marking times to obtain the abnormal marking ratio; The calculation logic of the occurrence frequency is: obtain the total number of times unexpected nodes appear in historical data, obtain the total number of times all nodes appear in historical data, and divide the total number of times unexpected nodes appear in historical data by the total number of times all nodes appear in historical data to obtain the occurrence frequency.

9. The anti-counterfeiting verification and tracking method based on RFID technology according to claim 8, characterized in that: Performing a sequence alignment on the first sequence and the second sequence to identify all unexpected nodes in the second sequence includes the following steps: Select the dynamic programming algorithm and initialize the comparison matrix. Start from the starting nodes of the standard path and the actual path, and compare each pair of nodes one by one. If the nodes in the two paths match, continue to compare the next pair of nodes. If not, mark the node as an unexpected node. For all the unmatched nodes that appear in the comparison, these nodes are recorded as unexpected nodes. Unexpected nodes refer to nodes in the actual path that do not appear in the standard path, that is, abnormal points in the circulation path; The position of each unexpected node in the second sequence is determined and recorded in the comparison result. The position information is used for subsequent risk analysis and path adjustment. A sequence comparison result report is generated, which contains a list of all unexpected nodes and their positions in the actual path.

10. An anti-counterfeiting verification and tracking system based on RFID technology, used to solve the tracking method described in any one of claims 1 to 9, characterized in that: It includes a marking module, a node identification module, a calculation module, and a management module; Marking module: scans the RFID tag of the tracked product, marks the standard path of the product as the first sequence, obtains the actual circulation path of the product, and marks the actual circulation path as the second sequence; Node identification module: perform sequence comparison on the first sequence and the second sequence, identify all unexpected nodes in the second sequence, analyze the historical data of the unexpected nodes, and obtain the confidence score of each unexpected node; Calculation module: If the confidence score of any unexpected node of the product is lower than the score threshold, a warning mark is placed on the RFID tag of the product. Otherwise, the influence weight of each unexpected node is obtained according to the correlation index, and the confidence scores and influence weights of all unexpected nodes are calculated through a fusion algorithm to generate a verification coefficient for the product. Management module: Generates corresponding management strategies for products based on the comparison results of verification coefficient and verification threshold, regularly obtains the number of unexpected node identifications for all products in the same series, and analyzes whether it is necessary to manage the same series of products in advance.

Images