A drug manufacturer monitoring system and method based on the Internet of Things

Abstract

The present invention relates to the field of Internet of Things monitoring technology, and in particular to a drug manufacturer monitoring system and method based on the Internet of Things, including: an Internet of Things node management module for classifying and marking Internet of Things nodes; a node self-organizing module for self-organizing Internet of Things nodes; a node self-repairing module for self-repairing Internet of Things nodes; an information collection module for collecting drug manufacturer information; an information transmission module for transmitting drug manufacturer information; an information verification module for verifying the transmitted drug manufacturer information; a monitoring and analysis module for outputting abnormal monitoring and analysis results; an intelligent decision-making module for outputting intelligent decisions; and a decision execution module for executing intelligent decisions. The present invention improves the monitoring efficiency of drug manufacturers and the security of drug manufacturer monitoring information by self-organizing and self-repairing Internet of Things nodes and encrypting and transmitting drug manufacturer monitoring information.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things monitoring technology, and in particular to a drug manufacturer monitoring system and method based on the Internet of Things. Background Art

[0002] When monitoring pharmaceutical manufacturers, traditional IoT nodes lack self-organization and self-repair capabilities. Once a node fails, the stability and reliability of the entire monitoring system will be seriously affected. Monitoring information is usually not encrypted during transmission, which makes data easy to be illegally intercepted or tampered with. Information security is low. When abnormal situations occur, the existing system often relies on manual intervention and lacks autonomous decision-making and control capabilities, resulting in low monitoring efficiency and long response time.

[0003] China Patent Publication No.: CN104899712B discloses a cold chain logistics monitoring system based on ZigBee technology, including at least one sensor node module, at least one handheld terminal and at least one coordinator node module; at least one sensor node module is wirelessly connected to a coordinator node module, and the coordinator node module is connected to the handheld terminal via a data line; at least one sensor node module and the coordinator node module form a ZigBee wireless communication network. However, the IoT nodes in this solution lack self-organization and self-repair functions. Once a node fails, the stability and reliability of the entire monitoring system will be seriously affected, and the monitoring efficiency of pharmaceutical manufacturers cannot be guaranteed. Summary of the invention

[0004] To this end, the present invention provides a pharmaceutical manufacturer monitoring system and method based on the Internet of Things, which is used to overcome the problems in the prior art of low pharmaceutical manufacturer monitoring efficiency and low security of pharmaceutical manufacturer monitoring information due to the lack of self-organization and self-repair of IoT nodes, failure to encrypt and transmit pharmaceutical manufacturer monitoring information, and failure to make autonomous decisions and control on abnormal situations.

[0005] To achieve the above objectives, on the one hand, the present invention provides a drug manufacturer monitoring system based on the Internet of Things, comprising:

[0006] An IoT node management module, used to classify and mark IoT nodes according to node service information, and also used to adjust the process of classifying and marking IoT nodes according to newly added IoT nodes;

[0007] A node self-organizing module is used to self-organize IoT nodes, optimize the self-organizing process according to the remaining power of each IoT node, and adjust the self-organizing optimization process according to the fluctuation properties of IoT nodes;

[0008] Node self-repair module, used to self-repair IoT nodes;

[0009] Information collection module, used to collect information about drug manufacturers through IoT nodes;

[0010] Information transmission module, used to transmit information about drug manufacturers;

[0011] Information verification module, used to verify the transmitted drug manufacturer information;

[0012] The monitoring and analysis module is used to perform abnormal monitoring and analysis on the verified drug manufacturer information, output the abnormal monitoring and analysis results, and calibrate the abnormal monitoring and analysis results according to the active setting signal;

[0013] Intelligent decision-making module, used to make decisions on drug manufacturers based on abnormal monitoring and analysis results, and output intelligent decisions based on the decision-making results;

[0014] The decision execution module is used to execute intelligent decisions.

[0015] Furthermore, the IoT node management module inputs the node service information into the TF-IDF vectorizer, and the TF-IDF vectorizer performs fitting conversion on the node service information, and converts the node service information into a TF-IDF feature matrix X, and calculates the Euclidean distance d between each IoT node and each cluster according to the TF-IDF feature matrix X of each IoT node. K， Setting K = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the example, the cluster corresponding to the minimum Euclidean distance is selected as the preset cluster of the IoT node, each IoT node is classified into its corresponding preset cluster, and the preset clustering objective function J is calculated. k is the total number of clusters, Ci is the set of the i-th cluster, and μi is the centroid of the i-th cluster.

[0016] Furthermore, the IoT node management module compares the preset clustering objective function J with the preset marking objective function J0, and classifies and marks each IoT node according to the comparison result, wherein:

[0017] When J≤J0, the preset clustering of each IoT node is used as the classification of each IoT node and marked;

[0018] When J>J0, the centroid of each cluster is adjusted, and the Euclidean distance d between each IoT node and each cluster is recalculated based on the TF-IDF feature matrix X of each IoT node. K， SettingK = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the process, the cluster corresponding to the minimum Euclidean distance is reselected as the preset cluster of the IoT node until the preset cluster objective function J is less than or equal to the preset marking objective function J0. The preset cluster of each IoT node is used as the classification of each IoT node and marked. If the preset cluster objective function J is still greater than the preset marking objective function J0 when the number of iterations n reaches the preset number of iterations n0, the preset cluster of each IoT node in the nth iteration is used as the classification of each IoT node and marked.

[0019] Further, the IoT node management module calculates the IoT node access time difference △T, sets △T=TA-T0, TA is the IoT access time, T0 is the current time, and compares the IoT node access time difference △T with the preset IoT node access time difference △T0, and judges the newly added IoT node according to the comparison result, wherein:

[0020] When △T>△T0, the IoT node management module determines that the IoT node is not a newly added IoT node;

[0021] When △T≤△T0, the Internet of Things node management module determines that the Internet of Things node is a newly added Internet of Things node and adjusts the node service information. The adjusted node service information is an information set after adding the newly added Internet of Things node service information. The Internet of Things node management module classifies and marks the newly added Internet of Things nodes according to the adjusted node service information.

[0022] Furthermore, the node self-organizing module sets the protocol of each IoT node as a distributed network protocol, compares the polling time interval TP with the preset polling time interval TP0, and judges the cluster head node of each cluster according to the comparison result, wherein:

[0023] When TP≤TP0, the node self-organizing module determines the cluster head node of each cluster as the current cluster head node;

[0024] When TP>TP0, the node self-organizing module determines the cluster head node of each cluster as the next cluster head node.

[0025] Furthermore, when the drug manufacturer is powered off, the node self-organizing module compares the remaining power A of each IoT node with the preset power A0, and judges the energy mode of each IoT node according to the comparison result, and optimizes the order of cluster head nodes of each cluster according to the judgment result, wherein:

[0026] When A≥A0, the node self-organizing module determines that the IoT node is in a high energy mode;

[0027] When A<A0, the node self-organizing module determines that the Internet of Things node is in low energy mode, optimizes the order of cluster head nodes of each cluster, removes the Internet of Things node from the order of cluster head nodes of the corresponding cluster, and uses the order of cluster head nodes of the corresponding cluster after removal as the order of cluster head nodes of each cluster after optimization.

[0028] Furthermore, the node self-organizing module compares the offline times P of each IoT node with the preset offline times P0, and judges the fluctuation attribute of the IoT node according to the comparison result, wherein:

[0029] When P≤P0, the node self-organizing module determines that the IoT node is a non-fluctuating node;

[0030] When P>P0, the node self-organizing module determines that the Internet of Things node is a fluctuating node, and adjusts the order of cluster head nodes of each cluster after optimization, and removes the Internet of Things node from the order of cluster head nodes of each cluster after optimization.

[0031] Furthermore, the node self-repair module obtains the heartbeat message of each IoT node in each heartbeat cycle to obtain a heartbeat message collection, and polls and compares each node ID in the heartbeat message collection with a preset node ID. After the polling and comparison, the missing node ID is obtained, and the missing node ID is used as a faulty IoT node. The faulty IoT node is removed from the cluster head node sequence of each cluster after optimization, self-repair is performed according to the distributed network protocol, and the location information of the faulty IoT node is sent to the user terminal.

[0032] Furthermore, the information collection module collects information about drug manufacturers through the Internet of Things nodes, and the drug manufacturer information includes closed transportation information of raw materials, quality information of raw materials, status of production equipment, production environment and quality information of finished products;

[0033] The information transmission module asymmetrically encrypts the information of the drug manufacturer involved, and transmits the information of the drug manufacturer involved and the asymmetrical encryption key to the information verification module through the Internet of Things node;

[0034] The information verification module decrypts the asymmetric encryption key and verifies the drug manufacturer information based on the decryption result, wherein:

[0035] When the decryption result is consistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the verified drug manufacturer information;

[0036] When the decryption result is inconsistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the drug manufacturer information that has not passed the verification;

[0037] The monitoring and analysis module obtains the real-time network traffic H in Wi reshark, and compares the real-time network traffic H with the preset real-time network traffic H0, and performs abnormal monitoring and analysis on the verified drug manufacturer information according to the comparison result, and outputs the abnormal monitoring and analysis result, wherein:

[0038] When H≤H0, the monitoring and analysis module determines that the verified drug manufacturer information at the current time is in a normal state;

[0039] When H≥H0, the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, and outputs the drug manufacturer information that has passed verification at the current time as an abnormal monitoring and analysis result;

[0040] When the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, the monitoring and analysis module obtains the signal source of the real-time network traffic corresponding to the drug manufacturer information that has passed verification at the current time, judges the signal property based on it, and calibrates the abnormal monitoring and analysis result based on the judgment result, wherein:

[0041] When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an external networking signal, the monitoring and analysis module determines that the signal nature is an abnormal access signal;

[0042] When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an internal networking signal, the monitoring and analysis module determines that the signal nature is an active setting signal, calibrates the abnormal monitoring and analysis results, and corrects the abnormal state of the verified drug manufacturer information at the current time to a normal state;

[0043] The intelligent decision-making module inputs the abnormal monitoring and analysis results obtained by the monitoring and analysis module into the decision-making model, and the decision-making model outputs the intelligent decision.

[0044] On the other hand, the present invention also provides a method for monitoring pharmaceutical manufacturers based on the Internet of Things, comprising:

[0045] Step S1, classifying and marking the IoT nodes according to the node service information through the IoT node management module;

[0046] Step S2, adjusting the process of classifying and marking the IoT nodes according to the newly added IoT nodes;

[0047] Step S3, self-organize the IoT nodes through the node self-organization module,

[0048] Step S4, optimizing the self-organizing process according to the remaining power of each IoT node;

[0049] Step S5, adjusting the self-organizing optimization process according to the fluctuation properties of the IoT nodes;

[0050] Step S6, self-repairing the IoT node through the node self-repair module;

[0051] Step S7, collecting information about drug manufacturers through IoT nodes;

[0052] Step S8, transmitting the drug manufacturer information through the information transmission module;

[0053] Step S9, verifying the transmitted drug manufacturer information through an information verification module;

[0054] Step S10, performing abnormal monitoring and analysis on the verified drug manufacturer information through the monitoring and analysis module, and outputting the abnormal monitoring and analysis results;

[0055] Step S11, calibrating the abnormal monitoring analysis result according to the active setting signal;

[0056] Step S12, making a decision on the drug manufacturer according to the abnormal monitoring and analysis results through the intelligent decision-making module, and outputting the intelligent decision according to the decision-making result;

[0057] Step S13, executing the intelligent decision through the decision execution module.

[0058] Compared with the prior art, the beneficial effect of the present invention lies in that the system classifies and marks the IoT nodes according to the node service information through the IoT node management module, so as to manage the IoT nodes according to the information type collected by the IoT nodes. The system also adjusts the process of classifying and marking the IoT nodes according to the newly added IoT nodes through the IoT node management module, so as to classify and mark the newly added IoT nodes when the IoT nodes are added, and further adjust the classification marks of the original IoT nodes when the newly added IoT nodes have an impact on the original IoT nodes. The system self-organizes the IoT nodes through the node self-organizing module to ensure the stability of the IoT nodes. The system also optimizes the self-organizing process according to the remaining power of each IoT node through the node self-organizing module to avoid the remaining power of each IoT node affecting the self-organizing process. The system also adjusts the self-organizing optimization process according to the fluctuation properties of the IoT nodes through the node self-organizing module to avoid the fluctuation properties of the IoT nodes affecting the self-organizing process. The system self-repairs the IoT nodes through the node self-repairing module to avoid the IoT node failure being unable to Self-repair is performed to improve the monitoring efficiency of the Internet of Things nodes on drug manufacturers. The system collects information on drug manufacturers through the information collection module to facilitate subsequent analysis of the information on drug manufacturers. The system transmits the information on drug manufacturers through the information transmission module to facilitate monitoring of drug manufacturers. The system verifies the transmitted information on drug manufacturers through the information verification module to facilitate secure transmission of the information on drug manufacturers. The system performs abnormal monitoring and analysis on the verified information on drug manufacturers through the monitoring and analysis module, and outputs the abnormal monitoring and analysis results to eliminate the abnormal monitoring results, thereby improving the Internet of Things monitoring efficiency of drug manufacturers. The system also calibrates the abnormal monitoring and analysis results according to the active setting signal through the monitoring and analysis module to further determine the accuracy of the abnormal monitoring and analysis results, thereby further improving the monitoring efficiency of drug manufacturers. The system makes decisions and judgments on drug manufacturers according to the abnormal monitoring and analysis results through the intelligent decision-making module, and outputs the intelligent decision according to the decision judgment results, and executes the intelligent decision through the decision execution module to make autonomous decisions and repair abnormal situations, thereby further improving the monitoring efficiency of drug manufacturers. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a schematic diagram of the structure of the drug manufacturer monitoring system based on the Internet of Things in this embodiment;

[0060] Figure 2 Schematic diagram of the process of the pharmaceutical manufacturer monitoring method based on the Internet of Things in this embodiment. DETAILED DESCRIPTION

[0061] 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.

[0062] The preferred embodiments 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 embodiments 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.

[0063] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0064] See also Figure 1 As shown, it is a schematic diagram of the structure of the drug manufacturer monitoring system based on the Internet of Things in this embodiment, and the system includes:

[0065] An IoT node management module, used to classify and mark IoT nodes according to node service information, and also used to adjust the process of classifying and marking IoT nodes according to newly added IoT nodes;

[0066] A node self-organizing module is used to self-organize the IoT nodes, optimize the self-organizing process according to the remaining power of each IoT node, and adjust the self-organizing optimization process according to the fluctuation properties of the IoT nodes. The node self-organizing module is connected to the IoT node management module;

[0067] A node self-repair module is used to self-repair the IoT node, and the node self-repair module is connected to the node self-organizing module;

[0068] An information collection module is used to collect information about drug manufacturers through IoT nodes. The information collection module is connected to the node self-repair module.

[0069] An information transmission module is used to transmit information about drug manufacturers, and the information transmission module is connected to the information collection module;

[0070] An information verification module is used to verify the transmitted drug manufacturer information, and the information verification module is connected to the information transmission module;

[0071] The monitoring and analysis module is used to perform abnormal monitoring and analysis on the verified drug manufacturer information, and output the abnormal monitoring and analysis results, and is also used to calibrate the abnormal monitoring and analysis results according to the active setting signal. The monitoring and analysis module is connected to the information verification module;

[0072] The intelligent decision-making module is used to make decisions on drug manufacturers based on the abnormal monitoring and analysis results, and output intelligent decisions based on the decision-making results. The intelligent decision-making module is connected to the monitoring and analysis module;

[0073] The decision execution module is used to execute the intelligent decision, and the decision execution module is connected to the intelligent decision module.

[0074] Specifically, the system is set in the monitoring equipment of the pharmaceutical manufacturer, and collects the information of the pharmaceutical manufacturer through the Internet of Things nodes of the pharmaceutical manufacturer to realize the management of the transportation, production and storage processes of the pharmaceutical manufacturer, thereby realizing safe and effective monitoring of the pharmaceutical manufacturer. The system self-organizes and self-repairs the Internet of Things nodes of the pharmaceutical manufacturer, encrypts the transmission of the monitoring information of the pharmaceutical manufacturer, and makes independent decisions and controls on abnormal situations, thereby improving the monitoring efficiency and information security of the pharmaceutical manufacturer. Among them, the system classifies and marks the Internet of Things nodes according to the node service information through the Internet of Things node management module, so as to manage the Internet of Things nodes according to the type of information collected by the Internet of Things nodes. The system also manages the Internet of Things nodes through the Internet of Things node management module. The network node management module adjusts the process of classifying and marking the Internet of Things nodes according to the newly added Internet of Things nodes, so as to classify and mark the newly added Internet of Things nodes when the Internet of Things nodes are added, and when the newly added Internet of Things nodes have an impact on the original Internet of Things nodes, the classification marks of the original Internet of Things nodes are further adjusted. The system self-organizes the Internet of Things nodes through the node self-organizing module to ensure the stability of the Internet of Things nodes. The system also optimizes the self-organizing process according to the remaining power of each Internet of Things node through the node self-organizing module to avoid the remaining power of each Internet of Things node affecting the self-organizing process. The system also optimizes the self-organizing process according to the fluctuation properties of the Internet of Things nodes through the node self-organizing module. The organization optimization process is adjusted to avoid the impact of the fluctuating properties of the IoT nodes on the self-organizing process. The system self-repairs the IoT nodes through the node self-repair module to avoid the inability of IoT nodes to self-repair due to failures, thereby improving the monitoring efficiency of IoT nodes on drug manufacturers. The system collects drug manufacturer information through the information collection module to facilitate subsequent analysis of the drug manufacturer information. The system transmits the drug manufacturer information through the information transmission module to facilitate monitoring of the drug manufacturers. The system verifies the transmitted drug manufacturer information through the information verification module to facilitate secure transmission of the drug manufacturer information. The system verifies the verified drug manufacturer information through the monitoring and analysis module. The information of drug manufacturers is monitored and analyzed for abnormalities, and the results of the abnormal monitoring and analysis are output to facilitate the elimination of abnormal monitoring results, thereby improving the Internet of Things monitoring efficiency of drug manufacturers. The system also calibrates the abnormal monitoring and analysis results according to the active setting signal through the monitoring and analysis module to further determine the accuracy of the abnormal monitoring and analysis results, thereby further improving the monitoring efficiency of drug manufacturers. The system makes decisions and judgments on drug manufacturers based on the abnormal monitoring and analysis results through the intelligent decision-making module, and outputs the intelligent decisions based on the decision-making judgment results, and executes the intelligent decisions through the decision execution module to make autonomous decisions and repair abnormal situations, thereby further improving the monitoring efficiency of drug manufacturers.

[0075] Specifically, the IoT node management module inputs the node service information into the TF-IDF vectorizer, and the TF-IDF vectorizer performs fitting conversion on the node service information, and converts the node service information into a TF-IDF feature matrix X, and calculates the Euclidean distance d between each IoT node and each cluster according to the TF-IDF feature matrix X of each IoT node. K， Setting K = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the example, the cluster corresponding to the minimum Euclidean distance is selected as the preset cluster of the IoT node, each IoT node is classified into its corresponding preset cluster, and the preset clustering objective function J is calculated. k is the total number of clusters, Ci is the set of the i-th cluster, μ i is the centroid of the ith cluster, the IoT node management module compares the preset clustering objective function J with the preset marking objective function J0, and classifies and marks each IoT node according to the comparison result, where:

[0076] When J≤J0, the preset clustering of each IoT node is used as the classification of each IoT node and marked;

[0077] When J>J0, the centroid of each cluster is adjusted, and the Euclidean distance d between each IoT node and each cluster is recalculated based on the TF-IDF feature matrix X of each IoT node. K， Setting K = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the process, the cluster corresponding to the minimum Euclidean distance is reselected as the preset cluster of the IoT node until the preset cluster objective function J is less than or equal to the preset marking objective function J0. The preset cluster of each IoT node is used as the classification of each IoT node and marked. If the preset cluster objective function J is still greater than the preset marking objective function J0 when the number of iterations n reaches the preset number of iterations n0, the preset cluster of each IoT node in the nth iteration is used as the classification of each IoT node and marked.

[0078] Specifically, the node service information refers to the information collected by the IoT node during the transportation, production and storage of drugs by the drug manufacturer, including the IoT location coordinates, the information content collected by the IoT and the type of production line collected by the IoT. The TF-IDF vectorizer refers to a tool for converting text data into numerical features. This embodiment does not specifically limit the process of the TF-IDF vectorizer converting the node service information into the TF-IDF feature matrix X. Those skilled in the art can freely set it according to the actual situation, as long as the conversion requirements of the TF-IDF feature matrix X are met. For example, the fi in the TF-IDF vectorizer can be set. The t_transform method simultaneously performs fitting and transformation operations on the node service information, and converts the node service information as text data into a TF-IDF feature matrix X, wherein the TF-IDF feature matrix X refers to a feature sparse matrix representing each document, and the preset marked objective function J0 refers to a preset value indicating that the objective function has reached convergence, and the preset marked objective function J0 can be set to 0.5, and the preset number of iterations n0 refers to a preset value indicating the number of iterations that reach the clustering standard, such as the preset number of iterations n0 can be set to 10, and the Euclidean distance between each IoT node and each cluster refers to the straight-line distance between two points. In the clustering algorithm, the Euclidean distance from each IoT node to the centroid of each cluster is calculated, and the centroid of each cluster refers to the average position of all nodes in the cluster, and the number of iterations refers to the number of iterations of the clustering algorithm from initialization to convergence.

[0079] Specifically, the IoT node management module calculates the IoT node access time difference △T, sets △T=TA-T0, TA is the IoT access time, T0 is the current time, and compares the IoT node access time difference △T with the preset IoT node access time difference △T0, and judges the newly added IoT node according to the comparison result, wherein:

[0080] When △T>△T0, the IoT node management module determines that the IoT node is not a newly added IoT node;

[0081] When △T≤△T0, the Internet of Things node management module determines that the Internet of Things node is a newly added Internet of Things node and adjusts the node service information. The adjusted node service information is an information set after adding the newly added Internet of Things node service information. The Internet of Things node management module classifies and marks the newly added Internet of Things nodes according to the adjusted node service information.

[0082] Specifically, the IoT node access time difference refers to the time interval from the IoT access moment to the current moment, the IoT access moment refers to the moment when the IoT node accesses the IoT node network of the pharmaceutical manufacturer to collect information about the pharmaceutical manufacturer, and the preset IoT node access time difference refers to the time interval indicating that the IoT node has just been connected to the IoT node network of the pharmaceutical manufacturer as a newly added IoT node, such as setting the preset IoT node access time difference to 10 minutes.

[0083] Specifically, the node self-organizing module sets the protocol of each IoT node to a distributed network protocol, compares the polling time interval TP with the preset polling time interval TP0, and judges the cluster head node of each cluster according to the comparison result, wherein:

[0084] When TP≤TP0, the node self-organizing module determines the cluster head node of each cluster as the current cluster head node;

[0085] When TP>TP0, the node self-organizing module determines the cluster head node of each cluster as the next cluster head node.

[0086] Specifically, the distributed network protocol refers to the transmission protocol of each IoT node. This embodiment does not limit the specific type of the distributed network protocol. Technical personnel in this field can freely set it according to actual conditions, as long as the self-organization requirements of each IoT node are met. For example, the distributed network protocol can be set to Zigbee. The polling time interval refers to the time interval between the current moment and the last polling moment. The preset polling time interval refers to the time interval for polling the cluster head nodes of each cluster, such as 1 hour. The cluster head node refers to an IoT node that plays a leading role and has responsibilities in each cluster of the IoT network. The node self-organizing module randomly sorts the cluster head nodes of each cluster to obtain the cluster head node order of each cluster.

[0087] Specifically, the node self-organizing module sets the protocol of each IoT node to a distributed network protocol. The distributed network protocol has a built-in self-organizing function so that it can automatically adjust when each IoT node joins or leaves the network, thereby giving the IoT node self-organizing and self-repairing capabilities, enabling it to automatically reorganize the network and restore normal functions when a failure occurs, thereby significantly improving the stability and reliability of the system. The node self-organizing module polls the cluster head nodes of each cluster so as to elect the cluster head in turn, thereby balancing the load of each node.

[0088] Specifically, when the drug manufacturer is powered off, the node self-organizing module compares the remaining power A of each IoT node with the preset power A0, and judges the energy mode of each IoT node according to the comparison result, and optimizes the order of cluster head nodes of each cluster according to the judgment result, wherein:

[0089] When A≥A0, the node self-organizing module determines that the IoT node is in a high energy mode;

[0090] When A<A0, the node self-organizing module determines that the Internet of Things node is in low energy mode, optimizes the cluster head node sequence of each cluster, removes the Internet of Things node from the cluster head node sequence of its corresponding cluster, and uses the cluster head node sequence of the corresponding cluster after removal as the cluster head node sequence of each cluster after optimization.

[0091] Specifically, the remaining power of each IoT node refers to the IoT node that can continue to collect information when the drug manufacturer is powered off, and the preset power refers to the preset value that distinguishes the power of the IoT node in high energy mode and low energy mode, such as setting the preset power to 30% of the total power.

[0092] Specifically, the node self-organizing module compares the offline times P of each IoT node with the preset offline times P0, and judges the fluctuation attribute of the IoT node according to the comparison result, wherein:

[0093] When P≤P0, the node self-organizing module determines that the IoT node is a non-fluctuating node;

[0094] When P>P0, the node self-organizing module determines that the Internet of Things node is a fluctuating node, and adjusts the order of cluster head nodes of each cluster after optimization, and removes the Internet of Things node from the order of cluster head nodes of each cluster after optimization.

[0095] Specifically, the number of offline times of each IoT node refers to the total number of times that each IoT node is separated from the IoT node network of the pharmaceutical manufacturer due to insufficient power, communication failure, equipment failure, etc., and the preset number of offline times refers to the preset value of the number of offline times for determining whether the IoT node is a fluctuating node.

[0096] Specifically, the node self-repair module obtains the heartbeat message of each IoT node in each heartbeat cycle, obtains a heartbeat message collection, polls and compares each node ID in the heartbeat message collection with the preset node ID, obtains the missing node ID after the polling and comparison, takes the missing node ID as the faulty IoT node, removes the faulty IoT node from the cluster head node sequence of each cluster after optimization, performs self-repair according to the distributed network protocol, and sends the location information of the faulty IoT node to the user terminal.

[0097] Specifically, the heartbeat cycle refers to the preset time interval for the IoT node to periodically send heartbeat messages, such as the heartbeat cycle can be set to 24 hours. The heartbeat message of each IoT node refers to the information periodically sent by each IoT node to indicate its online status. Each node ID refers to a combination of numbers representing the identity characteristics of each node. The preset node ID refers to the collection of all IoT node IDs that collect information on drug-related manufacturers. The location information of the faulty IoT node refers to the location information of the faulty IoT node in the drug-related manufacturer. The user terminal refers to a device terminal of a user who has monitoring responsibilities for the IoT node, such as a mobile terminal.

[0098] Specifically, the information collection module collects information about drug manufacturers through IoT nodes, and the drug manufacturer information includes closed transportation information of manufacturer's raw materials, raw material quality information, manufacturer's production equipment status, manufacturer's production environment information and manufacturer's finished product quality information.

[0099] Specifically, the information transmission module asymmetrically encrypts the information of the drug manufacturers, and transmits the drug manufacturers' information and the asymmetric encryption key to the information verification module through the Internet of Things node.

[0100] It can be understood that this embodiment does not limit the asymmetric encryption method. Those skilled in the art can freely set it according to the actual situation as long as the encryption requirements are met. For example, the drug manufacturer information can be asymmetrically encrypted by means of an elliptic encryption function. The asymmetric encryption key refers to the verification key for each IoT node to transmit the drug manufacturer information.

[0101] Specifically, the information verification module decrypts the asymmetric encryption key and verifies the drug manufacturer information based on the decryption result, wherein:

[0102] When the decryption result is consistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the verified drug manufacturer information;

[0103] When the decryption result is inconsistent with the preset asymmetric verification key, the information verification module determines that the drug-related manufacturer information corresponding to the asymmetric encryption key is drug-related manufacturer information that has not passed verification.

[0104] Specifically, the preset asymmetric verification key refers to the verification key corresponding to the asymmetric encryption key of each IoT node.

[0105] Specifically, the monitoring and analysis module obtains the real-time network traffic H in Wi reshark, and compares the real-time network traffic H with the preset real-time network traffic H0, and performs abnormal monitoring and analysis on the verified drug manufacturer information according to the comparison result, and outputs the abnormal monitoring and analysis result, wherein:

[0106] When H≤H0, the monitoring and analysis module determines that the verified drug manufacturer information at the current time is in a normal state;

[0107] When H≥H0, the monitoring and analysis module determines that the drug-related manufacturer information that has passed verification at the current time is in an abnormal state, and outputs the drug-related manufacturer information that has passed verification at the current time as an abnormal monitoring and analysis result.

[0108] Specifically, the Wi reshark refers to a tool for intelligently monitoring the traffic of IoT nodes, the real-time network traffic refers to the real-time traffic accessing each IoT node, and the preset real-time network traffic refers to the actual traffic with abnormal access behavior, such as exceeding 30% of the average network traffic.

[0109] Specifically, when the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, the monitoring and analysis module obtains the signal source of the real-time network traffic corresponding to the drug manufacturer information that has passed verification at the current time, judges the signal property based on it, and calibrates the abnormal monitoring and analysis result based on the judgment result, wherein:

[0110] When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an external networking signal, the monitoring and analysis module determines that the signal nature is an abnormal access signal;

[0111] When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an internal networking signal, the monitoring and analysis module determines that the signal nature is an active setting signal, calibrates the abnormal monitoring and analysis results, and corrects the abnormal state of the verified drug manufacturer information at the current time to a normal state.

[0112] Specifically, the signal source of the real-time network traffic refers to the source of the real-time traffic for accessing each IoT node, including external networking signals and internal networking signals. The external networking signal refers to the access signal outside the IoT node network of the pharmaceutical manufacturer, and the internal networking signal refers to the access signal within the IoT node network of the pharmaceutical manufacturer. This embodiment does not limit the identification method of the signal source of the real-time network traffic. Technical personnel in this field can freely set it according to actual conditions, such as identifying the signal source of the real-time network traffic by network segmentation.

[0113] Specifically, the intelligent decision-making module inputs the abnormal monitoring and analysis results obtained by the monitoring and analysis module into the decision-making model, and the decision-making model outputs the intelligent decision.

[0114] Specifically, the decision model refers to a decision tree model that takes the abnormal monitoring and analysis results as input and intelligent decision as output. The intelligent decision refers to a decision to adjust the abnormal monitoring and analysis results. This embodiment does not limit the construction process of the decision model. Technical personnel in this field can freely set it according to actual conditions, and only need to meet the output requirements for intelligent decision-making. For example, historical expert decision data can be set as training data to train the decision tree model, and the trained decision tree model that meets the verification conditions can be output as a decision model.

[0115] See also Figure 2 As shown, it is a flow chart of the method for monitoring drug manufacturers based on the Internet of Things in this embodiment, and the method includes:

[0116] Step S1, classifying and marking the IoT nodes according to the node service information through the IoT node management module;

[0117] Step S2, adjusting the process of classifying and marking the IoT nodes according to the newly added IoT nodes;

[0118] Step S3, self-organizing the IoT nodes through a node self-organizing module;

[0119] Step S4, optimizing the self-organizing process according to the remaining power of each IoT node;

[0120] Step S5, adjusting the self-organizing optimization process according to the fluctuation properties of the IoT nodes;

[0121] Step S6, self-repairing the IoT node through the node self-repair module;

[0122] Step S7, collecting information about drug manufacturers through IoT nodes;

[0123] Step S8, transmitting the drug manufacturer information through the information transmission module;

[0124] Step S9, verifying the transmitted drug manufacturer information through an information verification module;

[0125] Step S10, performing abnormal monitoring and analysis on the verified drug manufacturer information through the monitoring and analysis module, and outputting the abnormal monitoring and analysis results;

[0126] Step S11, calibrating the abnormal monitoring analysis result according to the active setting signal;

[0127] Step S12, making a decision on the drug manufacturer according to the abnormal monitoring and analysis results through the intelligent decision-making module, and outputting the intelligent decision according to the decision-making result;

[0128] Step S13, executing the intelligent decision through the decision execution module.

[0129] 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.

Claims

1. A drug manufacturer monitoring system based on the Internet of Things, characterized in that: include: An IoT node management module, used to classify and mark IoT nodes according to node service information, and also used to adjust the process of classifying and marking IoT nodes according to newly added IoT nodes; A node self-organizing module is used to self-organize IoT nodes, optimize the self-organizing process according to the remaining power of each IoT node, and adjust the self-organizing optimization process according to the fluctuation properties of IoT nodes; Node self-repair module, used to self-repair IoT nodes; Information collection module, used to collect information about drug manufacturers through IoT nodes; Information transmission module, used to transmit information about drug manufacturers; Information verification module, used to verify the transmitted drug manufacturer information; The monitoring and analysis module is used to perform abnormal monitoring and analysis on the verified drug manufacturer information, output the abnormal monitoring and analysis results, and calibrate the abnormal monitoring and analysis results according to the active setting signal; Intelligent decision-making module, used to make decisions on drug manufacturers based on abnormal monitoring and analysis results, and output intelligent decisions based on the decision-making results; Decision execution module, used to execute intelligent decisions; The information collection module collects information about drug manufacturers through IoT nodes, and the drug manufacturer information includes closed transportation information of raw materials, quality information of raw materials, status of production equipment, production environment and quality information of finished products; The information transmission module asymmetrically encrypts the information of the drug manufacturer involved, and transmits the information of the drug manufacturer involved and the asymmetrical encryption key to the information verification module through the Internet of Things node; The information verification module decrypts the asymmetric encryption key and verifies the drug manufacturer information based on the decryption result, wherein: When the decryption result is consistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the verified drug manufacturer information; When the decryption result is inconsistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the drug manufacturer information that has not passed the verification; The monitoring and analysis module obtains the real-time network traffic H in Wireshark, and compares the real-time network traffic H with the preset real-time network traffic H0, and performs abnormal monitoring and analysis on the verified drug manufacturer information according to the comparison result, and outputs the abnormal monitoring and analysis result, wherein: When H≤H0, the monitoring and analysis module determines that the verified drug manufacturer information at the current time is in a normal state; When H≥H0, the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, and outputs the drug manufacturer information that has passed verification at the current time as an abnormal monitoring and analysis result; When the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, the monitoring and analysis module obtains the signal source of the real-time network traffic corresponding to the drug manufacturer information that has passed verification at the current time, judges the signal property based on it, and calibrates the abnormal monitoring and analysis result based on the judgment result, wherein: When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an external networking signal, the monitoring and analysis module determines that the signal nature is an abnormal access signal; When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an internal networking signal, the monitoring and analysis module determines that the signal nature is an active setting signal, calibrates the abnormal monitoring and analysis results, and corrects the abnormal state of the verified drug manufacturer information at the current time to a normal state; The intelligent decision-making module inputs the abnormal monitoring and analysis results obtained by the monitoring and analysis module into the decision-making model, and the decision-making model outputs the intelligent decision.

2. The Internet of Things-based drug manufacturer monitoring system according to claim 1 is characterized in that: The IoT node management module inputs the node service information into the TF-IDF vectorizer, and the TF-IDF vectorizer performs fitting conversion on the node service information, and converts the node service information into a TF-IDF feature matrix X, and calculates the Euclidean distance d between each IoT node and each cluster according to the TF-IDF feature matrix X of each IoT node. K， Setting K = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the example, the cluster corresponding to the minimum Euclidean distance is selected as the preset cluster of the IoT node, each IoT node is classified into its corresponding preset cluster, and the preset clustering objective function J is calculated, and J= , k is the total number of clusters, Ci is the set of the i-th cluster, μ i is the centroid of the ith cluster.

3. The Internet of Things-based drug manufacturer monitoring system according to claim 2 is characterized in that: The IoT node management module compares the preset clustering objective function J with the preset marking objective function J0, and classifies and marks each IoT node according to the comparison result, where: When J≤J0, the preset clustering of each IoT node is used as the classification of each IoT node and marked; When J>J0, the centroid of each cluster is adjusted, and the Euclidean distance d between each IoT node and each cluster is recalculated based on the TF-IDF feature matrix X of each IoT node. K， Setting K = / / X-μ K / / , μ K is the centroid of each cluster, and the Euclidean distance d between each IoT node and each cluster K In the process, the cluster corresponding to the minimum Euclidean distance is reselected as the preset cluster of the IoT node until the preset cluster objective function J is less than or equal to the preset marking objective function J0. The preset cluster of each IoT node is used as the classification of each IoT node and marked. If the preset cluster objective function J is still greater than the preset marking objective function J0 when the number of iterations n reaches the preset number of iterations n0, the preset cluster of each IoT node in the nth iteration is used as the classification of each IoT node and marked.

4. The Internet of Things-based drug manufacturer monitoring system according to claim 3 is characterized in that: The IoT node management module calculates the IoT node access time difference △T, sets △T=TA-T0, TA is the IoT access time, T0 is the current time, and compares the IoT node access time difference △T with the preset IoT node access time difference △T0, and judges the newly added IoT node according to the comparison result, wherein: When △T>△T0, the IoT node management module determines that the IoT node is not a newly added IoT node; When △T≤△T0, the Internet of Things node management module determines that the Internet of Things node is a newly added Internet of Things node and adjusts the node service information. The adjusted node service information is an information set after adding the newly added Internet of Things node service information. The Internet of Things node management module classifies and marks the newly added Internet of Things nodes according to the adjusted node service information.

5. The Internet of Things-based drug manufacturer monitoring system according to claim 4 is characterized in that: The node self-organizing module sets the protocol of each IoT node as a distributed network protocol, compares the polling time interval TP with the preset polling time interval TP0, and judges the cluster head node of each cluster according to the comparison result, wherein: When TP≤TP0, the node self-organizing module determines the cluster head node of each cluster as the current cluster head node; When TP>TP0, the node self-organizing module determines the cluster head node of each cluster as the next cluster head node.

6. The Internet of Things-based drug manufacturer monitoring system according to claim 5 is characterized in that: When the drug manufacturer is powered off, the node self-organizing module compares the remaining power A of each IoT node with the preset power A0, and judges the energy mode of each IoT node according to the comparison result, and optimizes the order of cluster head nodes of each cluster according to the judgment result, wherein: When A≥A0, the node self-organizing module determines that the IoT node is in a high energy mode; When A<A0, the node self-organizing module determines that the Internet of Things node is in low energy mode, optimizes the order of cluster head nodes of each cluster, removes the Internet of Things node from the order of cluster head nodes of the corresponding cluster, and uses the order of cluster head nodes of the corresponding cluster after removal as the order of cluster head nodes of each cluster after optimization.

7. The Internet of Things-based drug manufacturer monitoring system according to claim 6 is characterized in that: The node self-organizing module compares the offline times P of each IoT node with the preset offline times P0, and judges the fluctuation attribute of the IoT node according to the comparison result, wherein: When P≤P0, the node self-organizing module determines that the IoT node is a non-fluctuating node; When P>P0, the node self-organizing module determines that the Internet of Things node is a fluctuating node, and adjusts the order of cluster head nodes of each cluster after optimization, and removes the Internet of Things node from the order of cluster head nodes of each cluster after optimization.

8. The Internet of Things-based pharmaceutical manufacturer monitoring system according to claim 1 is characterized in that: The node self-repair module obtains the heartbeat message of each IoT node in each heartbeat cycle to obtain a heartbeat message collection, performs polling comparison on each node ID in the heartbeat message collection and a preset node ID, obtains the missing node ID after the polling comparison, takes the missing node ID as a faulty IoT node, removes the faulty IoT node from the order of the cluster head nodes of each cluster after optimization, performs self-repair according to the distributed network protocol, and sends the location information of the faulty IoT node to the user terminal.

9. A method for applying to the Internet of Things-based drug manufacturer monitoring system as described in any one of claims 1 to 8, characterized in that: The method comprises: Step S1, classifying and marking the IoT nodes according to the node service information through the IoT node management module; Step S2, adjusting the process of classifying and marking the IoT nodes according to the newly added IoT nodes; Step S3, self-organize the IoT nodes through the node self-organization module, Step S4, optimizing the self-organizing process according to the remaining power of each IoT node; Step S5, adjusting the self-organizing optimization process according to the fluctuation properties of the IoT nodes; Step S6, self-repairing the IoT node through the node self-repair module; Step S7, collecting information about drug manufacturers through the IoT node; the drug manufacturer information includes closed transportation information of raw materials, raw material quality information, manufacturer production equipment status, manufacturer production environment information and manufacturer finished product quality information; Step S8, transmitting the drug manufacturer information through the information transmission module; asymmetrically encrypting the drug manufacturer information, and transmitting the drug manufacturer information and the asymmetrical encryption key to the information verification module through the Internet of Things node; Step S9, verify the transmitted drug manufacturer information through the information verification module; decrypt the asymmetric encryption key, and verify the drug manufacturer information according to the decryption result, wherein: When the decryption result is consistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the verified drug manufacturer information; When the decryption result is inconsistent with the preset asymmetric verification key, the information verification module determines that the drug manufacturer information corresponding to the asymmetric encryption key is the drug manufacturer information that has not passed the verification; Step S10, performing abnormal monitoring and analysis on the verified drug-related manufacturer information through the monitoring and analysis module, and outputting the abnormal monitoring and analysis results; obtaining the real-time network traffic H in Wireshark through the monitoring and analysis module, and comparing the real-time network traffic H with the preset real-time network traffic H0, and performing abnormal monitoring and analysis on the verified drug-related manufacturer information according to the comparison results, and outputting the abnormal monitoring and analysis results, wherein: When H≤H0, the monitoring and analysis module determines that the verified drug manufacturer information at the current time is in a normal state; When H≥H0, the monitoring and analysis module determines that the drug manufacturer information that has passed verification at the current time is in an abnormal state, and outputs the drug manufacturer information that has passed verification at the current time as an abnormal monitoring and analysis result; Step S11, calibrating the abnormal monitoring and analysis results according to the active setting signal; when the monitoring and analysis module determines that the drug-related manufacturer information that has passed verification at the current time is in an abnormal state, the signal source of the real-time network traffic corresponding to the drug-related manufacturer information that has passed verification at the current time is obtained, and the signal property is judged according to the signal source, and the abnormal monitoring and analysis results are calibrated according to the judgment result, wherein: When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an external networking signal, the monitoring and analysis module determines that the signal nature is an abnormal access signal; When the signal source of the real-time network traffic corresponding to the verified drug manufacturer information at the current time is an internal networking signal, the monitoring and analysis module determines that the signal nature is an active setting signal, calibrates the abnormal monitoring and analysis results, and corrects the abnormal state of the verified drug manufacturer information at the current time to a normal state; Step S12, making a decision on the drug manufacturer according to the abnormal monitoring and analysis results through the intelligent decision-making module, and outputting the intelligent decision according to the decision-making judgment results; inputting the abnormal monitoring and analysis results obtained by the monitoring and analysis module into the decision model through the intelligent decision-making module, and the decision model outputs the intelligent decision; Step S13, executing the intelligent decision through the decision execution module.

Images