Network big data secure transmission management system based on artificial intelligence

Abstract

The invention discloses a network big data secure transmission management system based on artificial intelligence, and relates to the technical field of network big data secure transmission management, a regional specification platform is set to be applied to enterprise internal networks, and for data acquisition and transmission between the internal networks, authority is acquired after verification of a login node unit; relatively convenient enterprise internal multi-area network data transmission is realized, searching and locking are performed through the login node unit when data leakage, damage and other problems occur, an external login environment is strictly examined when data access is performed outside an enterprise, and human face and contour acquisition is performed after multiple conditions are met. And for a malicious external login node data blasting means, the safety management unit adopts artificial intelligence to generate simulation data for external playing so as to complete protection of the external login node.

Description

Technical Field

[0001] The present invention relates to the technical field of network big data security transmission management, and specifically to a network big data security transmission management system based on artificial intelligence. Background Art

[0002] With the development of network technology, more and more enterprises divide their internal networks into multiple network areas for business needs and network security considerations. The multiple network areas are usually isolated by physical isolation, logical isolation, network gates, firewalls, etc., so as to ensure network independence and data security.

[0003] However, network isolation cannot objectively prevent the emergence of cross-network data transmission needs. In actual daily operations, enterprises will still have more cross-network transmission needs, such as data file transfer between different departments within the enterprise, and data file transfer between the enterprise and external partners. Therefore, a practical big data transmission management solution is needed to ensure data security.

[0004] To this end, we propose an artificial intelligence-based network big data security transmission management system. Summary of the invention

[0005] The purpose of the present invention is to provide a network big data security transmission management system based on artificial intelligence to solve the problem of data leakage in data file transmission between internal and external partners of an enterprise proposed in the above background technology;

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A network big data security transmission management system based on artificial intelligence includes a data security supervision platform and a data terminal, wherein the data terminal is used to encrypt the big data to be stored to obtain encrypted data and an encryption chain, and send the encrypted data and the encryption chain to the data security supervision platform in a unified manner;

[0008] The data security supervision platform is used to store and process encrypted data and encryption chains, and includes a data distribution unit, a database, a security management unit and a node management unit;

[0009] The data distribution unit distributes the encrypted data to the storage node for storage, and at the same time merges the encryption chain corresponding to the encrypted data with the node chain in the storage node storing the encrypted data to obtain a combined chain and sends it to the database for storage;

[0010] The node management unit is used to collect node information and send the node information to the security management unit;

[0011] The security management unit is used for security identification and allocation management of node information and generates a scheduling signal. At the same time, the security management unit generates accurate stack data or simulated stack data after preprocessing the scheduling signal and sends it to the database;

[0012] The database is used to store encrypted data, combination chains, smart terminal parameters and external personnel information.

[0013] Furthermore, the data end performs encryption processing on the big data to be stored, specifically including:

[0014] Step 1: Perform character and image recognition on the stored big data, set unique character values ​​corresponding to all characters, match the obtained character strings with their corresponding unique character values ​​in order to obtain the character string values; set unique color values ​​corresponding to all colors, magnify the obtained image along the center position to obtain a pixel grid, and match the obtained pixel grid with its corresponding unique color values ​​in order to obtain the color pixel map values;

[0015] Step 2: According to the formula Y=Ex+326, E is the corresponding string value or color pixel value in the stored big data, x is a manually set parameter, and Y is the combined value;

[0016] Step 3: Convert the combined value into a graphic value, obtain the combined value digit value, and construct an equipotential polygon according to the combined value digit value, fill the equipotential polygon with a sequential graphic according to the intersection points formed by two line segments of the equipotential polygon, set a sequential graphic corresponding to each of the digits from zero to nine, print the sequential graphics corresponding to the combined value at the intersection points of the equipotential polygons, obtain an encrypted graphic string, and obtain the encrypted data according to the encrypted graphic string;

[0017] Step 4: Separate and save the encrypted graphic strings obtained by the string value and the color pixel image value according to the storage time.

[0018] Furthermore, the data security supervision platform also includes a login node unit, which is used for the owners of smart terminals within the control range of the data security supervision platform to submit smart terminal parameters for registration through the smart terminal, and send the successfully registered smart terminal parameters to the database for storage. The database marks the successfully registered smart terminal as an internal test node, and generates a unique verification number for the internal test node, and marks the smart terminal parameters as internal test parameters, wherein the internal test parameters include registration time, location and registration number.

[0019] Furthermore, the node information includes the identity information of the login user, the initial time and the end time of the connection, the number of failed connections and the remaining memory of the connection node.

[0020] Furthermore, the node information security identification and allocation management operations of the security management unit specifically include:

[0021] Step 1: Calculate the time difference between the connection start time and the connection success time to obtain the single connection duration, compare the single connection duration with the preset duration threshold, and when the single connection duration is greater than the preset duration threshold, calculate the difference between the two to obtain the single overtime duration, and mark the single overtime duration as Dc, the number of failed connections as Wt, and the remaining memory of the connection node as Sy;

[0022] Step 2: According to the formula: Z = a1*DC+(a2Wt / a3Sy), where the risk scheduling coefficient Z, a1, a2 and a3 are correction factors, and a2>a3>a1>0; here, when the external user logs in again, when logging in without obtaining the combined chain, multiple attempts to fill in the wrong data will cause the connection node to record the wrong data and feed it back to the security management unit, and at the same time, it will also cause the single remaining memory at the connection node to decrease. The more times it fails, the less the remaining memory of the connection node, and the greater the risk scheduling coefficient Z;

[0023] Compare the risk scheduling coefficient Z with the risk scheduling coefficient threshold z:

[0024] When Z ≥ z, a first-level scheduling signal is generated;

[0025] When Z<z, a secondary scheduling signal is generated;

[0026] Step 3: When a first-level dispatch signal is generated in the security management unit, a "pay attention to login time" mark is actively generated and fed back to the smart terminal that is not supervised by the data security supervision platform;

[0027] When the number of first-level dispatch signal generation in the security management unit is ≥5, the simulation stack data is actively generated and sent to the database, where the simulation stack data is the simulation combination chain parsing data automatically generated by artificial intelligence according to the data filled in by the previous registrant, and the parsed simulation combination chain data is sent to the external registration node, where the external registration node is a smart terminal not monitored by the data security supervision platform;

[0028] Step 4: When a secondary scheduling signal is generated in the security management unit, a "multi-angle face extraction and recognition request" is actively generated on the external register node. After obtaining permission and collecting the face image, the precise stack data is automatically generated. The precise stack data is the combined chain parsed data input by the security management unit based on the external register node, and is fed back to the external register node.

[0029] Furthermore, the encryption chain is composed of the level, search directory and identification box of the encrypted data; the node chain is composed of the authentication number of the storage node and the corresponding characters.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] In the present invention, by setting up a regional specification platform for application in the internal network of the enterprise, when the enterprise obtains external big data, the data end first encrypts the data, and cooperates with the database to complete the storage of encrypted data and data chain. For data acquisition and transmission between internal networks, the authority is obtained after verification by the login node unit, so as to realize more convenient multi-regional network data transmission within the enterprise, and when problems such as data leakage and damage occur, the login node unit is used to search and lock. When data is accessed from outside the enterprise, the external login environment is strictly examined, and the face and contour are collected after multiple conditions are met. For malicious external login node data blasting methods, the security management unit will use artificial intelligence to generate simulated data for external playback to complete the protection of external login nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a block diagram of the network big data security transmission management system based on artificial intelligence of the present invention;

[0033] Figure 2 The schematic diagram of the arrangement of encrypted graphic strings formed by data encryption of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0035] See also Figure 1 , the present invention provides a technical solution:

[0036] Embodiment 1:

[0037] Based on the internal big data management of the enterprise, a data security supervision platform is established to encrypt the big data entering the enterprise. The encryption processing of the big data to be stored on the data end includes:

[0038] Perform character and image recognition on the stored big data, set all characters to correspond to unique character values, match the obtained string with its corresponding unique character values ​​in order to obtain the string value; set all colors to correspond to unique color values, zoom in on the obtained image along the center position to obtain a pixel grid, match the obtained pixel grid with its corresponding unique color value in order to obtain the color pixel map value, similar to the data end receiving the "purchase fee" character value, after the data end matching and conversion, each character corresponds to a unique character value and is converted to generate "52, 69, 441, 9632" where "," is the identification box in the encryption chain, which is used to isolate the data for easy identification;

[0039] According to the formula Y=Ex+326, E is the corresponding string value or color pixel value in the stored big data, x is a manually set parameter, and Y is a combined value. When the example character value "52694419632" is obtained, it is entered into the formula Y, where the x value is selected and input by the enterprise. When x=1, Y=52694419632*1+326, that is, 52694419958;

[0040] Convert the combined value into a graphic value, obtain the combined value digit value, and construct an equilateral polygon based on the combined value digit value. Fill the equilateral polygon with sequential graphics based on the intersection of two segments of the equilateral polygon. Set a sequential graphic corresponding to each of the digits from zero to nine. Print the sequential graphics corresponding to the combined value at the intersection of the equilateral polygon to obtain an encrypted graphic string. Obtain the encrypted data based on the encrypted graphic string. 52694419958 is eleven digits, which means that eleven polygons are generated. Then, convert the combined value from left to right to obtain eight different graphics and fill them in the corners of the eleven deformations in sequence, similar to Figure 2 As shown, the encrypted graphic string is obtained;

[0041] The encryption chain also includes the encryption data level and search directory, which are the generation time level and storage location of the encrypted graphic string. The data is ranked according to the time when the enterprise obtains the big data. The earliest encrypted graphic string is the first level, and so on.

[0042] The encrypted graphic string data obtained by the string value and the color pixel map value are stored separately according to the storage time. Since the string value may be the same as the encrypted graphic string value, they are placed in different areas in the database.

[0043] In order to ensure better data search in the later stage, the data distribution unit distributes the encrypted data to the storage node for storage, and at the same time merges the encryption chain corresponding to the encrypted data with the node chain in the storage node storing the encrypted data to obtain a combined chain and sends it to the database for storage. The combined chain here combines the encryption chain obtained after the input data is processed and the node chain stored in the entire database. The node chain is composed of the authentication number of the storage node and the corresponding character, similar to the database needing to store the string value and the encrypted graphic string value in different areas. The two areas will have corresponding authentication numbers. The combined chain produced by combining the encryption chain and the node chain is located in the database. When the external node needs to search for data in the later stage, the input combined chain can retrieve the corresponding encrypted graphic string. After being parsed by the security management unit, the complete data can be obtained, similar to the "1.2.31.26.34" combined chain, 1.2 means that the graphics and characters need to be obtained at the same time, 31 means the 31st warehouse of the graphics and characters database, 26 means the 26th layer, and 34 means the 34th level encrypted string and encrypted graphic string;

[0044] As for data access and transmission within the data security supervision platform, the required access verification is relatively simple. The login node unit set for this purpose is used for the owners of smart terminals within the control range of the data security supervision platform to submit smart terminal parameters for registration through the smart terminal, and the parameters of the successfully registered smart terminal are sent to the database for storage. The database marks the successfully registered smart terminal as an internal test node, and generates a unique verification number for the internal test node, and marks the smart terminal parameters as internal test parameters. The internal test parameters include registration time, location and registration number, that is, identity authentication and login within the scope specified by the enterprise, so as to obtain database data access rights and realize rapid data transmission between enterprises. The access path of the login node unit is one-way, and data cannot be downloaded from smart terminals within the control range of the data security supervision platform, which makes it easy to locate and search when data is damaged or leaked.

[0045] Embodiment 2:

[0046] When non-controlled smart terminals outside the data security supervision platform access data, there are many unknown conditions, so the identity authentication requirements are high;

[0047] The node management unit is used to collect node information and send it to the security management unit. The node information security identification and allocation management operations of the security management unit in the data security supervision platform include the identity information of the login user, the initial and end time of the connection, the number of failed connections and the remaining memory of the connection node.

[0048] The time difference between the connection start time and the connection success time is calculated to obtain the single connection duration, and the single connection duration is compared with the preset duration threshold. When the single connection duration is greater than the preset duration threshold, the difference between the two is calculated to obtain the single overtime duration, and the single overtime duration is calibrated as Dc, the number of failed connections is calibrated as Wt, and the remaining memory of the connection node is calibrated as Sy;

[0049] Step 2: According to the formula: Z = a1*DC + (a2Wt / a3Sy), where the risk scheduling coefficient Z, a1, a2 and a3 are correction factors, and a2>a3>a1>0;

[0050] Compare the risk scheduling coefficient Z with the risk scheduling coefficient threshold z:

[0051] When Z ≥ z, a first-level scheduling signal is generated;

[0052] When Z<z, a secondary scheduling signal is generated;

[0053] Step 3: When a first-level dispatch signal is generated in the security management unit, a "pay attention to login time" mark is actively generated and fed back to the smart terminal that is not supervised by the data security supervision platform to remind the login user to pay attention to the login time;

[0054] When the number of first-level dispatch signal generation times in the security management unit is ≥5, the simulation stack data is actively generated and sent to the database, where the simulation stack data is the simulation combination chain parsing data generated by artificial intelligence, and the parsed simulation combination chain data is sent to the external registration node, where the external registration node is an intelligent terminal not monitored by the data security supervision platform, to prevent the external registration node from receiving malicious data explosion and complete the protection of the external registration node;

[0055] Step 4: When the secondary scheduling signal is generated in the security management unit, a "multi-angle face extraction and recognition request" is actively generated on the external registration node. After obtaining permission, the face image is collected and sent to the node management unit for storage and automatic generation of precise stack data. The precise stack data is the parsed data generated by the security management unit based on the combined chain input by the external registration node, and is fed back to the external registration node to obtain the internal data of the enterprise. Prior to this, the enterprise and the partners obtained the combined chain only after signing a data confidentiality agreement, and the personnel information of the partners would be stored in the enterprise database. Therefore, the responsibility for data leakage problems at the external registration node lies with the partners.

[0056] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A network big data security transmission management system based on artificial intelligence, characterized by: It includes a data security supervision platform and a data terminal, wherein the data terminal is used to encrypt the big data to be stored to obtain encrypted data and an encryption chain, and send the encrypted data and the encryption chain to the data security supervision platform; The data security supervision platform is used to store and process encrypted data and encryption chains, and includes a data distribution unit, a database, a security management unit and a node management unit; The data distribution unit distributes the encrypted data to the storage node for storage, and at the same time merges the encryption chain corresponding to the encrypted data with the node chain in the storage node storing the encrypted data to obtain a combined chain and sends it to the database for storage; The node management unit is used to collect node information and send the node information to the security management unit; The security management unit is used for security identification and allocation management of node information and generates a scheduling signal. At the same time, the security management unit generates accurate stack data or simulated stack data after preprocessing the scheduling signal and sends it to the database; The database is used to store encrypted data, combination chains, smart terminal parameters and external personnel information.

2. The artificial intelligence-based network big data security transmission management system according to claim 1 is characterized by: The data end performs encryption processing on the big data to be stored, specifically including: Step 1: Perform character and image recognition on the stored big data, set unique character values ​​corresponding to all characters, match the obtained character strings with their corresponding unique character values ​​in order to obtain the character string values; set unique color values ​​corresponding to all colors, magnify the obtained image along the center position to obtain a pixel grid, and match the obtained pixel grid with its corresponding unique color values ​​in order to obtain the color pixel map values; Step 2: According to the formula Y=Ex+326, E is the corresponding string value or color pixel value in the stored big data, x is a manually set parameter, and Y is the combined value; Step 3: Convert the combined value into a graphic value, obtain the combined value digit value, and construct an equipotential polygon according to the combined value digit value, fill the equipotential polygon with a sequential graphic according to the intersection points formed by two line segments of the equipotential polygon, set a sequential graphic corresponding to each of the digits from zero to nine, print the sequential graphics corresponding to the combined value at the intersection points of the equipotential polygons, obtain an encrypted graphic string, and obtain the encrypted data according to the encrypted graphic string; Step 4: Separate and save the encrypted graphic strings obtained by the string value and the color pixel image value according to the storage time.

3. The network big data security transmission management system based on artificial intelligence according to claim 2 is characterized by: The data security supervision platform also includes a login node unit, which is used for the owners of smart terminals within the control range of the data security supervision platform to submit smart terminal parameters through the smart terminal for registration, and send the successfully registered smart terminal parameters to the database for storage. The database marks the successfully registered smart terminal as an internal test node, and generates a unique verification number for the internal test node, and marks the smart terminal parameters as internal test parameters, where the internal test parameters include registration time, location and registration number.

4. The artificial intelligence-based network big data security transmission management system according to claim 3 is characterized by: The node information includes the login identity information, the initial and end time of the connection, the number of failed connections and the remaining memory of the connection node.

5. The artificial intelligence-based network big data security transmission management system according to claim 1 is characterized by: The node information security identification and allocation management operations of the security management unit specifically include: Step 1: Calculate the time difference between the connection start time and the connection success time to obtain the single connection duration, compare the single connection duration with the preset duration threshold, and when the single connection duration is greater than the preset duration threshold, calculate the difference between the two to obtain the single overtime duration, and mark the single overtime duration as Dc, the number of failed connections as Wt, and the remaining memory of the connection node as Sy; Step 2: According to the formula: Z = a1*DC + (a2Wt / a3Sy), where the risk scheduling coefficient Z, a1, a2 and a3 are correction factors, and a2>a3>a1>0; Compare the risk scheduling coefficient Z with the risk scheduling coefficient threshold z: When Z ≥ z, a first-level scheduling signal is generated; When Z<z, a secondary scheduling signal is generated; Step 3: When a first-level dispatch signal is generated in the security management unit, a "pay attention to login time" mark is actively generated and fed back to the smart terminal that is not supervised by the data security supervision platform; When the number of first-level dispatch signal generation in the security management unit is ≥5, the simulation stack data is actively generated and sent to the database, where the simulation stack data is the simulation combination chain parsing data generated by artificial intelligence, and the parsed simulation combination chain data is sent to the external registration node, where the external registration node is a smart terminal that is not monitored by the data security supervision platform; Step 4: When the secondary scheduling signal is generated in the security management unit, a "multi-angle face extraction and recognition request" is actively generated on the external registration node. After obtaining permission and collecting the face image, the precise stack data is automatically generated. The precise stack data is the combined chain parsed data input by the security management unit based on the external registration node, and is fed back to the external registration node.

6. The artificial intelligence-based network big data security transmission management system according to claim 1 is characterized by: The encryption chain is composed of the level and search directory of the encrypted data; the node chain is composed of the authentication number of the storage node and the corresponding character.

Images