An information interaction method based on deep learning

Abstract

The present invention discloses an information interaction method based on deep learning, belonging to the field of information processing technology, comprising the following steps: S1, transmitting a data packet from a user terminal to a server, processing each data block of the data packet using a deep learning model, and generating a transmission necessary value for the data packet; S2, obtaining a stability matrix for the data packet based on IP datagrams between the user terminal and the server and the transmission necessary value for the data packet; S3, generating an information interaction key for the server's data packet based on the stability matrix, thereby completing the information interaction. The generation of the information interaction key in the present invention is closely related to the data packet transmission process, making it difficult for attackers to predict and crack the key, thereby improving the security of data transmission; and also simplifying the complexity of key management, reducing the cost and workload of key management.

Description

Technical Field

[0001] The present invention belongs to the field of information processing technology, and specifically relates to an information interaction method based on deep learning. Background Art

[0002] With the continuous development of terminal technology in recent years, not only has it continued to expand within its existing application areas, it has also entered new fields, giving rise to tablet computers, smartphones, and netbooks. With the widespread adoption of smart terminals, data exchange between them and servers is becoming increasingly frequent. However, ensuring data security, integrity, and tamper-proofing during data transmission has become a pressing issue. Existing secure exchange methods often have security vulnerabilities, making them vulnerable to hacker attacks and data leaks. Therefore, a more secure and reliable information exchange method is needed. Summary of the Invention

[0003] In order to solve the above problems, the present invention proposes an information interaction method based on deep learning.

[0004] The technical solution of the present invention is: an information interaction method based on deep learning comprises the following steps:

[0005] S1. Transmit the data packet from the user terminal to the server, use the deep learning model to process each data block of the data packet, and generate the necessary value for the transmission of the data packet;

[0006] S2. Obtaining a stability matrix of the data packet based on the IP datagram and the transmission necessary value of the data packet between the user terminal and the server;

[0007] S3. Generate an information interaction key for the server's data packet based on the stability matrix to complete the information interaction.

[0008] Furthermore, S1 includes the following sub-steps:

[0009] S11, splitting the data packet into several data blocks;

[0010] S12. Calculate the transmission carrying value for each data block using each neuron of the deep learning model;

[0011] S13. Generate a transmission necessary value for the data packet according to the transmission carrying value of each data block.

[0012] The beneficial effect of the above further scheme is: in the present invention, after the data packet is split, each data block can be accompanied by a checksum or other verification information, and the deep learning model can automatically extract useful feature representations from the data block, that is, the transmission carrying value; by analyzing the transmission carrying value of the data block, the necessary transmission value of the data packet can be generated for data key generation.

[0013] Furthermore, in S12, the transmission of the data block carries the value The calculation formula is:

[0014] ;

[0015] Where, represents the synaptic weight of the neuron corresponding to the data block, Represents the checksum of the data block, Indicates the priority weight of the data block.

[0016] The further embodiment of the present invention provides the following advantages: In the present invention, the data packet includes priority information, so each data block retains a corresponding priority, and the priority can be assigned a corresponding weight. Checksums play an important role in data processing and data communication. They are used to verify the sum of a group of data items at the destination to ensure data integrity and accuracy.

[0017] Furthermore, in S13, the necessary value is transmitted The calculation formula is:

[0018] ;

[0019] Where, Indicates the total number of data blocks, Indicates the The transmission of data blocks carries value, Indicates rounding operation. Indicates taking the maximum value.

[0020] Furthermore, S2 includes the following sub-steps:

[0021] S21, obtaining IP datagrams between the user terminal and the server;

[0022] S22, generating a data reassembly amount based on the fragment offset of each fragment in the IP datagram and the transmission necessary value of the data packet;

[0023] S23. Using the data packet loss rate and retransmission rate to correct the data reassembly amount, a stable matrix of the data packet is obtained.

[0024] The above further embodiment has the following beneficial effects: In the present invention, IP datagrams encapsulate the data to be transmitted, including important information such as the source address, destination address, and data, which is crucial for subsequent data processing and transmission. The data repackaging amount is generated based on the precise slice offset and transmission required value of each slice. This generates a more stable data packet transmission matrix, maintaining the stability and security of data transmission.

[0025] Furthermore, in S22, the data reload amount The calculation formula is:

[0026] ;

[0027] Where, Indicates the necessary value for transmission. Indicates the first The fragment offset of each fragment, Indicates the total number of fragments of an IP datagram.

[0028] An IP datagram is a packet transmitted by the Internet's Transmission Control Protocol. The fragment offset is a numeric value that indicates the relative position of a fragment within the original IP datagram. This parameter can be dimensionless if necessary.

[0029] Furthermore, in S23, the stability matrix of the data packet The expression is:

[0030] ;

[0031] Where, Indicates the amount of data reloaded, Indicates the packet loss rate, Indicates the packet retransmission rate.

[0032] Furthermore, S3 includes the following sub-steps:

[0033] S31, using the MD5 function to perform a hash operation on the maximum singular value of the stable matrix;

[0034] S32. Generate an information exchange key based on the hash operation result.

[0035] The above further solution has the following beneficial effects: In the present invention, MD5 is a widely used hash algorithm that can convert data of any length into a fixed-length (128-bit) hash value. The information exchange key combines the hash value and mathematical operations, increasing the complexity of the key.

[0036] Furthermore, in S31, the information exchange key The expression is: Where, Represents the hash operation result, Indicates rounding up.

[0037] The beneficial effects of the present invention are as follows: the present invention takes into account the powerful data processing capabilities of deep learning models, and utilizes deep learning models to quickly analyze each data block in a data packet, extract key information, and generate necessary transmission values; by comprehensively considering the necessary transmission values ​​of IP datagrams and data packets, a stable matrix reflecting the characteristics of the data packet can be obtained; information interaction keys are generated by the stable matrix, so that information security during data transmission can be ensured; the generation of information interaction keys in the present invention is closely related to the transmission process of the data packet, making it difficult for attackers to predict and crack the keys, thereby improving the security of data transmission; it also simplifies the complexity of key management and reduces the cost and workload of key management. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 Flowchart of the information interaction method based on deep learning. DETAILED DESCRIPTION

[0039] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0040] like Figure 1 As shown, the present invention provides an information interaction method based on deep learning, comprising the following steps:

[0041] S1. Transmit the data packet from the user terminal to the server, use the deep learning model to process each data block of the data packet, and generate the necessary value for the transmission of the data packet;

[0042] S2. Obtaining a stability matrix of the data packet based on the IP datagram and the transmission necessary value of the data packet between the user terminal and the server;

[0043] S3. Generate an information interaction key for the server's data packet based on the stability matrix to complete the information interaction.

[0044] In this embodiment of the present invention, S1 includes the following sub-steps:

[0045] S11, splitting the data packet into several data blocks;

[0046] S12. Calculate the transmission carrying value for each data block using each neuron of the deep learning model;

[0047] S13. Generate a transmission necessary value for the data packet according to the transmission carrying value of each data block.

[0048] In the present invention, after the data packet is split, each data block can be accompanied by a checksum or other verification information, and the deep learning model can automatically extract useful feature representations from the data block, that is, the transmission carrying value; by analyzing the transmission carrying value of the data block, the necessary transmission value of the data packet can be generated for data key generation.

[0049] In the embodiment of the present invention, in S12, the transmission carrying value of the data block The calculation formula is:

[0050] ;

[0051] Where, represents the synaptic weight of the neuron corresponding to the data block, Represents the checksum of the data block, Indicates the priority weight of the data block.

[0052] In the present invention, data packets contain priority information, so each data block retains a corresponding priority, and the priority can be assigned a corresponding weight. Checksums play an important role in data processing and data communication. They are used to verify the sum of a group of data items at the destination to ensure data integrity and accuracy.

[0053] The priority information may refer to the importance or urgency of the data block during transmission or processing, and the priority weight may also reflect the priority level.

[0054] In the embodiment of the present invention, in S13, the necessary value is transmitted The calculation formula is:

[0055] ;

[0056] Where, Indicates the total number of data blocks, Indicates the The transmission of data blocks carries value, Indicates rounding operation. Indicates taking the maximum value.

[0057] In this embodiment of the present invention, S2 includes the following sub-steps:

[0058] S21, obtaining IP datagrams between the user terminal and the server;

[0059] S22, generating a data reassembly amount based on the fragment offset of each fragment in the IP datagram and the transmission necessary value of the data packet;

[0060] S23. Using the data packet loss rate and retransmission rate to correct the data reassembly amount, a stable matrix of the data packet is obtained.

[0061] In this invention, IP datagrams encapsulate the data to be transmitted, including important information such as the source address, destination address, and data, which is crucial for subsequent data processing and transmission. The data repackaging amount is generated based on the precise fragment offset and transmission required value of each fragment. This generates a more stable data packet transmission matrix, maintaining the stability and security of data transmission.

[0062] In the embodiment of the present invention, in S22, the data reload amount The calculation formula is:

[0063] ;

[0064] Where, Indicates the necessary value for transmission. Indicates the first The fragment offset of each fragment, Indicates the total number of fragments of an IP datagram.

[0065] An IP datagram is a packet transmitted by the Internet's Transmission Control Protocol. The fragment offset is a numeric value that indicates the relative position of a fragment within the original IP datagram. This parameter can be dimensionless if necessary.

[0066] In the embodiment of the present invention, in S23, the stability matrix of the data packet The expression is:

[0067] ;

[0068] Where, Indicates the amount of data reloaded, Indicates the packet loss rate, Indicates the packet retransmission rate.

[0069] In this embodiment of the present invention, S3 includes the following sub-steps:

[0070] S31, using the MD5 function to perform a hash operation on the maximum singular value of the stable matrix;

[0071] S32. Generate an information exchange key based on the hash operation result.

[0072] In the present invention, MD5 is a widely used hash algorithm that can convert data of any length into a fixed-length (128-bit) hash value. The information exchange key combines the hash value and mathematical operations, increasing the complexity of the key.

[0073] In the embodiment of the present invention, in S31, the information exchange key The expression is: Where, Represents the hash operation result, Indicates rounding up.

[0074] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. An information interaction method based on deep learning, characterized in that: The following steps are involved: S1. Transmit the data packet from the user terminal to the server, use the deep learning model to process each data block of the data packet, and generate the necessary value for the transmission of the data packet; S2. Obtaining a stability matrix of the data packet based on the IP datagram and the transmission necessary value of the data packet between the user terminal and the server; S3. Generate an information interaction key for the server's data packet based on the stability matrix to complete the information interaction; The S1 comprises the following sub-steps: S11, splitting the data packet into several data blocks; S12. Calculate the transmission carrying value for each data block using each neuron of the deep learning model; S13, generating a transmission necessary value for the data packet according to the transmission carrying value of each data block; In the above S13, the necessary value is transmitted The calculation formula is: ； Where, Indicates the total number of data blocks, Indicates the The transmission of data blocks carries value, Indicates rounding operation. Indicates taking the maximum value; The S2 includes the following sub-steps: S21, obtaining IP datagrams between the user terminal and the server; S22, generating a data reassembly amount based on the fragment offset of each fragment in the IP datagram and the transmission necessary value of the data packet; S23, using the data packet loss rate and retransmission rate to correct the data reassembly amount to obtain a stable matrix of the data packet; In the step S22, the data reload amount The calculation formula is: ； Where, Indicates the necessary value for transmission. Indicates the first The fragment offset of each fragment, Indicates the total number of fragments of the IP datagram; In S23, the stability matrix of the data packet The expression is: ； Where, Indicates the amount of data reloaded, Indicates the packet loss rate, Indicates the packet retransmission rate.

2. The information interaction method based on deep learning according to claim 1, characterized in that: In the S12, the transmission of the data block carries the value The calculation formula is: ； Where, represents the synaptic weight of the neuron corresponding to the data block, Represents the checksum of the data block, Indicates the priority weight of the data block.

3. The information interaction method based on deep learning according to claim 1, characterized in that: The S3 includes the following sub-steps: S31, using the MD5 function to perform a hash operation on the maximum singular value of the stable matrix; S32. Generate an information exchange key based on the hash operation result.

4. The information interaction method based on deep learning according to claim 3, characterized in that: In S31, the information exchange key The expression is: Where, Represents the hash operation result, Indicates rounding up.

Images