Secure transmission method for 5G communication technology

Abstract

The invention discloses a secure transmission method for a 5G communication technology, and the method comprises the steps: carrying out the sensitivity evaluation, feature extraction, classification labeling and data block adjustment of to-be-transmitted data, and generating a data block set; generating a key in combination with a key exchange protocol, and selecting a proper encryption algorithm to encrypt the data blocks according to key characteristics to form an encrypted data block set; through preliminary authentication, hash chain verification in a transmission process and integrity verification at the end of transmission, security verification of identity and data is realized, and an authenticated encrypted data block is output; performing homomorphic encryption processing or data mask protection on the data stream to obtain an encrypted protection data stream, and transmitting the encrypted protection data stream; and monitoring the transmitted protection data stream in real time, predicting the traffic and an anomaly score through a traffic anomaly identification model based on the LSTM, and judging whether an anomaly exists based on a set self-adaptive risk threshold so as to perform source tracing and attack blocking.

Description

Technical Field

[0001] The present invention relates to the technical field of communication information security, and in particular to a secure transmission method for 5G communication technology. Background Art

[0002] With the rapid development of 5G communication technology, the network's high bandwidth, low latency and large-scale connection characteristics make 5G an important driving force for the digital transformation of various industries. However, the widespread application of 5G technology is also accompanied by many challenges in information security.

[0003] At present, traditional transmission security technology mainly relies on the encryption processing of data and the security protocol constraints during the transmission process. This method can prevent ordinary data theft and tampering to a certain extent. However, in the face of increasingly complex network attack methods, such as man-in-the-middle attacks, side-channel attacks and decryption threats brought by quantum computing, the existing security transmission methods have certain deficiencies in ensuring data integrity and privacy; in addition, due to the multi-level and heterogeneous nature of the 5G network architecture, the security verification and authentication of data during transmission also face higher challenges. Especially under the real-time requirements of data transmission, the existing security strategies are difficult to balance security and efficiency, which often leads to excessive complexity of encryption algorithms or long transmission authentication time, affecting the overall performance of data transmission.

[0004] In the existing technology, most encryption methods fail to effectively combine the sensitivity and feature information of the data for flexible encryption and classified protection. Instead, they adopt a unified encryption strategy, which not only causes over-protection of highly sensitive data and security risks of low-sensitivity data, but also increases the computational burden and difficulty of data recovery during the decryption process. In addition, anomaly detection and threat identification during transmission mainly rely on preset rules or fixed algorithms, which make it difficult to respond flexibly to rapidly changing attack patterns. Therefore, the existing technology has deficiencies in data sensitivity assessment, flexibility of transmission encryption, reliability of authentication and verification mechanisms, and real-time monitoring of the transmission process. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed.

[0007] Therefore, the technical problems solved by the present invention are: insufficient data sensitivity assessment, poor encryption flexibility and low transmission authentication reliability in the existing 5G transmission process.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: sensitivity assessment, feature extraction, classification annotation and data block adjustment are performed on the data to be transmitted to generate a data block set including sensitivity identification and feature annotation;

[0009] Generate a key in combination with a key exchange protocol, and select a suitable encryption algorithm according to key characteristics to encrypt the data blocks in the data block set to form an encrypted data block set;

[0010] By performing preliminary authentication on the encrypted data block set, hash chain verification during transmission, and integrity verification at the end of transmission, security verification of identity and data is achieved, and authenticated encrypted data blocks are output;

[0011] Performing homomorphic encryption processing or data mask protection on the authenticated encrypted data block to obtain an encrypted protected data stream, and transmitting the encrypted protected data stream;

[0012] The transmitted protection data flow is monitored in real time, the traffic flow and anomaly score are predicted through the LSTM-based traffic anomaly recognition model, and whether there is an anomaly is determined based on the set adaptive risk threshold to perform source tracing and attack blocking.

[0013] As a preferred solution of the secure transmission method for 5G communication technology according to the present invention, generating the data block set includes:

[0014] Receive a data stream to be transmitted in the 5G network, perform sensitivity analysis on the data to be transmitted through a sensitivity assessment model, output a sensitivity score, and divide the data to be transmitted into three levels of high sensitivity, medium sensitivity, and low sensitivity according to the sensitivity score;

[0015] Using an algorithm based on word frequency-inverse document frequency, extracting keywords from the data content of the data to be transmitted, and using the extracted keywords as data features;

[0016] According to the sensitivity level and network status, the data block size of the data to be transmitted is adjusted through the priority dynamic block adjustment algorithm, and the adjusted data block set Adjusted is output. D ata B locks, which contain sensitivity flags, feature keywords, and adjusted data blocks.

[0017] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the sensitivity score ranges from 0 to 1, where:

[0018] When the sensitivity score S>0.7, it is highly sensitive;

[0019] When 0.4≤sensitivity score S≤0.7, it is medium sensitivity;

[0020] When the sensitivity score S<0.4, it is low sensitivity.

[0021] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the encrypted data block set is formed, including:

[0022] Generate a key pair using the key exchange protocol (K A ,K B ), and encrypt and enhance the key through elliptic curve encryption algorithm;

[0023] Select the optimal encryption algorithm based on the sensitivity level and characteristic keywords of the data block;

[0024] Among them, the AES-256-GCM algorithm is selected for high-sensitivity data blocks, the ChaCha20 algorithm is selected for medium-sensitivity data blocks, and the AES-128 encryption algorithm is selected for low-sensitivity data blocks. The encrypted data block set Encrypted is output. D ata B locks, each data block contains the corresponding key and encryption algorithm identifier.

[0025] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the encryption enhancement of the key by using the elliptic curve encryption algorithm includes:

[0026] Generate the public keys of both parties (P A ,Q A ) and a private key pair (P B ,Q B ), wherein the two parties are a transmitting end and a receiving end;

[0027] The two parties generate a shared key K through public key exchange shared ,in,

[0028] The shared key K is encrypted using an elliptic curve encryption algorithm. shared Encryption is performed to form the final key K, and the output is Encrypted K ey.

[0029] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the output of the authenticated encrypted data block includes:

[0030] Before data transmission, the elliptic curve digital signature algorithm is used to authenticate the identities of both parties, and both parties generate signatures Sig A and Sig B , verified by the other party’s public key. If the verification fails, the system immediately terminates the transmission and triggers an alarm;

[0031] During the transmission process, the hash value H is calculated for each data block to form a hash chain;

[0032] After the transmission is completed, the receiving end verifies the hash chain. If the hash chain is broken, it is identified that the data packet has been tampered with;

[0033] The integrity of the transmitted data is then verified using the digital certificate at the receiving end. Once verified, the verified encrypted data block is output.

[0034] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the encrypted protected data stream includes:

[0035] For highly sensitive data blocks, the Paillier homomorphic encryption algorithm is used to encrypt them, and the homomorphic encrypted protected data stream is obtained. E ncrypted B locks;

[0036] For sensitive data, data masking technology is used to replace sensitive fields to obtain the data stream Masked protected by data mask D ata B locks;

[0037] For low-sensitivity data, the AES-128 algorithm is used for encryption, and the calculation method is:

[0038] C L =AES128_Encrypt(K AES128 ,D L )

[0039] Among them, C L Indicates the encrypted data block, and AES128_Encrypt indicates the AES-128 encryption operation.

[0040] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, a data flow anomaly recognition model is constructed based on the long short-term memory network algorithm to monitor the data flow processed by privacy protection in real time. The long short-term memory network algorithm identifies abnormal traffic by learning the time series changes of data traffic. The output of the data flow anomaly recognition model is the traffic anomaly score E. s core;

[0041] According to the statistical characteristics of historical traffic data, including mean and standard deviation, an adaptive risk threshold T is set. risk , the calculation formula is:

[0042] T risk =μ+k·σ

[0043] Among them, k is the sensitivity coefficient, ranging from 2 to 3, μ is the average value of historical flow, and σ is the standard deviation of flow.

[0044] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, the abnormal score E of the current time period is s core and adaptive risk threshold T risk For comparison, where:

[0045] If E s core>T risk , it is considered that abnormal traffic is detected and an abnormal alarm is triggered;

[0046] If E s core≤T risk , the traffic is considered normal and no alarm needs to be triggered.

[0047] As a preferred solution of the secure transmission method for 5G communication technology described in the present invention, after the alarm is triggered, the abnormal traffic is analyzed by using the tracing algorithm. a Identify the source of the attack, including:

[0048] Perform feature analysis on the abnormal traffic to generate a feature vector V a ;

[0049] Determine the feature vector V by similarity matching a The closest path P a ;

[0050] For the path P a Perform blocking operations and record attack source information.

[0051] Beneficial effects of the present invention:

[0052] 1. Improve the pertinence and flexibility of data encryption, reduce the encryption burden, and improve data transmission efficiency and security;

[0053] 2. Improved the security protection level of data, optimized the efficiency of encryption algorithms in different scenarios, ensured data confidentiality, and improved system flexibility and transmission efficiency;

[0054] 3. Through multi-level authentication and verification steps, it can effectively prevent the security threats of man-in-the-middle attacks and data tampering, realize the dual verification of identity and data, enhance the security of data transmission, and improve the overall reliability of the system;

[0055] 4. Through the combination of homomorphic encryption and mask protection, double encryption protection of sensitive data is achieved, which effectively prevents the risks of external theft and internal leakage and improves the security of data transmission;

[0056] 5. It enhances the dynamic security during data transmission, and can respond and cope with possible attacks in a timely manner, improving the overall security and defense capabilities of data transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0058] Figure 1 A schematic diagram of a flow chart of a secure transmission method for 5G communication technology shown in the present invention;

[0059] Figure 2 A schematic diagram of the relationship between the sizes of data blocks of different sensitivity data shown in the present invention;

[0060] Figure 3 It is a schematic diagram of the sensitivity scoring curve of each data shown in the present invention. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0062] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without making any creative work should fall within the scope of protection of the present invention.

[0063] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0064] According to an embodiment of the present invention, Figure 1The flowchart shown is a secure transmission method for 5G communication technology, which specifically includes the following steps:

[0065] S1. Perform sensitivity assessment, feature extraction, classification and labeling, and data block adjustment on the data to be transmitted to generate a data block set containing sensitivity identification and feature labeling. It should be noted that generating a data block set includes:

[0066] Receive the data stream to be transmitted in the 5G network, perform sensitivity analysis on the data to be transmitted through the sensitivity assessment model, output a sensitivity score, and divide the data to be transmitted into three levels of high sensitivity, medium sensitivity, and low sensitivity according to the sensitivity score;

[0067] As an example, the calculation formula for the output data sensitivity score S is:

[0068] S=α·C+β·P+γ·V

[0069] Among them, C represents the confidentiality of data, P represents the priority of data, V represents the risk of sensitive content exposure of data, and α, β, and γ are weight coefficients;

[0070] Furthermore, the sensitivity score ranges from 0 to 1, where:

[0071] When the sensitivity score S>0.7, it is highly sensitive;

[0072] When 0.4≤sensitivity score S≤0.7, it is medium sensitivity;

[0073] When the sensitivity score S<0.4, it is low sensitivity;

[0074] An algorithm based on word frequency-inverse document frequency is used to extract keywords from the data content of the data to be transmitted, and the extracted keywords are used as data features;

[0075] Calculate the term frequency tf and inverse document frequency idf of each data packet Di. The formula is as follows:

[0076]

[0077] Where w is the feature word, N is the total number of packets, and df(w) is the number of packets containing w;

[0078] Select the top k keywords with the highest tf_idf score as data feature output;

[0079] The data features and sensitivity levels will be annotated in the data packet and output as the annotated data packet set Data B locks, as input for subsequent steps;

[0080] According to the sensitivity level and network status, the data block size of the data to be transmitted is adjusted through the priority dynamic block adjustment algorithm, and the adjusted data block set Adjusted is output. D ata B locjs, which contains sensitivity flags, feature keywords, and adjusted data blocks.

[0081] S2. Generate a key in combination with the key exchange protocol, select a suitable encryption algorithm according to the key characteristics to encrypt the data blocks in the data block set, and form an encrypted data block set. Among them, what needs to be explained in this step is:

[0082] Generate a key pair using the key exchange protocol (K A ,K B ), and encrypt and enhance the key through elliptic curve encryption algorithm;

[0083] Key encryption is enhanced through elliptic curve cryptography, including:

[0084] Generate the public keys of both parties (P A ,Q A ) and a private key pair (P B ,Q B ), where the two parties are the transmitting end and the receiving end;

[0085] The two parties generate a shared key K through public key exchange shared ,in,

[0086] The shared key K is encrypted using the elliptic curve algorithm shared Encryption is performed to form the final key K, and the output is Encrypted K ey;

[0087] Select the optimal encryption algorithm based on the sensitivity level and characteristic keywords of the data block;

[0088] Among them, the AES-256-GCM algorithm is selected for high-sensitivity data blocks, the ChaCha20 algorithm is selected for medium-sensitivity data blocks, and the AES-128 encryption algorithm is selected for low-sensitivity data blocks. The encrypted data block set Encrypted is output. D ata B locks, each data block contains the corresponding key and encryption algorithm identifier.

[0089] S3: Perform preliminary authentication on the encrypted data block set, perform hash chain verification during transmission, and perform integrity verification at the end of transmission to achieve security verification of identity and data, and output authenticated encrypted data blocks. It should be noted that the output of authenticated encrypted data blocks includes:

[0090] Before data transmission, the elliptic curve digital signature algorithm is used to authenticate the identities of both parties, and both parties generate signatures Sig A and Sig B , both parties verify each other's signature through their respective public keys:

[0091] Verify(Sig,PK)=H(Data)=R+S

[0092] Among them, Sig represents the signature result, PK is the public key, H is the hash function, H(Data) is the result of the hash function H operation on the signature data Data, R is a signature parameter, which represents a randomly generated number, and it is the intermediate result of the signature algorithm. S is another important parameter of the ECDSA signature, which is associated with the private key d and R and is used to generate the signature;

[0093] If the verification fails, the system immediately aborts the transmission and triggers an alarm;

[0094] If the verification is successful, the data transmission verification will be started. During the transmission process, the hash value G(D n ), forming a hash chain H chain :

[0095] H chain =H(H(D1)+H(D2)+…+H(D n ))

[0096] The hash chain adds each hash value on the chain to form a unique checksum.

[0097] After the transmission is completed, the integrity of the transmitted data is verified through the digital certificate of the receiving end. chain The H calculated by the sender chain If they are consistent, the verification is passed and the verified encrypted data block is output;

[0098] If the hash chain is broken, that is, inconsistent, the data packet is identified as tampered with and the verification fails. The system records it and prevents further transmission.

[0099] S4, homomorphic encryption processing or data mask protection is performed on the authenticated encrypted data block to obtain an encrypted protected data stream, and transmit it. It should be noted that the encrypted protected data stream includes:

[0100] For highly sensitive data blocks, the Paillier homomorphic encryption algorithm is used to encrypt them, and the homomorphic encrypted protected data stream is obtained. E ncrypted B locks;

[0101] Exemplarily, define a plaintext data block as m, and its encrypted form as E(m), which satisfies the following properties:

[0102] E(m1+m2)=E(m1)×E(m2);

[0103] For sensitive data, data masking technology is used to replace sensitive fields with "******" to obtain the data stream Masked D ata B locks;

[0104] For low-sensitivity data, the AES-128 algorithm is used for encryption, and the calculation method is:

[0105] C L =AES128_Encrypt(K AES128 ,D L )

[0106] Among them, C L Indicates the encrypted data block, and AES128_Encrypt indicates the AES-128 encryption operation.

[0107] S5: Monitor the transmitted protection data flow in real time, predict the flow and anomaly score through the LSTM-based flow anomaly recognition model, and determine whether there is anomaly based on the set adaptive risk threshold to trace the source and block the attack. Among them, the following points need to be explained in this step:

[0108] Based on the long short-term memory network algorithm (LSTM), a data flow anomaly recognition model is built to monitor the privacy-protected data flow in real time. The long short-term memory network algorithm learns the time series changes of data flow and identifies abnormal traffic. The output of the data flow anomaly recognition model is the predicted traffic X. pred and anomaly score E s core;

[0109] The privacy-preserving data stream is segmented in chronological order to form a time series data set X = x1, x2, ..., x n , where each x n Indicates the data flow value of a time slice;

[0110] The model predicts future traffic values ​​by learning the time series characteristics of historical traffic. The input of LSTM is the time series dataset X. Each data point x n Represents the data flow within a period of time, and the output of LSTM is the predicted flow X pred And the anomaly score E_score, the calculation formula of the anomaly score E_score is:

[0111] E_score=|X t -X pred |

[0112] Among them, X is the time series data set, that is, the current real traffic, X pred To predict traffic, the anomaly score E_score represents the difference between the actual traffic and the predicted traffic;

[0113] According to the statistical characteristics of historical traffic data, including mean and standard deviation, an adaptive risk threshold T is set. risk , the calculation formula is:

[0114] T risk =μ+k·σ

[0115] Among them, k is the sensitivity coefficient, ranging from 2 to 3, μ is the average value of historical flow, and σ is the standard deviation of flow;

[0116] Adaptive risk threshold detection compares the anomaly score output by the model with the set risk threshold T risk , determine whether to trigger an abnormal alarm;

[0117] The anomaly score E for the current time period s core and adaptive risk threshold T risk For comparison, where:

[0118] If E s core>T risk , it is considered that abnormal traffic is detected and an abnormal alarm is triggered;

[0119] If E s core≤T risk , the traffic is considered normal and no alarm needs to be triggered;

[0120] After the alarm is triggered, the source tracing algorithm is used to analyze the abnormal traffic and the source tracing path P is used to a Identify the source of the attack, including:

[0121] Perform feature analysis on abnormal traffic and generate feature vector V a ;

[0122] Determine the feature vector V by similarity matching a The closest path P a ;

[0123] For path P a Perform blocking operations and record attack source information.

[0124] As an example, the communication system performs feature analysis on abnormal traffic and generates a feature vector V of the abnormal traffic.a ,The feature vector contains at least the IP address, port number, packet size, and transmission rate of the abnormal traffic.

[0125] As an example, using the feature vector V a Match it with the known attack signature library in the system to find the closest path P a , the similarity calculation formula of the path is:

[0126]

[0127] Among them, V k is the feature vector in the attack feature library, and Sim represents the cosine similarity between vectors.

[0128] As an example, find the most similar path P a After that, the system will locate the specific location of the attack source, such as the attacker's IP address and data packet path.

[0129] As an example, the system blocks the attack source. Specific operations include closing the corresponding network port and banning the attacking IP address. At the same time, the system will record the detailed information of the attack for subsequent analysis and adjustment of strategies.

[0130] It should be noted that the method of the present invention realizes flexible encryption and classification protection of data by performing sensitivity assessment and feature extraction on the data to be transmitted, and at the same time adopts a key exchange protocol to generate keys, and combines the data block adjustment strategy to further improve the transmission security and integrity verification of data; through preliminary authentication, hash chain verification during transmission, integrity verification at the end of transmission and homomorphic encryption and other multiple security measures, the credibility of identity and data is ensured; in addition, an adaptive anomaly detection algorithm is used to monitor the protection data stream in real time, effectively identify potential threats and trace and block attacks in a timely manner.

[0131] Preferably, in order to verify the effectiveness of the multi-level data security transmission method proposed in the embodiment of the present invention, this embodiment selects a typical 5G network environment as an experimental platform. The experimental environment includes a transmission end and a receiving end. Both sides are configured with servers that support elliptic curve encryption algorithm and long short-term memory network (LSTM) model. Various types of data streams are used in the experiment, including financial transaction data (Data A )、User Browsing Records (Data B )、News Information C ), social media information (Data D )、Positioning Information(Data E ), sensor data F ) and system log (Data G ).

[0132] The data stream to be transmitted is preprocessed, and the sensitivity evaluation model is used to perform sensitivity scoring on each data packet. The calculation formula is:

[0133] S=αC+βP+γV

[0134] Among them, C represents the confidentiality of the data, P represents the priority of the data, V represents the risk of sensitive content exposure of the data, α, β, and γ are weight coefficients. In this experiment, the weight coefficients are set to α = 0.5, β = 0.3, and γ = 0.2;

[0135] According to the calculation, the sensitivity score of each data is obtained: Data A is 0.85, Data B is 0.65, Data C is 0.30, Data D is 0.55, Data E is 0.90, Data F is 0.40, Data G is 0.70; according to the score, Data A and Data E Classified as high-sensitivity data, Data B 、Data D and Data G Classified as medium-sensitivity data, Data C and Data F Classified as low-sensitivity data;

[0136] The algorithm based on term frequency-inverse document frequency (TF-IDF) is used to extract keywords from the data content. i , calculate the term frequency tf and inverse document frequency idf, the formula is as follows:

[0137]

[0138] Among them, w is the feature word, N is the total number of data packets, df(w) is the number of data packets containing w, and the top k keywords with the highest scores are selected as data feature outputs, for example, Data A Five keywords including "account", "password" and "transaction amount" were extracted. B Four keywords were extracted: Data C 3 keywords were extracted, and so on;

[0139] According to the sensitivity level and network status, the data block size is adjusted through the priority dynamic block adjustment algorithm. The high-sensitivity data block is adjusted to 512 bytes, the medium-sensitivity data block is adjusted to 1024 bytes, and the low-sensitivity data block is adjusted to 2048 bytes. This adjustment ensures the security of high-sensitivity data and the transmission efficiency of low-sensitivity data;

[0140] In the key generation phase, the transmission end and the receiving end generate a key pair using the key exchange protocol, and encrypt and enhance the key using the elliptic curve encryption algorithm. Both parties generate their own public and private keys, and generate a shared key K through public key exchange. shared , and further encrypted to form the final key K;

[0141] According to the sensitivity level of the data block, select the appropriate encryption algorithm. A 、Data E ) uses AES-256-GCM algorithm, medium sensitivity data (Data B 、Data D 、Data G ) uses ChaCha20 algorithm, low sensitivity data (Data C 、Data F ) Using the AES-128 algorithm, each data block in the encrypted data block set contains a corresponding key and encryption algorithm identifier;

[0142] Before data transmission, the elliptic curve digital signature algorithm is used to authenticate the identities of both parties, generate a signature Sig, and verify the signature of the other party through their respective public keys. If the verification is successful, the data transmission process begins; otherwise, the transmission is terminated and an alarm is triggered;

[0143] During the transmission process, the hash value is calculated for each data block to form a hash chain:

[0144] H chain =H(H(D1)+H(D2)++H(D n ))

[0145] After the transmission is completed, the integrity of the transmitted data is verified through the digital certificate of the receiving end to ensure that the data has not been tampered with;

[0146] For highly sensitive data blocks (Data A 、Data E ), and further encrypt it using the Paillier homomorphic encryption algorithm to satisfy the homomorphic property:

[0147] E(m1)×E(m2)=E(m1+m2)

[0148] Medium Sensitivity Data B、Data D 、Data G )Use data masking technology to replace sensitive fields with "******", and low-sensitivity data (Data C 、Data F ) Continue to use AES-128 algorithm encryption;

[0149] During the data transmission process, the LSTM-based traffic anomaly recognition model is used to monitor the data flow in real time, and the privacy-protected data flow is segmented in chronological order to form a time series data set, which is input into the LSTM model to predict future traffic values. And calculate the anomaly score and set the adaptive risk threshold based on historical traffic data;

[0150] If the anomaly score exceeds the threshold, an anomaly alarm is triggered, and a tracing algorithm is used to locate and block the attack source.

[0151] The experimental results are shown in the following table:

[0152] Table 1. Multi-level data security transmission experiment data table

[0153]

[0154]

[0155] Referring to Table 1, it can be seen intuitively that the sensitivity score of the method of the present invention reflects the accurate assessment of the sensitivity of the data. For example, Data A and Data E The scores of Data C and Data F The scores of the two data types are lower, 0.30 and 0.40 respectively, and they are classified as low-sensitivity data. This accurate classification helps to take targeted encryption and protection measures in subsequent steps.

[0156] Preferably, in order to intuitively demonstrate the transmission of data with different sensitivities, the simulation experiment generates a schematic diagram of the relationship between the sensitivity score and the data block size through MATLAB, such as Figure 2 As shown in the figure, different types of lines are used to represent data transmission trends of different sensitivity levels. The solid line represents high-sensitivity data, the dotted line represents medium-sensitivity data, and the dotted line represents low-sensitivity data.

[0157] Reference Figure 2 , it can be seen intuitively that high sensitivity data (Data A 、Data E) corresponds to a smaller block size of 512 bytes, which is beneficial to improve security. B 、Data D 、Data G ) block size is 1024 bytes, taking into account both security and efficiency, low-sensitivity data (Data C 、Data F ) The block size is relatively large, 2048 bytes, which improves the transmission efficiency. This dynamic adjustment strategy of data block size enables the system to achieve a good balance between security and performance.

[0158] Preferably, in order to further demonstrate the effectiveness of sensitivity scoring for data classification processing, based on the experimental data in Table 1, a sensitivity scoring curve diagram of each data is also generated, such as Figure 3 shown.

[0159] Reference Figure 3 , you can intuitively see the distribution of sensitivity scores for each data, such as Data A 、Data E The sensitivity scores of Data were 0.85 and 0.90, which were located at a higher position in the curve graph, indicating high sensitivity; B 、Data D 、Data G The sensitivity scores of Data were 0.65, 0.55, and 0.70, respectively, which were located in the middle area of ​​the curve, indicating that these data have a moderate degree of sensitivity; C 、Data F The sensitivity scores of are low, 0.30 and 0.40, respectively, which are located in the lower area of ​​the curve, which justifies their classification as low sensitivity data.

[0160] The curve chart clearly shows the sensitivity score of each data, verifying the effectiveness of the sensitivity assessment model. The score of high-sensitivity data is significantly higher than that of medium and low-sensitivity data, supporting the accuracy of data classification. The sensitivity score curve chart intuitively demonstrates the advantages of the present invention in data sensitivity assessment. Compared with the existing technology, it can accurately assess the sensitivity of data and help to achieve refined security management.

[0161] Preferably, the anomaly scores of all data are lower than the set threshold (assuming the threshold is 0.1), and no alarm is triggered, indicating that the system is stable.

[0162] Compared with traditional methods, the present invention has achieved significant improvements in data security and transmission efficiency. For example, in terms of data block size adjustment, the method of the present invention can dynamically adjust according to the sensitivity level, thereby improving the security of high-sensitivity data and the transmission efficiency of low-sensitivity data. The LSTM-based traffic anomaly recognition model effectively monitors the data transmission process without false alarms or omissions, thus providing a guarantee for secure data transmission in the 5G network environment.

[0163] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A secure transmission method for 5G communication technology, characterized in that: include: Perform sensitivity assessment, feature extraction, classification annotation and data block adjustment on the data to be transmitted, and generate a data block set containing sensitivity identification and feature annotation; Generate a key in combination with a key exchange protocol, and select a suitable encryption algorithm according to key characteristics to encrypt the data blocks in the data block set to form an encrypted data block set; By performing preliminary authentication on the encrypted data block set, hash chain verification during transmission, and integrity verification at the end of transmission, security verification of identity and data is achieved, and authenticated encrypted data blocks are output; Performing homomorphic encryption processing or data mask protection on the authenticated encrypted data block to obtain an encrypted protected data stream, and transmitting the encrypted protected data stream; The transmitted protection data flow is monitored in real time, the traffic flow and anomaly score are predicted through the LSTM-based traffic anomaly recognition model, and whether there is an anomaly is determined based on the set adaptive risk threshold to perform source tracing and attack blocking.

2. The secure transmission method for 5G communication technology according to claim 1, characterized in that: Generating the data block set includes: Receive a data stream to be transmitted in the 5G network, perform sensitivity analysis on the data to be transmitted through a sensitivity assessment model, output a sensitivity score, and divide the data to be transmitted into three levels of high sensitivity, medium sensitivity, and low sensitivity according to the sensitivity score; Using an algorithm based on word frequency-inverse document frequency, extracting keywords from the data content of the data to be transmitted, and using the extracted keywords as data features; According to the sensitivity level and network status, the data block size of the data to be transmitted is adjusted through the priority dynamic block adjustment algorithm, and the adjusted data block set Adjusted is output. D ata B locks, which contain sensitivity flags, feature keywords, and adjusted data blocks.

3. The secure transmission method for 5G communication technology according to claim 2, characterized in that: The sensitivity score ranges from 0 to 1, where: When the sensitivity score S>0.7, it is highly sensitive; When 0.4≤sensitivity score S≤0.7, it is medium sensitivity; When the sensitivity score S<0.4, it is low sensitivity.

4. The secure transmission method for 5G communication technology according to claim 1, characterized in that: The encrypted data block set is formed, comprising: Generate a key pair using the key exchange protocol (K A , K B ), and encrypt and enhance the key through elliptic curve encryption algorithm; Select the optimal encryption algorithm based on the sensitivity level and characteristic keywords of the data block; Among them, the AES-256-GCM algorithm is selected for high-sensitivity data blocks, the ChaCha20 algorithm is selected for medium-sensitivity data blocks, and the AES-128 encryption algorithm is selected for low-sensitivity data blocks. The encrypted data block set Encrypted is output. D ata B locks, each data block contains the corresponding key and encryption algorithm identifier.

5. The secure transmission method for 5G communication technology according to claim 4, characterized in that: The encryption enhancement of the key by using the elliptic curve encryption algorithm includes: Generate the public keys of both parties (P A , Q A ) and a private key pair (P B , Q B ), wherein the two parties are a transmitting end and a receiving end; The two parties generate a shared key K through public key exchange shared ,in, The shared key K is encrypted using an elliptic curve encryption algorithm. shared Encryption is performed to form the final key K, and the output is Encrypted K ey.

6. The secure transmission method for 5G communication technology according to claim 1, characterized in that: The output authenticated encrypted data block includes: Before data transmission, the elliptic curve digital signature algorithm is used to authenticate the identities of both parties, and both parties generate signatures Sig A and Sig B , verified by the other party’s public key. If the verification fails, the system immediately terminates the transmission and triggers an alarm; During the transmission process, the hash value H is calculated for each data block to form a hash chain; After the transmission is completed, the receiving end verifies the hash chain. If the hash chain is broken, it is identified that the data packet has been tampered with; The integrity of the transmitted data is then verified using the digital certificate at the receiving end. Once verified, the verified encrypted data block is output.

7. The secure transmission method for 5G communication technology according to claim 1, characterized in that: The encrypted protected data stream includes: For highly sensitive data blocks, the Paillier homomorphic encryption algorithm is used to encrypt them, and the homomorphic encrypted protected data stream is obtained. E ncrypted B locks; For sensitive data, data masking technology is used to replace sensitive fields to obtain the data stream Masked protected by data mask D ata B locks; For low-sensitivity data, the AES-128 algorithm is used for encryption, and the calculation method is: C L =AES128_Encrypt(K AES128 ,D L ) Among them, C L Indicates the encrypted data block, and AES128_Encrypt indicates the AES-128 encryption operation.

8. The secure transmission method for 5G communication technology according to claim 1 or 7, characterized in that: A data flow anomaly recognition model is constructed based on the long short-term memory network algorithm to monitor the data flow after privacy protection in real time. The long short-term memory network algorithm identifies abnormal traffic by learning the time series changes of data traffic. The output of the data flow anomaly recognition model is the traffic anomaly score E. s core; According to the statistical characteristics of historical traffic data, including mean and standard deviation, an adaptive risk threshold T is set. risk , the calculation formula is: T risk =μ+k·σ Among them, k is the sensitivity coefficient, ranging from 2 to 3, μ is the average value of historical flow, and σ is the standard deviation of flow.

9. The secure transmission method for 5G communication technology according to claim 8, characterized in that: The anomaly score E for the current time period s core and adaptive risk threshold T risk For comparison, where: If E s core>T risk , it is considered that abnormal traffic is detected and an abnormal alarm is triggered; If E s core≤T risk , the traffic is considered normal and no alarm needs to be triggered.

10. The secure transmission method for 5G communication technology according to claim 9, characterized in that: After the alarm is triggered, the abnormal traffic is analyzed using the source tracing algorithm. a Identify the source of the attack, including: Perform feature analysis on the abnormal traffic to generate a feature vector V a ; Determine the feature vector V by similarity matching a The closest path P a ; For the path P a Perform blocking operations and record attack source information.

Images