Network data encryption and privacy protection system in cloud environment

Abstract

The invention relates to the technical field of cloud computing, in particular to a network data encryption and privacy protection system in a cloud environment, which comprises a key management unit driven by a wolf pack algorithm, an encryption algorithm optimization unit, a privacy protection strategy dynamic adjustment unit and a safety monitoring and abnormity response unit. The invention discloses a cloud environment network data encryption and privacy protection system constructed based on a wolf pack algorithm. High-security keys are dynamically generated and distributed through a key management unit, security performance and resource consumption are balanced through an encryption algorithm optimization unit, multi-target dynamic gaming and compliance guarantee are achieved through a privacy protection strategy unit, distributed attack detection and cooperative defense are completed through a security monitoring unit, and the security performance is improved through a cooperative feedback mechanism between the units. Intelligent encryption protection, dynamic strategy adjustment and efficient attack response of the full life cycle of the data in the cloud environment are realized, and the system security, the resource utilization rate and the compliance capability are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing, and in particular to a network data encryption and privacy protection system in a cloud environment. Background Art

[0002] A cloud environment refers to a virtualized, distributed computing environment built based on cloud computing technology. It dynamically allocates and manages computing resources (such as servers, storage, network equipment, software services, etc.) through the network to achieve on-demand use and elastic expansion of resources. With the popularization of cloud computing technology, the scale and complexity of data storage, transmission, and processing in cloud environments are growing exponentially. Data security and privacy protection have become core challenges. In existing technologies, traditional cloud environment data encryption and privacy protection schemes have problems such as low key management efficiency, insufficient adaptability of encryption algorithms, static privacy policies, and poor coordination of anomaly detection.

[0003] Based on this, the present invention provides a network data encryption and privacy protection system in a cloud environment to solve the above-mentioned technical problems. Summary of the Invention

[0004] The purpose of the present invention is to provide a network data encryption and privacy protection system in a cloud environment. The present invention is based on a cloud environment network data encryption and privacy protection system constructed based on the wolf pack algorithm. The key management unit dynamically generates and distributes high-security keys, the encryption algorithm optimization unit balances security performance and resource consumption, the privacy protection strategy unit realizes multi-objective dynamic game and compliance assurance, and the security monitoring unit completes distributed attack detection and collaborative defense. Relying on the collaborative feedback mechanism between units, it realizes intelligent encryption protection, dynamic strategy adjustment and efficient attack response for the entire life cycle of data in the cloud environment, which significantly improves the system security, resource utilization and compliance capabilities.

[0005] To achieve the above object, the present invention provides the following technical solutions: The present invention provides a network data encryption and privacy protection system in a cloud environment, including a key management unit driven by a wolf pack algorithm, an encryption algorithm optimization unit, a privacy protection strategy dynamic adjustment unit, and a security monitoring and exception response unit, wherein: The wolf pack algorithm driven key management unit is used to optimize key generation, update and distribution by utilizing the wolf pack algorithm; The encryption algorithm optimization unit is used to perform multi-objective collaborative optimization on multiple encryption algorithms and their parameters through the wolf pack algorithm to balance security, performance and resource consumption; The privacy protection strategy dynamic adjustment unit: a multi-objective dynamic game mechanism based on the wolf pack algorithm, through strategy weight distribution, cross-unit collaborative feedback and compliance adaptive adjustment; The security monitoring and anomaly response unit is used to build a distributed anomaly detection model through the wolf pack algorithm, and the agent nodes collaborate to identify attack behaviors and trigger dynamic defense.

[0006] The key management unit driven by the wolf pack algorithm includes a key generation module, a key update module, and a key distribution module, wherein: The key generation module simulates the collaborative mechanism of a wolf pack: "scouting, alpha decision-making, and fierce wolf attacking," maps the key space to the wolf pack search space, and uses the wolf pack algorithm to iteratively optimize the key sequence with high complexity and low repetition rate, with the objective function of maximizing the key entropy and minimizing the repetition rate. The key update module: based on the dynamic collaboration characteristics of the wolf pack, sets a periodic update cycle and monitors the risks of abnormal login and key leakage in real time; The key distribution module is used to detect the cloud network topology in real time through the wolf detection node, generate a weighted link graph, the head wolf node calculates multiple non-overlapping shortest paths through the wolf pack algorithm path optimization model, and the fierce wolf node splits the key and transmits it to the target node in parallel along different paths.

[0007] The encryption algorithm optimization unit includes an algorithm parameter optimization module, an algorithm combination selection module, and a dynamic feedback optimization module, wherein: The algorithm parameter optimization module is used to construct the fitness function of the wolf pack algorithm with encryption strength, calculation delay and resource occupancy as optimization targets; The algorithm combination selection module is used to dynamically match the optimal encryption algorithm combination using the wolf pack algorithm according to data type, data sensitivity and business requirements; The dynamic feedback optimization module is used to receive the attack type fed back by the security monitoring unit, and the wolf pack algorithm automatically switches the defensive parameters and adjusts the complexity of the encryption algorithm.

[0008] The encryption algorithms include symmetric encryption algorithms, asymmetric encryption algorithms and homomorphic encryption algorithms. The wolf pack algorithm is dynamically matched according to the data type, such as real-time traffic, stored data, and privacy data.

[0009] The dynamic feedback optimization module receives the attack type fed back by the security monitoring unit, and the wolf pack algorithm automatically switches the defensive parameters and adjusts the complexity of the encryption algorithm. The specific operations are as follows: A1: Attack feature extraction: Receive attack type data fed back by the security monitoring unit and extract key parameters such as attack frequency f, attack duration t, and attack intensity s; A2: Defense parameter calculation: Based on the wolf pack algorithm, a fitness function F(x) is constructed, where x is the encryption algorithm parameter vector. ,in, is the weight coefficient, and , calculation parameter adjustment priority; A3: Dynamic parameter adjustment: According to the calculation results, if the attack strength S>80, the number of rounds n of the AES encryption algorithm will be increased from the default 10 rounds to rounds, while reducing the parallelism of data encryption to reduce the risk of side channel attacks; A4: Strategy learning optimization: record the defense effect E after each parameter adjustment, and iteratively update the weight coefficient through the wolf pack algorithm , the formula is ,in, is the updated weight coefficient , is the weight coefficient of the previous iteration , ensuring that subsequent adjustment strategies are more suitable for attack scenarios.

[0010] The privacy protection policy dynamic adjustment unit includes a policy weight distribution module, a cross-unit collaborative feedback module, and a compliance adaptive adjustment module, wherein: The strategy weight allocation module is used to dynamically allocate the weights of various strategies based on the "siege prey" model of the wolf pack algorithm, using the data desensitization level, access control granularity, and compliance indicators as game factors; The cross-unit collaborative feedback module is used to establish data interaction between the key management unit driven by the wolf pack algorithm, the encryption algorithm optimization unit, and the security monitoring and abnormal response unit, and receive data on key update frequency, algorithm complexity, and threat level; The compliance adaptive adjustment module is used for a built-in compliance rule base, and the wolf pack algorithm regularly scans the matching degree between the privacy policy and the rule base.

[0011] The strategy weight allocation module is used to dynamically allocate the weights of various strategies based on the "siege prey" model of the wolf pack algorithm, using the data desensitization level, access control granularity, and compliance indicators as game factors. The specific operations are as follows: B1: Data desensitization is divided into three levels: complete desensitization, partial desensitization, and anonymization, with values ​​of 1, 0.6, and 0.3, respectively. Access control granularity is assigned based on scope, such as user level = 1, department level = 0.7, and global level = 0.4. Compliance indicators are scored based on the degree of alignment between current policies and regulations. B2: Let the size of the wolf pack be N, and the position vector of each wolf be expressed as , corresponding to the weights of data desensitization, access control, and compliance, the initial weight and the weights that need to be met ; B3: Construct the fitness function, the formula is , where D, A, and C are the quantitative values ​​of data desensitization level, access control granularity, and compliance indicators, respectively. is the empirical coefficient, such as: ; B4: The "siege prey" mechanism based on the wolf pack algorithm, through the formula Iteratively update the weights, where is the current global optimal solution, is the randomly selected wolf position, is a random number between 0 and 1 until the fitness function converges; B5: Apply the final optimized weight vector to the privacy protection strategy and dynamically adjust the priorities of data desensitization, access control, and compliance-related measures.

[0012] The compliance adaptive adjustment module is used to build a compliance rule base. The wolf pack algorithm regularly scans the match between the privacy policy and the rule base. The specific operations are as follows: C1: Compliance rule parsing: Parsing the terms in the built-in compliance rule base into a set of triples ,in: is the rule number, Rule attributes, such as "data storage period" and "user access rights"; Compliance thresholds, such as "≤30 days" or "Requires explicit user consent"; C2: Strategy-Rule Matching Calculation: Constructing Matching Function , where P is the current privacy policy parameter vector, and the calculation formula is: ;in, is the consistency function, 1 for complete match, otherwise 0; C3: Wolf pack algorithm optimization scan: Initialization: Set the wolf pack size N, the position vector of each wolf represents the scanning priority weight of m rules, and ; Fitness function: ,in is the rule importance weight; Iterative update: Through the "siege" mechanism of the wolf pack algorithm, according to the formula Adjust the weights, where is the current optimal solution, is the randomly selected wolf position, is the shrinkage factor; C4: Compliance Risk Assessment: Calculation of Comprehensive Compliance Risk Index ,when Trigger strategy modification; C5: Dynamic correction strategy: Based on the rule priority optimized by the wolf pack algorithm, adjust the privacy policy parameter P to maximize RI. The correction formula is: ,in, is the learning rate, is the gradient of the fitness function.

[0013] The security monitoring and abnormal response unit includes a distributed detection module, a collaborative identification module, and a dynamic defense module, wherein: The distributed detection module is used to adopt a layered architecture, with edge nodes acting as "wolf detection agents" to detect traffic anomalies in real time based on a local lightweight wolf pack algorithm model. The central node acts as a "lead wolf controller" to aggregate the detection results of each agent and identify attack chains using a global wolf pack algorithm model. The collaborative identification module is used to adopt a "voting mechanism" between agent nodes. When multiple wolf-detecting agents detect the same type of anomaly, the leader wolf controller determines it as a valid attack; The dynamic defense module is used to trigger dynamic defense measures once an attack is confirmed, including instructing the key management unit to urgently update the key of the affected area.

[0014] The voting mechanism includes that when ≥M wolf detection agents detect the same type of anomalies, the head wolf controller determines it as a valid attack and triggers dynamic defense measures, including: ① the key management unit driven by the wolf pack algorithm urgently updates the key of the affected area; ② the encryption algorithm optimization unit enhances the encryption level of the target data; ③ the privacy protection policy dynamic adjustment unit improves the access control granularity to the user level.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention uses a cloud environment network data encryption and privacy protection system built based on the wolf pack algorithm. The key management unit dynamically generates and distributes high-security keys, the encryption algorithm optimization unit balances security performance and resource consumption, the privacy protection strategy unit realizes multi-objective dynamic game and compliance assurance, and the security monitoring unit completes distributed attack detection and collaborative defense. Relying on the collaborative feedback mechanism between units, it realizes intelligent encryption protection, dynamic strategy adjustment and efficient attack response for the entire life cycle of data in the cloud environment, which significantly improves the system security, resource utilization and compliance capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a system diagram of the network data encryption and privacy protection system in the cloud environment of the present invention.

[0017] Figure 2 This is an architectural diagram of the network data encryption and privacy protection system in the cloud environment of the present invention.

[0018] Figure 3 This is a state diagram of the dynamic adjustment of privacy policy in the network data encryption and privacy protection system under the cloud environment of the present invention.

[0019] Description of Figure Numbers: 100. Wolf pack algorithm-driven key management unit; 101. Key generation module; 102. Key update module; 103. Key distribution module; 200. Encryption algorithm optimization unit; 201. Algorithm parameter optimization module; 202. Algorithm combination selection module; 203. Dynamic feedback optimization module; 300. Privacy protection policy dynamic adjustment unit; 301. Policy weight distribution module; 302. Cross-unit collaborative feedback module; 303. Compliance adaptive adjustment module; 400. Security monitoring and anomaly response unit; 401. Distributed detection module; 402. Collaborative identification module; 403. Dynamic defense module. DETAILED DESCRIPTION

[0020] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0021] Example: like Figure 1-Figure 3 As shown, this embodiment provides a network data encryption and privacy protection system in a cloud environment, including a key management unit 100 driven by a wolf pack algorithm, an encryption algorithm optimization unit 200, a privacy protection policy dynamic adjustment unit 300, and a security monitoring and anomaly response unit 400, wherein: the key management unit 100 driven by the wolf pack algorithm is used to optimize key generation, update and distribution by utilizing the wolf pack algorithm; the encryption algorithm optimization unit 200 is used to perform multi-objective collaborative optimization of multiple encryption algorithms and their parameters through the wolf pack algorithm to balance security, performance and resource consumption; the privacy protection policy dynamic adjustment unit 300 is based on the multi-objective dynamic game mechanism of the wolf pack algorithm, through policy weight distribution, cross-unit collaborative feedback and compliance adaptive adjustment; the security monitoring and anomaly response unit 400 is used to build a distributed anomaly detection model through the wolf pack algorithm, and the agent nodes collaboratively identify attack behaviors and trigger dynamic defense.

[0022] In this embodiment, it should be noted that: the wolf pack algorithm-driven key management unit 100 provides dynamic key support for the encryption algorithm optimization unit 200, the encryption algorithm optimization unit 200 provides algorithm strength feedback for the privacy protection policy dynamic adjustment unit 300, the privacy protection policy dynamic adjustment unit 300 defines a policy baseline for the security monitoring and exception response unit 400, and the security monitoring and exception response unit 400 feeds back the dynamic adjustment of the wolf pack algorithm-driven key management unit 100 and the encryption algorithm optimization unit 200 through exception detection.

[0023] In the present invention, the key management unit 100 driven by the wolf pack algorithm includes a key generation module 101, a key update module 102, and a key distribution module 103, wherein: the key generation module 101: by simulating the collaborative mechanism of the wolf pack "scouting wolf reconnaissance-leader wolf decision-fierce wolf attack", the key space is mapped to the wolf pack search space, with the maximization of the key entropy and the minimization of the repetition rate as the objective function, the wolf pack algorithm is used to iteratively search for optimization to generate a high-complexity, low-repetition key sequence; the key update module 102: based on the dynamic collaborative characteristics of the wolf pack, a periodic update cycle is set, and the risk of abnormal login and key leakage is monitored in real time; the key distribution module 103: is used to detect the cloud network topology in real time through the wolf scouting node, generate a weighted link graph, the alpha wolf node calculates multiple non-overlapping shortest paths through the wolf pack algorithm path optimization model, and the ferocious wolf node fragments the key and transmits it to the target node in parallel along different paths.

[0024] In this embodiment, it should be noted that: the swarm intelligence optimization of the key generation module 101 generates highly secure keys, the dynamic collaboration mechanism of the key update module 102 ensures the security of the key lifecycle, and the multi-path fragmented transmission of the key distribution module 103 achieves reliable distribution.

[0025] In addition, it should be noted that the periodic update interval ranges from T1 = 5 to 30 minutes, and the specific indicator for abnormal monitoring is three consecutive invalid key attempts triggering an emergency update. The path optimization model uses a combination of the Dijkstra algorithm and wolf pack iteration. The key sharding security design is designed such that if the number of shards is ≥ 3, all shards must be reached before the key can be reconstructed.

[0026] In the present invention, the encryption algorithm optimization unit 200 includes an algorithm parameter optimization module 201, an algorithm combination selection module 202, and a dynamic feedback optimization module 203, wherein: the algorithm parameter optimization module 201: is used to construct the fitness function of the wolf pack algorithm with encryption strength, computing delay, and resource occupancy as optimization targets; the algorithm combination selection module 202: is used to dynamically match the optimal encryption algorithm combination using the wolf pack algorithm according to the data type, data sensitivity, and business requirements; the encryption algorithm includes a symmetric encryption algorithm, an asymmetric encryption algorithm, and a homomorphic encryption algorithm, and the wolf pack algorithm dynamically matches the algorithm combination according to the data type such as real-time traffic, stored data, and privacy data. The dynamic feedback optimization module 203: is used to receive the attack type feedback from the security monitoring unit, and the wolf pack algorithm automatically switches the defensive parameters and adjusts the complexity of the encryption algorithm. The specific operations are as follows: A1: Attack feature extraction: Receive the attack type data feedback from the security monitoring unit, and extract the key parameters of attack frequency f, attack duration t, and attack intensity s; A2: Defense parameter calculation: Construct a fitness function F(x) based on the wolf pack algorithm, where x is the encryption algorithm parameter vector, through ,in, is the weight coefficient, and , calculate the parameter adjustment priority; A3: Dynamic parameter adjustment: According to the calculation results, if the attack strength S>80, the number of rounds n of the AES encryption algorithm is increased from the default 10 rounds to round, while reducing the parallelism of data encryption to reduce the risk of side channel attacks; A4: Strategy learning optimization: record the defense effect E after each parameter adjustment, and iteratively update the weight coefficient through the wolf pack algorithm , the formula is ,in, is the updated weight coefficient , is the weight coefficient of the previous iteration , ensuring that subsequent adjustment strategies are more suitable for attack scenarios.

[0027] In this embodiment, it should be noted that the algorithm parameter optimization module 201 constructs a multi-objective fitness function, the algorithm combination selection module 202 implements intelligent algorithm matching, and the dynamic feedback optimization module 203 completes attack-responsive parameter adjustment.

[0028] In addition, it should be noted that the fitness function Boundary conditions: Set resource usage thresholds (e.g., CPU utilization not exceeding 80%) and minimum encryption strength standards (e.g., AES key length ≥ 128 bits) to prevent optimization results from breaching security or performance limits. If optimization results violate these constraints, a secondary optimization is triggered, forcing the parameter vector x to meet the requirements. Real-time traffic: For real-time data such as video streaming and voice calls, lightweight symmetric encryption algorithms (e.g., ChaCha20) are preferred, combined with hash message authentication codes to ensure data integrity. The Wolf Pack algorithm dynamically adjusts the encryption block size based on network bandwidth. For example, in low-bandwidth environments, the encryption block size can be reduced from the default 128 bytes to 64 bytes to reduce transmission latency. For long-term storage of financial data and medical records, a hybrid encryption scheme combining symmetric encryption (AES-256) and asymmetric encryption (RSA-4096) is employed. The Wolf Pack algorithm optimizes key management strategies based on data update frequency. For example, for rarely updated data, the key update cycle can be extended to three months, while for frequently modified data, it can be shortened to one week. Privacy data: When processing private information such as user ID numbers and biometrics, homomorphic encryption algorithms (such as CKKS) are enabled to support data analysis in ciphertext. The wolfpack algorithm dynamically adjusts the homomorphic operation depth based on the privacy protection level. For example, in face recognition scenarios, to ensure recognition accuracy, the operation depth is set to 3 layers, while reducing computational overhead by optimizing parameters.

[0029] In the present invention, the privacy protection policy dynamic adjustment unit 300 includes a policy weight allocation module 301, a cross-unit collaborative feedback module 302, and a compliance adaptive adjustment module 303, wherein: the policy weight allocation module 301 is used to dynamically allocate the weights of each policy based on the "siege prey" model of the wolf pack algorithm, using the data desensitization level, access control granularity, and compliance index as game factors; the specific operations are as follows: B1: the data desensitization level is divided into three levels: complete desensitization, partial desensitization, and anonymization, and the values ​​are assigned to 1, 0.6, and 0.3 respectively; the access control granularity is assigned according to the range size, such as user level = 1, department level = 0.7, and global level = 0.4; the compliance index is calculated based on the matching degree between the current policy and the regulations; B2: the wolf pack size is set to N, and the position vector of each wolf is expressed as , corresponding to the weights of data desensitization, access control, and compliance, the initial weight and the weights that need to be met ; B3: Construct fitness function, the formula is , where D, A, and C are the quantitative values ​​of data desensitization level, access control granularity, and compliance indicators, respectively. is the empirical coefficient, such as: ; B4: The "siege prey" mechanism based on the wolf pack algorithm, through the formula Iteratively update the weights, where is the current global optimal solution, is the randomly selected wolf position, is a random number between 0 and 1 until the fitness function converges; B5: Apply the final optimized weight vector to the privacy protection strategy, and dynamically adjust the priority of data desensitization, access control, and compliance-related measures. Cross-unit collaborative feedback module 302: Used for the wolf pack algorithm-driven key management unit 100, encryption algorithm optimization unit 200, and security monitoring and abnormal response unit 400 to establish data interaction and receive data on key update frequency, algorithm complexity, and threat level; Compliance adaptive adjustment module 303: Used for the built-in compliance rule base, the wolf pack algorithm regularly scans the matching degree between the privacy policy and the rule base. The specific operations are: C1: Compliance rule parsing: Parse the terms in the built-in compliance rule base into a set of triples ,in: is the rule number, Rule attributes, such as "data storage period" and "user access rights"; Compliance thresholds, such as "≤30 days" and "Require user explicit consent"; C2: Policy-rule matching calculation: Constructing a matching function , where P is the current privacy policy parameter vector, and the calculation formula is: ;in, is the consistency function, which is 1 if the match is complete and 0 otherwise; C3: Wolf pack algorithm optimization scan: Initialization: Set the wolf pack size N, the position vector of each wolf represents the scanning priority weight of m rules, and ; Fitness function: ,in is the rule importance weight; iterative update: through the "siege" mechanism of the wolf pack algorithm, according to the formula Adjust the weights, where is the current optimal solution, is the randomly selected wolf position, is the shrinkage factor; C4: Compliance risk assessment: calculate the comprehensive compliance risk index ,when C5: Dynamic correction strategy: Based on the rule priority optimized by the wolf pack algorithm, the privacy policy parameter P is adjusted to maximize RI. The correction formula is: ,in, is the learning rate, is the gradient of the fitness function.

[0030] In this embodiment, it should be noted that: the strategy weight allocation module 301 implements multi-objective dynamic game decision-making, the cross-unit collaborative feedback module 302 obtains global security situation information, and the compliance adaptive adjustment module 303 ensures strategy compliance.

[0031] In addition, it is important to note that the "siege prey" model maps the strategy optimization process, such as "the alpha wolf represents the current optimal strategy, the scout wolf explores new strategy spaces, and the fierce wolf quickly executes high-priority strategy adjustments." The update mechanism of the compliance rule library, such as the automatic synchronization of official regulatory documents every week, takes an example of a compliance risk amendment: "When it is detected that the data storage period exceeds the 30 days stipulated by GDPR, a policy amendment is automatically triggered to adjust the storage period parameter to 25 days.

[0032] In the present invention, the security monitoring and anomaly response unit 400 includes a distributed detection module 401, a collaborative identification module 402, and a dynamic defense module 403. The distributed detection module 401 utilizes a layered architecture, with edge nodes acting as "wolf detection agents" to detect traffic anomalies in real time based on a local lightweight wolf pack algorithm model. The central node acts as a "lead wolf controller," aggregating the detection results of each agent and identifying attack chains using the global wolf pack algorithm model. The collaborative identification module 402 utilizes a "voting mechanism" among agent nodes. When multiple wolf detection agents detect the same anomaly, the leader wolf controller determines it to be a valid attack. The voting mechanism includes the following: when ≥M wolf detection agents detect the same anomaly, the leader wolf controller determines it to be a valid attack and triggers dynamic defense measures, including: ① the wolf pack algorithm-driven key management unit 100 urgently updates the affected area's key; ② the encryption algorithm optimization unit 200 enhances the target data encryption level; and ③ the privacy protection policy dynamic adjustment unit 300 increases access control granularity to the user level. The dynamic defense module 403 triggers dynamic defense measures upon confirming an attack, including instructing the key management unit to urgently update the affected area's key.

[0033] In this embodiment, it should be noted that: the distributed detection module 401 implements preliminary attack identification, the collaborative identification module 402 completes attack verification and confirmation, and the dynamic defense module 403 performs multi-dimensional emergency response.

[0034] Furthermore, it is important to note the distribution of computing power between edge nodes and central nodes. For example, edge nodes handle 90% of real-time traffic detection, while central nodes are responsible for complex attack chain analysis. The voting mechanism's threshold M (e.g., M = 3) and false alarm rate control methods, such as introducing a time window where voting is triggered only when more than five similar anomalies occur within three minutes, are also important. The time delay requirement for multi-unit collaboration, such as ≤10 seconds from attack detection to key update completion, is based on the defensive measures determined by the technical solution of this invention.

[0035] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0036] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A network data encryption and privacy protection system in a cloud environment, characterized by: The system comprises a key management unit (100) driven by a wolf pack algorithm, an encryption algorithm optimization unit (200), a privacy protection strategy dynamic adjustment unit (300), and a security monitoring and abnormal response unit (400), wherein: The wolf pack algorithm driven key management unit (100) is used to optimize key generation, update and distribution by utilizing the wolf pack algorithm; The encryption algorithm optimization unit (200) is used to perform multi-objective collaborative optimization on multiple encryption algorithms and their parameters through a wolf pack algorithm to balance security, performance and resource consumption; The privacy protection strategy dynamic adjustment unit (300): a multi-objective dynamic game mechanism based on the wolf pack algorithm, through strategy weight distribution, cross-unit collaborative feedback and compliance adaptive adjustment; The security monitoring and anomaly response unit (400) is used to construct a distributed anomaly detection model through a wolf pack algorithm, and the agent nodes collaborate to identify attack behaviors and trigger dynamic defense.

2. The network data encryption and privacy protection system in a cloud environment according to claim 1, characterized in that: The wolf pack algorithm driven key management unit (100) comprises a key generation module (101), a key update module (102), and a key distribution module (103), wherein: The key generation module (101) simulates the collaborative mechanism of a wolf pack, i.e., "scouting wolf - alpha wolf decision - fierce wolf attack", maps the key space into the wolf pack search space, takes maximizing the key entropy and minimizing the repetition rate as the objective function, and uses the wolf pack algorithm to iteratively search for the optimal solution, thereby generating a key sequence with high complexity and low repetition rate; The key update module (102) sets a periodic update cycle based on the dynamic collaboration characteristics of the wolf pack and monitors the risks of abnormal login and key leakage in real time; The key distribution module (103) is used to detect the cloud network topology in real time through the wolf detection node, generate a weighted link graph, the head wolf node calculates multiple non-overlapping shortest paths through the wolf pack algorithm path optimization model, and the fierce wolf node splits the key and transmits it to the target node in parallel along different paths.

3. The network data encryption and privacy protection system in a cloud environment according to claim 1, characterized in that: The encryption algorithm optimization unit (200) comprises an algorithm parameter optimization module (201), an algorithm combination selection module (202), and a dynamic feedback optimization module (203), wherein: The algorithm parameter optimization module (201) is used to construct a fitness function of the wolf pack algorithm with encryption strength, computing delay, and resource occupancy as optimization targets; The algorithm combination selection module (202) is used to dynamically match the optimal encryption algorithm combination using the wolf pack algorithm according to data type, data sensitivity and business requirements; The dynamic feedback optimization module (203) is used to receive the attack type fed back by the security monitoring unit, and the wolf pack algorithm automatically switches the defensive parameters and adjusts the complexity of the encryption algorithm.

4. The network data encryption and privacy protection system in a cloud environment according to claim 3, characterized in that: The encryption algorithms include symmetric encryption algorithms, asymmetric encryption algorithms and homomorphic encryption algorithms. The wolf pack algorithm is dynamically matched according to the data type, such as real-time traffic, stored data, and privacy data.

5. The network data encryption and privacy protection system in a cloud environment according to claim 3, characterized in that: The dynamic feedback optimization module (203) receives the attack type fed back by the security monitoring unit, and the wolf pack algorithm automatically switches the defensive parameters and adjusts the complexity of the encryption algorithm. The specific operations are as follows: A1: Attack feature extraction: Receive attack type data fed back by the security monitoring unit and extract key parameters such as attack frequency f, attack duration t, and attack intensity s; A2: Defense parameter calculation: Based on the wolf pack algorithm, a fitness function F(x) is constructed, where x is the encryption algorithm parameter vector. ,in, is the weight coefficient, and , calculation parameter adjustment priority; A3: Dynamic parameter adjustment: According to the calculation results, if the attack strength S>80, the number of rounds n of the AES encryption algorithm will be increased from the default 10 rounds to rounds, while reducing the parallelism of data encryption to reduce the risk of side channel attacks; A4: Strategy learning optimization: record the defense effect E after each parameter adjustment, and iteratively update the weight coefficient through the wolf pack algorithm , the formula is ,in, is the updated weight coefficient , is the weight coefficient of the previous iteration , ensuring that subsequent adjustment strategies are more suitable for attack scenarios.

6. The network data encryption and privacy protection system in a cloud environment according to claim 1, characterized in that: The privacy protection policy dynamic adjustment unit (300) includes a policy weight allocation module (301), a cross-unit collaborative feedback module (302), and a compliance adaptive adjustment module (303), wherein: The strategy weight allocation module (301) is used to dynamically allocate the weights of various strategies based on the "siege prey" model of the wolf pack algorithm, using the data desensitization level, access control granularity, and compliance indicators as game factors; The cross-unit collaborative feedback module (302) is used to establish data interaction between the wolf pack algorithm-driven key management unit (100), the encryption algorithm optimization unit (200), and the security monitoring and abnormal response unit (400), and receive data on key update frequency, algorithm complexity, and threat level; The compliance adaptive adjustment module (303) is used to build in a compliance rule base, and the wolf pack algorithm regularly scans the matching degree between the privacy policy and the rule base.

7. The network data encryption and privacy protection system in a cloud environment according to claim 6, characterized in that: The strategy weight allocation module (301) is used to dynamically allocate the weights of various strategies based on the "siege prey" model of the wolf pack algorithm, using the data desensitization level, access control granularity, and compliance indicators as game factors. The specific operations are as follows: B1: Data desensitization is divided into three levels: complete desensitization, partial desensitization, and anonymization, with values ​​of 1, 0.6, and 0.3, respectively. Access control granularity is assigned based on scope, such as user level = 1, department level = 0.7, and global level = 0.

4. Compliance indicators are scored based on the degree of alignment between current policies and regulations. B2: Let the size of the wolf pack be N, and the position vector of each wolf be expressed as , corresponding to the weights of data desensitization, access control, and compliance, the initial weight and the weights that need to be met ; B3: Construct the fitness function, the formula is , where D, A, and C are the quantitative values ​​of data desensitization level, access control granularity, and compliance indicators, respectively. is the empirical coefficient, such as: ; B4: The "siege prey" mechanism based on the wolf pack algorithm, through the formula Iteratively update the weights, where is the current global optimal solution, is the randomly selected wolf position, is a random number between 0 and 1 until the fitness function converges; B5: Apply the final optimized weight vector to the privacy protection strategy and dynamically adjust the priorities of data desensitization, access control, and compliance-related measures.

8. The network data encryption and privacy protection system in a cloud environment according to claim 6, characterized in that: The compliance adaptive adjustment module (303) is used to build in a compliance rule base. The wolf pack algorithm periodically scans the matching degree between the privacy policy and the rule base. The specific operation is: C1: Compliance rule parsing: Parsing the terms in the built-in compliance rule base into a set of triples ,in: is the rule number, Rule attributes, such as "data storage period" and "user access rights"; Compliance thresholds, such as "≤30 days" and "Requires explicit user consent"; C2: Strategy-Rule Matching Calculation: Constructing Matching Function , where P is the current privacy policy parameter vector, and the calculation formula is: ;in, is the consistency function, 1 for complete match, otherwise 0; C3: Wolf pack algorithm optimization scan: Initialization: Set the wolf pack size N, the position vector of each wolf represents the scanning priority weight of m rules, and ; Fitness function: ,in is the rule importance weight; Iterative update: Through the "siege" mechanism of the wolf pack algorithm, according to the formula Adjust the weights, where is the current optimal solution, is the randomly selected wolf position, is the shrinkage factor; C4: Compliance Risk Assessment: Calculation of Comprehensive Compliance Risk Index ,when Trigger strategy modification; C5: Dynamic correction strategy: Based on the rule priority optimized by the wolf pack algorithm, adjust the privacy policy parameter P to maximize RI. The correction formula is: ,in, is the learning rate, is the gradient of the fitness function.

9. The network data encryption and privacy protection system in a cloud environment according to claim 1, characterized in that: The security monitoring and abnormal response unit (400) comprises a distributed detection module (401), a collaborative identification module (402), and a dynamic defense module (403), wherein: The distributed detection module (401) is used to adopt a layered architecture, with edge nodes acting as "wolf detection agents" to detect traffic anomalies in real time based on a local lightweight wolf pack algorithm model, and central nodes acting as "head wolf controllers" to aggregate detection results of each agent and identify attack chains through a global wolf pack algorithm model; The collaborative identification module (402) is used to adopt a "voting mechanism" between the agent nodes. When multiple wolf-detecting agents detect the same type of anomaly, the head wolf controller determines it as a valid attack; The dynamic defense module (403) is used to trigger dynamic defense measures once an attack is confirmed, including instructing the key management unit to urgently update the key of the affected area.

10. The network data encryption and privacy protection system in a cloud environment according to claim 9, characterized in that: The voting mechanism includes that when ≥M wolf detection agents detect the same type of anomaly, the head wolf controller determines it as a valid attack and triggers dynamic defense measures, including: ① the key management unit (100) driven by the wolf pack algorithm urgently updates the key of the affected area; ② the encryption algorithm optimization unit (200) enhances the encryption level of the target data; ③ the privacy protection policy dynamic adjustment unit (300) improves the access control granularity to the user level.

Images