303 results about "Advantage" patented technology

The invention discloses a data security and privacy management method and system based on a block chain technology, and the method comprises the following steps: S1, collecting original data in a target system, and carrying out the preprocessing of the original data; s2, constructing an access control strategy set; s3, converting into a structured strategy configuration file, and deploying a smart contract; s4, matching and judging the access request and a preset strategy by the intelligent contract; s5, judging whether the access request is legal or not; s6, performing structured processing on the access log, and writing the access log into a block chain account book; s7, intelligently selecting an encryption mechanism, and providing access return data to the access party; and S8, carrying out modeling prediction on the access mode by adopting a federal gradient lifting tree algorithm. The block chain intelligent contract and the federal gradient boosting tree algorithm are fused, differential privacy, zero-knowledge proof and homomorphic encryption technologies are combined, data intelligence and refined privacy protection are realized, and the method has the advantages of high credibility, high intelligence and high traceability.

The invention relates to a method for obtaining a hint injection attack test case for a large language model, which combines a conditional variation auto-encoder cVAE and a Markov chain, and obtains a large language model test case through data generation and context modeling, gradual exposure of malicious instructions and simulation of multiple rounds of dialogue attacks in reality. Realizing multiple rounds of dialogue attacks on the large language model, and challenging the defense capability of the large language model; the design scheme introduces a concealment technology, role play attack, state transition and other technologies, improves the complexity and concealment of attacks, has the core advantages of automation, higher concealment, wide coverage, batch testing and the like, can evaluate the security defense capability of the large language model more truly and comprehensively, and improves the security defense capability of the large language model. The defects of an existing defense mechanism are found, and research on multi-round prompt injection attacks and improvement of a security defense mechanism are promoted.

The invention relates to the technical field of cloud computing financial security, and provides a financial data encryption transmission and storage method. The method is characterized by comprising the following steps of: executing sensitivity-driven data grading fragmentation on a user terminal; dynamic elliptic curve encryption is carried out on transmission data through a two-channel encryption engine, and fully homomorphic encryption is carried out on storage data; dynamically distributing the fragments to heterogeneous cloud nodes by the multi-cloud routing based on reinforcement learning; a distributed key management matrix is constructed, key fragments are dispersed and stored in a block chain and a hardware security module, and reconstruction is activated through biological characteristics. The method has the advantages that full-link ciphertext operation is realized, and the plaintext exposure risk is eliminated; ciphertext state financial calculation is supported; single point failure is resisted; the APT attack is defended dynamically; and the quantum security evolution capability is realized. The method is suitable for mobile banks, cross-border payment and other scenes.

The invention discloses an efficient and compact key packaging, encryption and decryption method based on NTRU. The invention belongs to the technical field of lattice passwords, and particularly relates to an NTRU-based secret key packaging, encrypting and decrypting method with scalable ciphertext compression and balanced performance. According to the method, a public key compression and scaling ciphertext compression technology is provided for an NTRU password system, only one polynomial is needed in the encryption and decryption process, and the technology is also suitable for other password schemes based on NTRU. The method has the advantages of being shorter in ciphertext size, more flexible in parameter selection, capable of proving security under classical and quantum random oracle models, tighter in theoretical security protocol advantage, higher in known attack resistance, efficient in implementation, negligible in error rate and the like.

The invention provides an adaptive data encryption method based on risk driving, and belongs to the technical field of data security based on deep learning. The method comprises the following steps: firstly, constructing a multi-dimensional input feature by collecting context information such as a communication behavior sequence, a topological structure and an equipment state; secondly, designing a communication risk intelligent assessment module based on an echo state network and a multi-head attention mechanism to realize joint identification of communication behavior types and risk levels; then, constructing an encryption strategy decision module composed of a rule tree and a strategy neural network cooperatively, and obtaining an optimal disturbance level, a disguise level, an encryption algorithm, a channel type and other strategy combinations in a limited strategy space according to an identification result; and finally, executing structure disturbance, behavior camouflage and data encryption packaging according to a strategy result, and performing secure transmission through a hidden channel. The method breaks through the limitations of lack of context modeling, uncontrollable strategy acquisition, extensive encryption granularity and the like in the traditional communication encryption technology, and has remarkable technical advantages.

The invention provides a storage type XSS attack detection method and device based on a large language model, and the method comprises the steps: receiving an initial cue word, inputting the initial cue word into a cue word expander, carrying out the extension of the initial cue word, carrying out the performance evaluation to screen out a cue word meeting a preset performance index, carrying out the multi-round extension and evaluation, and obtaining a detection result. Until the optimized cue word is obtained; the method comprises the following steps: receiving a to-be-detected code, generating a program dependency graph of the to-be-detected code, decomposing the program dependency graph of the to-be-detected code into a plurality of sub-graphs with consistent program structures to obtain a plurality of sub-program dependency graphs, traversing nodes in each sub-program dependency graph, and reserving an original structure of a program; the nodes are sequentially converted into an original code format to obtain code representation fragments, the code representation fragments and the optimized cue words are input into a large language model to judge whether each code fragment is an attack vector of a storage type XSS or not, and a final speculation result is obtained. The method has the advantages of low detection cost, high accuracy and high efficiency.

The invention discloses a hybrid encryption and decryption method based on a chaotic system and an optical key, and the encryption method comprises the steps: generating a chaotic key through the initial information of a to-be-encrypted object and an initial key, and generating a chaotic sequence for the chaotic key through employing a five-dimensional Hamiltonian conservative chaotic system (FHCCS); scrambling and grouping the initial information according to the chaotic sequence, and selecting a corresponding diffraction distance as an optical key; generating a computer-generated hologram by adopting a layered angular spectrum method according to the diffraction distance; encrypting the computer-generated hologram by adopting the chaos sequence to obtain an optical password book; and encrypting a ciphertext of the optical password book by using the chaotic sequence to obtain a ciphertext of a final encrypted object. According to the method, an encryption mode of combining the optical key and the chaotic key is adopted, and the optical password book inherits the advantages of efficient information storage, high decryption speed, high parallelism and the like of optical encryption, and has the advantages of high shear resistance, high security and the like which are not possessed by a traditional password book.

The invention discloses a data security transmission method based on an ML-KEM algorithm and a PUF, and the method comprises the steps: in a registration stage, mobile equipment generates an anti-quantum key through employing a PUF technology, and a CA issues an anti-quantum digital certificate based on a Falcon algorithm and a cryptographic accumulator; in an identity authentication stage, a mobile device and an edge gateway complete bidirectional identity authentication by exchanging identity labels, anti-quantum certificates and authentication key parameters in combination with a Falcon algorithm and PUF challenge-response, and negotiate to generate a shared key by using an ML-KEM algorithm, thereby effectively defending threats such as man-in-the-middle attack, replay attack and identity counterfeiting; in a data transmission stage, two communication parties realize data encryption transmission by adopting a symmetric encryption algorithm, and data integrity and source credibility are guaranteed in combination with a Falcon digital signature; the whole scheme has the advantages of quantum security resistance, efficient authentication, dynamic key updating, lightweight encryption and the like.

The invention discloses a digital archive multi-protection-based data security transmission method, and relates to the field of data security transmission, and the method comprises the steps: generating a random seed in real time based on a quantum state, carrying out the iterative blocking of the quantum seed through combining with Logistic chaotic mapping, outputting variable-length secret key fragmentation, carrying out the symmetric encryption of archive data through employing a lightweight SM4 algorithm, and carrying out the encryption of the archive data. Obtaining encrypted data of the fragments; a behavior anomaly probability is calculated through a hidden Markov model, a risk score is calculated through a safety interval, and when a score value exceeds a safety threshold value, a grading response is triggered; key fragments are generated by adopting n nodes, at least k fragments can reconstruct signatures to monitor the node state of the block chain, and when the node anomaly rate exceeds a node threshold value, a three-layer architecture is automatically switched. The method has the advantages that an active multi-protection system is constructed through a quantum dynamic key, risk-driven hierarchical response and block chain disaster recovery switching, and quantum attack resistance and dynamic adaptation to transmission of network risks are realized.

The invention provides a dynamic adaptive encryption protocol selection mechanism. The mechanism comprises a network condition monitoring module, a data sensitivity evaluation module, an encryption protocol dynamic selection module and a performance feedback adjustment module. Indexes such as network bandwidth and delay are monitored in real time, and the network condition is evaluated; evaluating a sensitivity level of the data to be transmitted; dynamically selecting the most suitable encryption protocol from the encryption protocol library according to the network condition and the data sensitivity; encrypting and transmitting data by using the selected encryption protocol; and monitoring the transmission performance, and adjusting an encryption protocol selection strategy according to feedback. The method can automatically judge network and data conditions, dynamically select an encryption protocol, optimize network performance and resource utilization efficiency on the premise of ensuring data security, improve transmission efficiency and reduce calculation overhead, and has the advantages of automation, security, performance optimization, efficient resource utilization and the like; the method is suitable for the field of network communication with high requirements for safety and performance.

The invention provides a block chain and machine learning combined big data secure transmission method in the technical field of information security and distributed computing crossing. The method comprises the steps that S1, a transmission end carries out data fragmentation and preprocessing on to-be-transmitted big data to obtain data blocks; s2, acquiring real-time security monitoring data, inputting the real-time security monitoring data into the data transmission risk identification model to obtain a transmission risk identification result, matching an encryption rule from an encryption rule set based on the transmission risk identification result, encrypting the data block to obtain a data encryption block, and compressing the data encryption block to obtain a data compression block; s3, calculating a data fingerprint uploading block chain of each data compression block, and sending each data compression block to a receiving end; s4, performing security defense through the abnormal transmission behavior recognition model in the receiving process of the receiving end; and S5, performing integrity verification through the data fingerprints, and performing decompression, decryption and splicing to obtain big data. The method has the advantage that the security and efficiency of big data transmission are greatly improved.

The invention discloses a mobile payment encryption system based on a data random algorithm, and the system comprises a collection module which is used for collecting multi-modal data, and constructing a cross-modal entropy source pool; the random seed generation module is used for extracting feature vectors from the cross-modal entropy source pool and generating initial random seeds; the time sequence self-adaptive random chain module is used for generating a random factor sequence based on the initial random seed recursion and forming a random chain structure; the encryption module is used for jointly encoding the random factor sequence and transaction data to be encrypted, generating a dynamic encryption parameter set and encrypting the transaction data; and the chain type updating module is used for collecting network environment parameters, updating the time sequence self-adaptive random chain and generating a new dynamic encryption parameter set so as to form a chain type evolved encrypted data sequence. According to the method, the cross-modal entropy source and the time sequence random chain are utilized to realize dynamic encryption of the mobile payment data, and the method has the advantages of high entropy randomness, strong attack resistance and high transmission security.

The invention provides an attack technique clustering and attributing method for attack modes, and relates to the technical field of network security. The attack technique clustering and attributing method comprises the following steps: collecting threat intelligence original data and extracting data based on a large language model to obtain attack information; inputting the attack information into a retrieval enhancement generation framework to generate a triple, and fusing the triple with a network security knowledge graph to obtain TTPs data; clustering TTPs data based on a Gaussian mixture clustering model, and enhancing the Gaussian mixture clustering model based on the structure entropy of TTP nodes to obtain a soft clustering result; and constructing attack trees based on a soft clustering result, carrying out weight distribution based on an attention mechanism, and carrying out matching on the attack trees in the knowledge graph based on a similarity calculation method to obtain an attribution result. According to the method, a complete attack behavior attribution implementation framework is provided, the advantages of the frontier technology are combined, and efficient, accurate, reliable, extensible and high-interpretability attribution can be achieved.

The invention provides a distributed quantum security encryption method, system and device. The method comprises the following steps: monitoring the quality of a quantum key distribution link in real time, and dynamically selecting a quantum key, a post-quantum session key or a fusion key of the quantum key and the post-quantum session key as a working key; and when the node cannot be directly reached, the trusted relay node generates an end-to-end key seed by using a quantum key and a post-quantum cryptography shared key which are respectively established with the two ends, the end-to-end key seed is encrypted and distributed by a public key and then a session key is independently derived by the two communication parties, and the relay node cannot decrypt. The system adopts a distributed Mesh network architecture supporting a terminal / relay dual mode. The device integrates a quantum random number generator, a post quantum cryptography coprocessor and a mixed key engine. The method integrates the advantages of quantum physical security and post quantum cryptography, has high security and availability, and is suitable for constructing a key infrastructure security communication network resisting quantum computing attack.

The invention relates to the technical field of image processing and secret communication, in particular to a remote sensing image encryption algorithm based on a novel spatiotemporal chaotic system. Comprising the following steps: constructing a spatiotemporal chaotic system based on a coupling mapping lattice CML model, scrambling a plaintext by using a point-taking blocking Arnold periodic scrambling algorithm, eliminating a boundary generated by a blocking effect by using a bit-level arrangement algorithm, and designing a diffusion algorithm based on the spatiotemporal chaotic system and a Hash index chain. The method has the advantages that more complex behaviors of the spatiotemporal chaotic system can be described, and the spatiotemporal chaotic system not only has a larger parameter range, but also has more chaotic sequences; from the perspective of anti-decoding capability of an encryption algorithm, the dynamic characteristic of the spatio-temporal chaotic system is very suitable for image encryption; on the basis of a spatiotemporal chaotic system, the remote sensing image encryption algorithm is designed in combination with a point-taking partitioning algorithm, so that the spatial features of the remote sensing image can be processed more comprehensively, and the encryption algorithm has relatively high capability of resisting cracking.

Disclosed herein is a means that takes advantage of the immutable security provided by blockchain data structures and further implements measures that enable data in blockchain records to remain private to their respective users via the use of multiple ciphertexts that are subjected to zero-knowledge proofs and manipulated homomorphically. Ciphertexts make use of cryptographic key pair encryption where participants make use of public keys to encrypt data intended for one another alone.

The invention relates to the technical field of information security, in particular to a data governance sensitive data desensitization method, which comprises the following steps of: performing hash fragmentation on a primary key of a database, generating semantic, statistical and topological fused enhanced vectors, and obtaining risk form fingerprints through a classification model; splicing the fingerprint and the vector in the homomorphic encryption domain to calculate a risk score, and decrypting in the trusted execution environment to obtain a high-risk feature tensor when a threshold value is met; training a volumetric neural rendering network according to the feature tensor, generating a proxy record under the constraint of a matrix product state tensor chain, and writing the proxy record into a proxy data lake by combining random salt with primary key hash; after query is received, a query hypergraph is constructed to calculate incremental mutual information, leakage indexes are obtained based on persistent coherence, Groth16 zero-knowledge proof is generated in combination with risk fingerprints, a result is returned after verification is passed by deducting privacy budget, and an aggregation value of Laplacian noise disturbance is output when the budget is insufficient. The method has the advantages of high-accuracy desensitization and verifiable privacy budget control.

The invention relates to the field of time capsule interaction, and discloses a time capsule interaction method and system, and the method comprises the steps: carrying out the logic unit segmentation of a to-be-sealed content, and building an encryption baseline which continuously evolves along with a time factor; under the support of an encryption baseline, performing fragmentation processing on the evolved and encrypted content, and triggering copy reconstruction through a witness mechanism when a node failure is detected; on the basis of a content storage link, continuously executing periodic health check in a storage period, and generating a state index in combination with a node online rate, key evolution integrity and user interaction history; when the key has an unrecoverable risk in the state index, triggering a multi-role collaborative recovery mechanism, and requiring a user side to submit a standby identity certificate; and after missing key reconstruction is completed, when preset opening time arrives, decryption operation is executed based on the key set which is finally evolved and completed and the recovered fragments, and an opening response with an interaction attribute is generated. The method has the advantage of improving the trust of the user.

Large Language Model (LLM) security includes monitoring an LLM; detecting an attack on the LLM and defining an attack type of a plurality of attack types based on the monitoring, providing a notification of the attack; and causing a defense to the attack based on the attack type. Advantageously, the security can be configured to be executed between a user outside of the LLM. Further, the security can be configured to defend against multi-turn attacks.

The invention belongs to the technical field of integrated circuit hardware security, and discloses a double-dynamic key encryption circuit and method for dynamically changing dynamic keys after interpretation, and the circuit comprises an external dynamic key module, an internal dynamic key module and a circuit original logic module. And the external dynamic key module generates a dynamic key signal through the linear feedback shift register unit, and the internal dynamic key module generates a D / T selection signal according to a parity check result of the dynamic key signal to dynamically change the type of a trigger in the state machine. The state machine is divided into a group A and a group B, key updating is triggered at an entrance state through a state detection unit, and it is ensured that the key is still dynamically changed after being interpreted. Through a double-dynamic key mechanism and state machine jump confusion, probe detection and bypass attacks are effectively resisted, and the safety of the circuit is remarkably improved. And when the initial key is wrong, circuit jump is abnormal, so that IP protection is realized. The method has the advantages of high attack resistance, dynamic key change, high security and the like.

The invention relates to the technical field of privacy computing, and discloses a privacy computing protocol composition method based on parameter security conversion, comprising clients for data conversion, a plurality of MPC servers and a communication interface, S100, each client executes model training on a local private data set, generates local model update parameters, and sends the local model update parameters to the client; the method comprises the following steps: S200, a parameter encryption and conversion stage: a client calls an MPC encryption module to carry out security encryption on local model parameters, secret state conversion is carried out on the parameters by adopting a secret sharing and homomorphic encryption mechanism, and ciphertext parameter representation is generated, and S300, an encryption parameter transmission stage: the client transmits the secret state parameters to an MPC server through a communication interface, and the client transmits the encrypted parameters to the MPC server through a communication interface. The data transmission is protected by adopting an encrypted channel, S400, a security aggregation stage: the MPC server side receives encrypted state parameters from each client side, and S500, an aggregation result decryption and synchronization stage, and the device has the advantages of realizing protocol deep fusion, enhancing security and robustness and the like.

The invention discloses a cryptanalysis task breakpoint continuation type fault-tolerant scheduling method, which relates to the technical field of distributed computing scheduling, and is characterized by comprising the following steps of: acquiring task attribute information, historical scheduling data and distributed node running state data of a cryptanalysis task; performing feature recognition on the task attribute information to obtain a task feature result, and matching the task feature result with tasks and nodes in the node operation state data to determine fault-tolerant scheduling core parameters; the method has the advantages that accurate breakpoint positioning and continuous calculation state recovery are achieved in combination with the breakpoint continuous calculation model, repeated calculation after task interruption is avoided, time and resource consumption is greatly reduced, the execution continuity of a cryptographic analysis task is improved, the method is convenient to adapt to a large-scale distributed resource scene of an E-level super-calculation cluster, and the method is suitable for large-scale distributed resource scenes of the E-level super-calculation cluster. The parallel computing power advantage of the E-level super computing cluster can be fully exerted, and the execution continuity of the cryptographic analysis task is improved.

The invention discloses an open source software supply chain security intelligent monitoring and risk assessment method, relates to the technical field of software supply chain security, and solves the problem that only existence of indirect dependence can be identified generally in the prior art, but whether a vulnerability code can be called on an actual execution path cannot be accurately judged. According to the method, through dependence locking and Hash fingerprint embedding during construction, cryptographic-level accurate traceability of a component version is realized, and version ambiguity is thoroughly eliminated; in combination with deep interception of a JVM probe for method calling and class loading events during operation, a loading source and a call stack context of each class in a real execution path can be captured, bidirectional association mapping is performed on dynamic behavior data and a static dependency tree, and whether a vulnerability component is located on an executable call chain or not is accurately identified. Compared with the defect that coarse-grained warning is carried out only through version matching in a traditional tool, the method has the advantage that by means of dynamic-static data fusion, it is avoided that a large number of unavailable indirect dependency vulnerabilities are misjudged to be high in risk.

The invention relates to the field of product authentication and traceability, and discloses a block chain-based product authentication and traceability method and system, and the method comprises the steps: obtaining the authentication information of a product, carrying out the encryption processing of data through a Hash algorithm, and uploading the processed authentication information to a block chain system; based on the product identification code, calling an intelligent contract deployed in a block chain network, and performing automatic verification on the traceability information uploaded by each participation node in the product circulation link; collecting traceability information corresponding to the product identification code, and uploading the traceability information to a block chain system in sequence; based on the tracing information, dividing information bound with the product identification code into common information and sensitive information, performing encryption processing on the sensitive information according to an authority strategy, and generating a hierarchical access data representation structure; and based on the encrypted data representation structure, generating a visual traceability path diagram of the product in combination with each time node and the product circulation path. The method has the advantage of improving the traceability of the authentication information.

The invention belongs to the technical field of network security, and discloses a federated learning-based distributed network security threat prediction system, which is characterized in that adaptive iteration is driven by threat entropy through a dynamic federated learning scheduling module, emergency update is triggered when a threat feature entropy value exceeds a threshold value, and a model iteration period is greatly compressed; a hierarchical sub-model architecture is combined, so that different domain nodes can pertinently update feature layer parameters, the recognition speed of novel variation threats is remarkably improved, the problem that traditional model recognition is lagged is solved, and it is guaranteed that a system follows up threat changes in time; a privacy enhanced aggregation module is adopted, and parameter security aggregation is realized on the premise of not leaking original data through differential privacy and homomorphic encryption double-layer protection; and meanwhile, heterogeneous data is mapped to a unified feature space by virtue of a federal transfer learning engine, so that the cross-scene model adaptability is improved, the problem of difficulty in data fusion under a centralized architecture is solved, the collaborative advantage of multi-source data is fully played, and the global model precision is improved.

The invention discloses a high-capacity file encryption method and system based on block processing. According to the method, a high-capacity file is read and encrypted in blocks according to the fixed size, the situation that the high-capacity file is wholly loaded to a memory is avoided, memory occupation is reduced, and efficient encryption is achieved; two symmetric encryption algorithms of SM4 and AES are supported, and the corresponding algorithm is automatically identified and called during decryption by embedding an algorithm identifier in an encrypted file; and generating a check code for the encrypted file by using SM3 for integrity check before decryption. A file path coding mode irrelevant to a platform is adopted, a Unicode path is supported, and mainstream operating systems such as Windows and Linux can be compatible; and meanwhile, a buffer area and a multi-thread mechanism are introduced, so that file I / O operation and encryption calculation are executed in parallel, and the encryption efficiency is improved by utilizing the advantages of a multi-core processor. According to the method, the problems of high memory occupation, inconvenience in algorithm switching, poor cross-platform compatibility and low encryption efficiency of high-capacity file encryption are solved, and the method has a wide application prospect.

The invention discloses a computer network data storage encryption method and system, and relates to the technical field of storage encryption, and the method comprises the steps: blocking the forgery possibility from a data source through an environment parameter collection mechanism based on space-time constraint, converting a physical environment feature into an encryption element, and enabling an attacker to carry out the tampering through synchronously counterfeiting a multi-dimensional environment parameter; based on a dynamic feature vector generation technology driven by a chaos model, environmental disturbance is amplified in an exponential level by utilizing initial value sensitivity of a Lorenz system, so that an encryption key every time has uniqueness and unpredictability, and a time window vulnerability of a fixed key rotation mechanism is broken; through combination of a geofence strategy and a distributed storage architecture and through fragment-level geographical constraint and BFT consensus verification, a dynamic decryption permission control system is established while the tamper-proof advantage of a block chain is reserved, and the problem of permanent solidification of pollution data is avoided.

The present invention provides a method and device for enhancing confidentiality based on a combination of hash functions, which belongs to the technical field of quantum secure communication. The present invention includes: two parties in quantum key distribution obtain a shared consistent low-security key S; one party in quantum key distribution randomly selects a square modulus hash function g from the square modulus hash function family SQH. x , and share it with the other party, quantum key distribution uses the square modulus hash function g x Compress the shared low-security key S to obtain a consistent intermediate key S'=g x (S); One of the two parties in quantum key distribution randomly selects a number-theoretic transformation hash function h from the number-theoretic transformation hash function family NH c , and share it with the other party, quantum key distribution both parties use number theory to transform the hash function h c Compress the intermediate key S' to obtain the consistent final security key K = h c (S'). The confidentiality enhancement method proposed in the present invention has the advantages of low computational complexity, simple implementation, and high computational efficiency. It can also solve the problem in the prior art that the computational complexity increases rapidly with the length of the original key processing when performing confidentiality enhancement calculations based on general-purpose hash functions, and can effectively improve the speed of confidentiality enhancement calculations.

The invention discloses a data transmission optimization system based on a communication network. According to the invention, the symmetric encryption module adopts an AES algorithm, and the high-speed encryption performance of the symmetric encryption module is used for quickly encrypting a large amount of data, so that the real-time performance of data transmission is ensured. And the asymmetric encryption module carries out security encryption on the symmetric key by using an RSA algorithm, so that the security of the key in the transmission process is ensured, and the risk of key leakage is prevented. The hybrid encryption controller module ingeniously combines the advantages of symmetric encryption and asymmetric encryption, generates a key in real time through the dynamic key generation module, and optimizes encrypted data through the ciphertext processing module, thereby constructing a multi-level and all-around security protection system. According to the design, various network attacks are effectively resisted, the integrity and non-tampering property of data transmission are guaranteed, and a solid security guarantee is provided for data transmission in a communication network.

The invention discloses a deep neural network model back door defense method based on frequency domain disturbance. According to the method, frequency domain disturbance and lightweight retraining are combined, a retraining sample is constructed through frequency domain disturbance, and a back door in a suspicious model is removed through lightweight retraining. Specifically, a frequency domain variant dataset is generated by applying perturbations to the amplitude spectrum of the image and keeping the phase spectrum unchanged. In combination with a lightweight retraining process, the model is guided to ignore high-frequency-band information possibly containing a backdoor attack trigger, and finally a healing model is obtained. The repaired model can effectively resist backdoor attacks initiated by opponents, and pollution samples containing triggers can be identified as correct categories with high probability. Compared with most related works, the scheme provided by the invention has obvious advantages.