404 results about "Random seed" patented technology

The invention discloses a data encryption and dynamic key management method based on a quantum security protocol, and the method comprises the steps: 1, selecting an initial quantum random seed and a classical entropy source seed, and generating a key at an initial moment; at each moment, forming an update increment through a nonlinear chaos increment and a fractal function, constructing a key update function, and updating the key; 2, preprocessing the original message into an integer vector; constructing a dynamic transformation matrix by using a secret key at the current moment, sampling from multi-dimensional Gaussian distribution to obtain a short vector, and constructing a public random matrix; encrypting the integer vector by combining the dynamic transformation matrix, the short vector and the public random matrix to obtain a ciphertext; and step 3, decrypting by using the secret key at the current moment and the dynamic transformation matrix which are synchronously updated, and recovering the original message. The invention provides a data encryption scheme with high security and quantum attack resistance.

The invention discloses an anti-counterfeiting traceability system and method based on Beidou satellite positioning and a block chain. The method comprises the following steps: generating a current positioning data tuple; performing validity judgment on the satellite signal quality index; generating current event summary information, performing encryption calculation on the event summary information by using a private key embedded in a commodity and improving a verifiable random function, and outputting a random seed and a verification proof; constructing an event fingerprint and generating an event signature by using a private key; a challenge request is generated when the commodity request state is migrated, the commodity end device responds to the challenge request and collects new Beidou positioning information, and a challenge response data packet is constructed; and performing consistency verification on the challenge response data packet according to the virtual state resume smart contract, writing the verified data into a geographic sub-chain, recording an event abstract hash path by a main chain, and updating a commodity state transfer path. According to the invention, based on Beidou positioning and an improved verifiable random function, the credible verification of the commodity position and the traceable management of the state are realized.

The invention discloses a fine tuning system of a large-scale pre-training model in a federated learning environment and application thereof. The system comprises a local disturbance gradient estimation module, a differential privacy protection module and a global model aggregation and update module. The local disturbance gradient estimation module is used for calculating a global model loss value by combining forward propagation with a zero-order optimization method so as to estimate a gradient and realize all-parameter fine tuning; the differential privacy protection module performs differential privacy protection processing on the estimated disturbance gradient to prevent gradient information from leaking user sensitive data; and the global model aggregation and update module reconstructs a disturbance vector and completes global model update based on a random seed and a scalar gradient uploaded by a client. Compared with the prior art, on the premise of not depending on back propagation, all-parameter fine tuning of a large-scale pre-training model is achieved, data privacy is guaranteed, meanwhile, calculation and memory expenses are remarkably reduced, and the method is suitable for a resource-limited distributed calculation environment.

The invention discloses a meteorological data transmission method based on an edge protection gateway algorithm, and belongs to the technical field of data transmission, and the method comprises the steps: building communication connection with meteorological observation equipment through a serial port service controller, carrying out the standardization processing of data, and carrying out the fragmentation storage and backup; generating a random seed by using a hardware random number generator, generating a round key through a key expansion function, performing encryption operation on the grouped data, and calculating a message authentication code; sending a password suite list to establish handshake connection, exchanging a key through elliptic curve operation, segmenting compressed data into records, and adding an authentication tag for transmission; identity authentication is carried out by adopting a digital certificate, and security protection is carried out on data through a firewall state table and a feature matching algorithm. Through the scheme of the application, the security of meteorological observation data transmission can be improved, data leakage, tampering and attacking are prevented, and the security and reliability of meteorological data are guaranteed.

This document describes techniques and apparatuses directed at stateful hash-based signing with a single public key and multiple independent signers. Upon obtaining a Leighton-Micali signature (LMS) randomized parameter, a provisioning server may share the LMS randomized parameter among multiple signers. Next, the provisioning server may associate a unique, starting leaf index number to each signer and notify each of the signers. The signers may then create a random SEED for Leighton-Micali one-time signature (LM-OTS) signatures and generate local LM-OTS and LMS public keys. After generating the local public keys, the signers may share local LMS public keys with the provisioning server. Upon receipt of the local LMS public keys, the provisioning server may then order the local LMS public keys and generate a common LMS public key. The provisioning server can then provision the ordered list, the common LMS public key, and a Merkle tree path to each of the signers.

The invention relates to a master-slave communication technology, and discloses an automatic slave address allocation method of a master-slave communication system and the master-slave communication system, and the method comprises the following steps: when all slave addresses need to be reallocated, a host sends a first broadcast frame containing a first instruction identifier and a random seed; and after receiving and analyzing, each slave generates random delay and a slave address in combination with the unique identifier of the slave and the random seed, and sends a response frame containing the slave address and equipment information to the host according to the random delay. And the host receives the response frame, updates the slave equipment information table and establishes a special communication link for the slave. And when a new slave is accessed, the host sends a second broadcast frame containing a second instruction identifier, a random seed and an allocated address list. And the slave with the allocated address does not respond, the new slave generates an address sending response frame according to a rule, and the host establishes a link for the new slave. The invention further discloses a computer readable storage medium. The invention aims to improve the efficiency of allocating addresses to the slave by the master-slave communication system and facilitate the expansion and access of the new slave.

The invention discloses a medical data security access control method and system based on cloud network side-end cooperation. The method and system are used for solving the problems of privacy disclosure, data abuse, access control failure and the like in cross-institution medical data cooperation. The method provided by the invention comprises the steps that the edge device collects medical data of a user; performing preprocessing and homomorphic encryption on the medical data on the edge computing node, and uploading the encrypted medical data to a cloud platform; when a use request of a calculation participant for target medical data of a target user is received, obtaining a smart contract; checking whether the authority of the calculation participant accords with an authorization rule or not through the smart contract; if yes, activating an SMPC protocol, and generating a random seed through a VRF or a chaotic mapping function; performing encryption calculation by using a random seed to obtain an encrypted calculation result; and when the calculation participant requests to obtain the calculation result, decrypting and authorizing to access the calculation result according to the verification result of the random seed.

The invention discloses a small sample modeling stability evaluation method and system based on Bootstrap resampling, and particularly relates to the technical field of computers and data intellectualization, the method comprises the following steps: completing data preprocessing and stable stage identification under a unified time base, and forming a segment set; an average block length is used as a decision quantity, an SBB is optimized to construct a sample space, and an optimal block length is determined in combination with variance consistency and nominal coverage rate consistency criteria; carrying out re-sampling training around short / medium / long scales, and collecting layered indexes such as prediction, parameters and features; stability calculation and empirical coverage rate calibration are completed based on intra-scale statistics and inter-scale weighted mixing, and a pseudo-stationary diagnosis score is output; and finally, engineering judgment and backspacing optimization are carried out according to the coverage rate, the correlation maintenance and the risk threshold. The system records random seeds, block metadata and model version generation evidence pointers, has the characteristics of traceability and reverifiability, and is suitable for scenes of production lines, network traffic, finance 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 rapid path planning method based on collision constraint and path point optimization. The method comprises the following steps: acquiring a path planning starting point, a target point and obstacle information; obtaining a feasible sampling point set based on collision constraints; setting a time step t as 1; at the current time step, randomly sampling from the feasible sampling point set to obtain a new path point, selecting the nearest existing path point as a father node, and calculating the shortest distance between a connecting line of the two paths and an obstacle; reselection of an RRT * algorithm and optimization of a reconnection mechanism are carried out; judging whether the current path is feasible; if t is not feasible, if t does not reach the maximum number of iterations, t is set to be t + 1, and if t reaches the maximum number of iterations, an expansion search random seed is initialized, t is reset to be 1, and currently sampled path point information is reserved; and iteratively carrying out the process until a feasible path is found, and optimizing the feasible path to obtain a final path. By adopting the scheme, the path planning efficiency is improved, and the final path is smoother.

The invention provides a text watermark adding and detecting method and device, equipment and a storage medium, and can be applied to various fields such as text creation. The method comprises the steps of obtaining cue words of a to-be-generated text and a word list of a language model; for the ith token to be generated of the text to be generated, T generated tokens are obtained from the current generated token sequence, the T generated tokens are processed through a language model, and probability distribution corresponding to the ith token to be generated is obtained; selecting K generated tokens from the current generated token sequence, determining semantic feature information of the K generated tokens, and determining a random seed corresponding to the ith to-be-generated token based on the semantic feature information of the K generated tokens; and adjusting the probability in the probability distribution based on the random seed, determining a target token of the i-th token to be generated based on the adjusted probability distribution, and obtaining a text added with a watermark based on the target token. According to the invention, the robustness of text watermark addition is 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 discloses a jewelry transaction data security and privacy protection method based on a 5G fusion application. The method comprises the following steps: establishing a distributed database; generating a first transaction public key and a first transaction private key of two transaction parties through a pre-established first asymmetric encryption algorithm; carrying out random multivariate function calculation on the first transaction public key and the first transaction private key of the two transaction parties, and desensitizing the random multivariate function through an encryption algorithm and storing the random multivariate function in a distributed database; solving a random seed of the random multivariate function as a second transaction private key of the two transaction parties through a key derivation function of a second Hash algorithm; in the encryption period, the transaction data of the two transaction parties are encrypted through an encryption algorithm, and meanwhile, the encrypted data are desensitized and stored; in the decryption period, the two transaction parties obtain plaintext data through the second transaction private key; on the basis of an existing encryption algorithm, the neural network algorithm is introduced to generate the purchaser key, and meanwhile, desensitization processing is performed on the ciphertext data, so that jewelry transaction data is safer.

The invention provides a parallel multi-target random parameter optimization method and device.The parallel multi-target random parameter optimization method comprises the steps that the number of nodes of a parallel computing cluster is configured, and a random seed sequence is generated; carrying out structure loading and compiling on the to-be-optimized model, and obtaining global configuration of the to-be-optimized model; sending the global configuration to all processes, generating an initial parameter population meeting the constraint of the global configuration by each process based on the random seed, and performing serial evaluation to obtain a target function value; independently iteratively executing shuffling complex evolution by each process until a convergence condition is met; each process outputs the respective optimal parameter vector and the corresponding objective function value to a root process, and a global optimal solution is determined through the root process. By means of the method and device, the accuracy and efficiency of parameter optimization can be remarkably improved, and the problem that the model optimization effect is poor in the prior art is solved.

A system and methods are provided for encrypting and decrypting data payloads, receiving an unencrypted data payload; generating a random seed value; generating in FPGA firmware an encryption hash key from seed parameters including the seed value, XORing the encryption hash key with the unencrypted data payload to generate an encrypted data payload; transmitting the encrypted data packet with the seed value and the encrypted data payload to a second FPGA that regenerates the hash key from the see parameters and XORing the hash key with the encrypted data payload to regenerate an unencrypted data payload.

The invention discloses a secondary power modulus-based post-error-correction-free quantum lattice signature method, which comprises a counter, and comprises the following steps of: S1, sampling a random seed, and expanding the seed through a hash function to obtain a byte stream element I, a byte stream element II and a byte stream element III; s2, calculating and generating a public key and a private key based on the throttling element I, the byte stream element II and the byte stream element III; s3, assigning initial values to the counter and the signature, and calculating and generating a hash value I and a hash value II based on the private key and the signature message; s4, performing signature loop calculation based on the hash value I, the hash value II, the private key and the initial value of the counter to generate a legal signature final value, and returning the legal signature final value; s5, a signature verification variable is generated through calculation based on the public key and the signature final value, if the signature verification variable meets a verification condition, signature verification is passed, otherwise verification is not passed, the post-quantum lattice signature calculation efficiency, bandwidth optimization and stability are greatly improved, and a more efficient solution is provided for practical application.

The invention relates to the technical field of optical detection imaging, in particular to a single-pixel imaging method and system. Comprising the steps of obtaining optical signals to obtain a reference image and target optical field information, taking the reference image and the target optical field information as a mark set and a signal set, and establishing a neural optical field model and a feature extraction network; dividing corresponding data sets and random samples into a training set, a test set and a verification set according to the same random seed function; training with a neural light field model, and outputting as a model file when the minimum value of a loss function obtained through training is smaller than a set threshold value; the trained model file is verified through a verification set, if the obtained accuracy rate is larger than a first threshold value range, the result is qualified, and otherwise, iteration model training and verification are repeatedly conducted through a new random seed function, no complex optical-mechanical structure and spatial modulation mode exist, high-precision imaging of an object can be achieved through a single-pixel sensor, and high-precision imaging of the object is achieved. And meanwhile, a semi-supervised deep learning algorithm is adopted, so that the training speed is improved, and hardware resources are saved.

The invention discloses a dynamic encryption and security level adaptive adjustment method for flight data, particularly relates to the field of data security, and is used for solving the security and integrity problems of the flight data in an unstable network environment. According to the method, the mass of random seeds is quantified in real time in the mixing process, incremental seeds are added once the dispersion is reduced, and weak random source transmission is blocked; the historical track continuously provides threshold correction, so that the secret key updating rhythm and the link fluctuation are kept at the same frequency; a new secret key with sufficient safety margin is output by depending on a structured spin mixing and real-time source supplementing mechanism, and a secret key material keeps double characteristics of high discretization and high independence, so that randomness attenuation caused by packet missing, retransmission or node out-of-step is avoided; meanwhile, security level adjustment logic is deeply coupled with a key generation process, so that adaptive balance of encryption strength and operation load is realized, and the flight data is enabled to maintain collaborative excellent performance of confidentiality, integrity and real-time performance in a complex environment.

The embodiment of the invention provides a large language model optimization method based on variance reduction and momentum acceleration. The method comprises the following steps: a gradient estimation stage of a large language model: initializing a seed list and a projection list; executing multiple independent query iterations, calling a disturbance subprogram, generating a random seed for the large language model, and storing a table; in the disturbance subprogram, the determined gradient projection value is stored in a projection list, and iteration is carried out for next query; after executing multiple independent query iterations, storing a plurality of random seeds and a plurality of gradient projection values corresponding to the random seeds; in the weight updating stage of the large language model, a gradient norm subprogram is called for each layer of the large language model, and a random seed is obtained to reset a random number generator; and determining variance-reduced gradient estimation according to the gradient projection value extracted from the projection list and the reproduced disturbance vector. According to the embodiment of the invention, gradient information queried for multiple times is aggregated to generate low-noise gradient estimation, and fine tuning of a large language model is completed.

The invention provides a system-on-chip verification method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining a plurality of random excitation components and a tag associated with each random excitation component, obtaining a basic excitation component and a random seed sequence according to the recorded tags, and carrying out the verification of the basic excitation component and the random seed sequence; and a verification failure signal is reproduced by using the basic excitation component and the random seed sequence. According to the method, labels are associated with a plurality of random excitation components, random seeds used by the random excitation components are recorded by using the labels, a basic excitation component and a random seed sequence are obtained according to the recorded labels, and specific random seeds forming the random excitation components are obtained by reading the labels associated with the random excitation components. Therefore, in the reproduction process, the same random excitation component can be generated according to the label when the failure signal appears for the first time, and the verification failure signal is reproduced by using the basic excitation component and the random seed sequence, so that the failure signal is reproduced more easily, and the verification efficiency is improved.

The invention discloses a security encryption method, device, equipment and medium for a vehicle-gauge-level HSM, and relates to the technical field of vehicle network security, and the method comprises the steps: obtaining a target algorithm parameter corresponding to a target encryption task in an algorithm parameter register after receiving the target encryption task, and determining a corresponding encryption algorithm type; if the target algorithm type is a post quantum cryptography algorithm type, a true random number generation component of a vehicle gauge level HSM is called through a post quantum cryptography coprocessor to generate a true random seed, and a target encryption task is executed based on the target algorithm parameter and the true random seed; in the process of executing the target encryption task, second-order mask operation is carried out on the fast number theory transformation operation, whether error injection attacks exist or not is continuously detected in the sampling process, and operation results of the key steps are stored through a preset composite register set.

The invention discloses a security measurement system for data transmission encryption, and particularly relates to the field of data disaster tolerance, and the system comprises an identification module, an acquisition module, a seed generation module, an AST construction module, a variation module, a verification module and a packaging module. The authentication module verifies whether the equipment or the software is in an unauthorized state or not by constructing a preposed feature vector; and the acquisition module is used for acquiring the internal environment characteristics and the external environment characteristics of the equipment or the software when the judgment result of the identification module is in an unauthorized state, splicing the internal environment characteristics and the external environment characteristics into character strings, calculating a hash value for the splicing result of the character strings through a hash function, and outputting an environment entropy character string. Through AST variation based on environment entropy and double random seed drive, dynamic randomization of backup and recovery scripts is realized, supply chain attack is prevented, even if an attacker tampers a configuration file, the script security is relatively ensured by a dynamic generation and key protection mechanism depended in the recovery process, and thus the integrity of data disaster tolerance is ensured.

The invention relates to the technical field of data encryption and decryption, in particular to a solid state disk data encryption and rapid decryption co-processing system which comprises a key management module, a task scheduling distribution module, a cache optimization module, an identity verification module and a controller monitoring module. According to the method, the unique random seed is generated by combining the user identity characteristic parameters and the SSD controller state, the key attribute is dynamically adjusted by periodically disturbing the data, the key complexity and safety are improved, cross matching is carried out according to the concurrent task priority and the data safety level in the task scheduling link, and the complexity and safety of the key are improved. The method comprises the following steps: distributing special channels and buffer resources, ensuring that a key task has a higher execution priority, performing cache optimization, dynamically dividing a multi-level cache region, realizing priority scheduling and processing of a high-frequency data block according to the access frequency of the data block, fusing authentication parameters and equipment and environment information in an identity authentication link, realizing multi-identity verification, and realizing high-efficiency and high-efficiency scheduling of the high-frequency data block according to the access frequency of the high-frequency data block. And the data decryption authority is strictly controlled.

The invention discloses a fine tuning system of a large-scale pre-training model in a federated learning environment and application thereof. The system comprises a local disturbance gradient estimation module, a differential privacy protection module and a global model aggregation and update module. The local disturbance gradient estimation module is used for calculating a global model loss value by combining forward propagation with a zero-order optimization method so as to estimate a gradient and realize all-parameter fine tuning; the differential privacy protection module performs differential privacy protection processing on the estimated disturbance gradient to prevent gradient information from leaking user sensitive data; and the global model aggregation and update module reconstructs a disturbance vector and completes global model update based on a random seed and a scalar gradient uploaded by a client. Compared with the prior art, on the premise of not depending on back propagation, all-parameter fine tuning of a large-scale pre-training model is achieved, data privacy is guaranteed, meanwhile, calculation and memory expenses are remarkably reduced, and the method is suitable for a resource-limited distributed calculation environment.

The invention discloses a pore modeling method for a gas diffusion layer of a proton exchange membrane fuel cell in the field of pore modeling, which comprises the following steps of: S1, initializing the size of a fiber layer, namely initializing parameters such as the length and width of the fiber layer of the gas diffusion layer, the number and porosity of the fiber layer, the number of random seeds, the size of a grid and the volume fraction of PTFE (Polytetrafluoroethylene); according to the method, after points are randomly generated, all points meeting the conditions are screened out in circulation, it is ensured that all generated fibers are in a simulation area, the innovative modeling method is provided, visual data of gas diffusion layer two-dimensional fiber pores can be efficiently generated, and the method has the advantages of being high in practicability and high in practicability. And an accurate and visual tool is provided for research on improvement of the internal drainage performance of the battery. According to the method, an MATLAB script parametric modeling technology is combined, the number of seeds and target porosity characteristics are accurately set, a structure meeting the actual diffusion layer porosity requirement is randomly generated, and reasonable distribution of substances in the modeling process is ensured.

The invention relates to the technical field of quantum computers, in particular to a secret key generation method and system used between two terminals and electronic equipment. And the intermediate public key is obtained based on the XOR operation of the public keys of the two terminals which are interactively shared by the two parties, so that the confidentiality and the security of the negotiation public key determined by the public random seed and the intermediate public key and the private key are effectively ensured. The data transmission frequency is reduced; and the signature security is improved.

The invention relates to a physical unclonable function modeling attack resisting method based on a neural network. The method comprises the following steps: generating a multi-bit random bit stream by using a random seed generated by an LFSR module, and simultaneously acting PUF excitation on a plurality of physical unclonable functions to generate a multi-bit physical unclonable function response; the method comprises the following steps: constructing an initial feature matrix by utilizing a Decoder module, and equally dividing position decoding excitation into a plurality of binary position decoding sub-excitation with a certain length; sequentially guiding multi-bit physical unclonable function responses to storage units of corresponding addresses to form a physical unclonable function response feature matrix, and inputting the physical unclonable function response feature matrix into the neural network model to obtain one-bit confusion response; and the other bit of confusion response of the random seed is reutilized until a final confusion response sequence with a preset bit is generated, and the server compares the final confusion response sequence with an expected confusion response to complete identity authentication. By adopting the method, the hardware security can be improved.

A data processing method performed on a verification node and including obtaining task event information and a signature for the task event information from a scheduling node. The task event information includes a first computing node set participating in execution of a computing task and a random seed provided by a task submission end of the computing task. The method further includes performing verification on the signature, deriving a random number set based on the random seed in response to the verification of the signature succeeds, selecting, for the computing task, a second computing node set associated with the random number set, verifying validity of the first computing node set by using the second computing node set, to obtain a verification result, and adding a block including the verification result to a blockchain of a verification node cluster including the verification node.

The invention provides a sequence decision strategy acquisition method and system for high-neuroplasm crowds. The method comprises the following steps: establishing a full-screen drawing canvas, and loading a task set comprising a decision map type, an emotion titer ratio, a reward parameter, a random seed and a sequence control parameter; collecting user gender selection and binding a corresponding emotion picture library; generating a decision map based on a Markov decision principle under the control of the parameter set, defining nodes and connecting edges, randomly scheduling and laying emotion pictures according to a titer ratio, and assigning rewards to form an emotion and reward coupling environment; based on the environment, recording a mouse position and a mouse path in an execution exploration stage, recording key pressing time, an action direction and an award in an execution stage, and establishing step sequence association according to a timestamp to form sequence decision data; a Latin square balance control task sequence is adopted, and the process is propelled under the trigger of a user; and finally, filing the data in a binary system and a text file for subsequent analysis. According to the invention, the authenticity and repeatability of data are improved.

The invention provides a Python bytecode obfuscation method and system based on dynamic screening and random replacement, and the method comprises the steps: calculating a plurality of safety indexes and a comprehensive safety index SDI of bytecode files, and screening out the bytecode files which have high safety requirements and need to be obfuscated; a linear congruence generator is constructed, a pseudo-random number sequence is provided for obfuscation operation of byte codes, and operation code dynamic replacement obfuscation operation is performed in byte code instructions by generating random seeds, constructing random dynamic operation code mapping tables and applying the independent operation code mapping tables to different byte code files needing to be obfuscated; and performing obfuscation state detection on the obfuscated bytecode file, if the detection result is the obfuscated bytecode file, modifying bytecode analysis and instruction scheduling logic, increasing a runtime de-obfuscation process, embedding an Opcode reflection mechanism, re-compiling to generate a customized Python interpreter supporting the execution of the obfuscated bytecode, and executing the obfuscated bytecode file. According to the invention, byte codes can be confused.