49 results about "Random Number Seed" patented technology

The invention discloses a drainage basin monitoring network layout optimization method based on mutual information active learning. The method comprises the following steps: data processing; constructing an integrated learning prediction model composed of a plurality of LSTMs, initializing each LSTM model parameter by setting different random number seeds, and iteratively updating the model parameters; constructing a covariance matrix of prediction results of the potential monitoring points by using the trained integrated learning prediction model; on the basis of an active learning algorithm of mutual information maximization, mutual information gains of all candidate sites added into the current monitoring network are calculated, candidate points enabling the mutual information gains to be maximum are selected, and information value sorting is carried out on potential monitoring points according to the sizes of the mutual information gains; and outputting a monitoring station optimization suggestion list. According to the method, uncertainty is scientifically quantified by adopting an ensemble learning method, and global information value is evaluated by adopting a mutual information maximization strategy, so that point selection decision does not depend on subjective experience any more, but strict calculation based on data and information theory, and the scientificity of decision is remarkably improved.

The application belongs to the technical field of artificial intelligence, and particularly relates to a private AI model calling method and system based on encrypted data interaction, which comprises the following steps: S1, a client generates a series of structured random numbers through a key derivation function based on a main random number seed and each dimension index, encrypts an original input vector into an initial ciphertext vector by using a partial homomorphism encryption public key and the structured random numbers, generates a conditional re-encoding key pair for nonlinear calculation in the AI model, and sends a re-encoding public key to a server. The application can eliminate the confusion ciphertext, so that the server side constructs a quasi-gradient noise based on the intermediate state of the calculation process and injects the result, resisting side channel attacks on the output ciphertext; and the client can reconstruct and remove the noise, so that the lossless calling result is obtained without sacrificing any calculation accuracy, and end-to-end privacy protection of the whole model reasoning process is realized.

The embodiment of the invention provides a vehicle cloud interaction data transmission method, device and equipment, a medium and a vehicle, and relates to the technical field of data transmission. The method comprises the following steps: acquiring a dynamic key, and a VIN code and a TCODE code of a vehicle; wherein the VIN code can uniquely identify a vehicle, and the TCODE code can uniquely identify a vehicle-mounted terminal in the vehicle; on the basis of a specified random number seed, performing supplementation adjustment on the VIN code and the TCODE code to obtain a first plaintext and a second plaintext; symmetrically encrypting the first plaintext and the second plaintext according to the dynamic key to obtain a first ciphertext corresponding to the first plaintext and a second ciphertext corresponding to the second plaintext; splicing the first ciphertext and the second ciphertext to obtain a spliced ciphertext, and performing asymmetric encryption on the spliced ciphertext to obtain a third ciphertext; and determining a target key according to the specified length and the third ciphertext, and performing encryption or decryption verification on the vehicle cloud interaction data based on the target key, so that the possibility that the key is cracked and exposed can be reduced.

The invention provides an automatic construction method and system for large-scale image editing data. Generating M initial setting editing instructions according to the context general object data set; expanding the initial setting editing instruction by utilizing the basic large language model to obtain N basic expansion editing instructions; taking a real image generated by the non-AI as an anchor point, improving a diffusion model by using image filling, and generating a source image Os and a target image Qt according to the real image and the corresponding image description; operating P times of image filling to improve the diffusion model by using different random number seeds, and screening by using an automatic mixing index to filter out low-quality generation samples; the image filling improved diffusion model uses a distillation training stable diffusion framework, and the data generation speed is increased by P times.

Embodiments of the present application relate to the technical field of communication, and disclose an access method, system, electronic device and computer readable storage medium. The access method comprises: receiving a login request of a client; generating a random number seed and sending the random number seed to the client after the login request is verified successfully; receiving an access request sent by the client; wherein the access request carries a name-value pair, the name in the name-value pair is a value obtained by the client after a serialization operation on the random number seed; and the name in the name-value pair is verified, and after the verification is passed, the value in the name-value pair is determined to be legal, so that system resources can be saved while ensuring access security.

Provided in the present application are a secure multi-party computation method and an electronic device. The method comprises: a first device and a second device negotiating a pseudo-random number seed; first, in a trusted execution environment, the first device generating a random number on the basis of the pseudo-random number seed, and generating an auxiliary parameter on the basis of the random number, a first parameter Y, and a computation task corresponding to Y; then, receiving an encrypted data fragment x1 sent by the second device, and in the trusted execution environment, determining z1 on the basis of x1, Y, the auxiliary parameter and the computation task; and subsequently, sending z1 to the second device, and the second device determining a sum of M2 and z1 as a computation result of a computation task corresponding to X and Y. The technical solution provided in the present application can reduce the communication overheads during secure multi-party computation.

The invention discloses an AIS weak signal blind separation robustness method for an unmanned ship, and the method comprises the following steps: 1) constructing a random signal matrix set, and generating an AIS data set simulating a real scene through random number seeds, signal source dynamic change and signal length adjustment; 2) designing a generator network fusing one-dimensional convolution and an LSTM time sequence module, extracting a space-time joint feature of the mixed signal and reconstructing a target signal; 3) establishing a multi-constraint optimization mechanism, and dynamically adjusting generator parameters by combining mean square error loss, interference suppression loss, statistical correlation loss and adversarial loss; 4) constructing a generative adversarial framework, and adopting a multi-layer perceptron discriminator to realize signal authenticity identification through cross entropy loss to form a parameter game mechanism; and 5) under a 5dB low signal-to-noise ratio condition, verifying the robust separation performance of the model in a complex multipath environment. According to the method, through adversarial training and multi-dimensional feature fusion, the blind separation precision of weak AIS signals under strong noise interference is remarkably improved.

The invention belongs to the technical field of nuclear reactor physical simulation and safety analysis, and discloses a nuclear reactor passive start-up simulation method based on a Monte Carlo method, which comprises the following steps: calculating a single sample based on a Monte Carlo direct simulation method, modifying a random number seed in an input card to obtain a plurality of calculation results of mutually independent samples, and calculating the single sample according to the calculation results. And finally, obtaining a probability distribution result of the weak neutron field through statistics. The method can accurately describe the physical process of the weak neutron field stage, and is suitable for a three-dimensional complex reactor.

This invention relates to the field of industrial control system technology and discloses a fault diagnosis method for pipe production lines based on digital twins. The method includes: providing a discrete physical state library, wherein each address mapping node maps to a feature parameter vector and a control feedforward compensation matrix; generating multi-dimensional projection basis vectors based on random number seeds and fixing them into a mapping projection matrix; mapping the feature parameter vectors to projection sequences and generating mapping feature codes using a positive / negative sign quantization criterion; reconstructing the index topology of the discrete physical state library based on the mapping feature codes; acquiring real-time feedback noise sensing signals and generating real-time mapping feature codes; matching and addressing address mapping nodes in the addressing memory interval and retrieving the corresponding matrix; and implementing control command phase deviation correction. This invention improves the physical tolerance determinism of discrete digital model addressing and eliminates control command phase lag caused by analytical calculation delays.

The invention discloses a power dispatching certificate generation and verification method, device, system and equipment and a storage medium, and the generation method comprises the steps: receiving a quantum random number seed generated after a certificate application is approved, taking the quantum random number seed as an input, and employing a software and hardware collaborative architecture to execute a national secret SM2 algorithm to generate an asymmetric key pair; and performing digital signature by using an asymmetric key pair and user identity information, generating double digital certificates in combination with an SM3 algorithm, and issuing the double digital certificates to a user, a program and equipment, a signature certificate private key being only known by a client, and a private key of an encryption certificate being backed up and managed by a KMC. The invention further provides a certificate verification method based on the power dispatching security partition and the CAs and the KMCs at all levels. Partition verification is carried out on the double digital certificates through public keys stored in a trust list by two communication parties. According to the invention, the security and reliability of the power dispatching digital certificate are improved.

The embodiment of the invention provides a data communication method and device, a storage medium and electronic equipment, the method is applied to a vehicle-mounted system, the vehicle-mounted system comprises a first controller and a second controller, and the method comprises the following steps: executing a forward verification method; after the first controller verifies the first signature value, a reverse verification method is executed; after the second controller verifies the second signature value, the second controller performs data communication with the first controller, and before data communication, the first controller and the second controller perform bidirectional verification of identity information based on the random number seed and the signature value; after the identity information of the first controller and the identity information of the second controller pass verification, the first controller and the second controller conduct data communication, an attacker can be prevented from counterfeiting the identity, it is ensured that data cannot be tampered or stolen in the transmission process, and therefore the safety of data communication in the vehicle-mounted system is improved.

The invention discloses an all-digital phase-locked unit and a jitter clock signal generation method, relates to the field of semiconductor integrated circuit design, and provides the all-digital phase-locked unit aiming at the problem that the jitter characteristic of a random number source generated by a true random number at present is difficult to accurately control. The all-digital phase locking unit comprises an adjustable filtering module, a jitter control module and an output module, and adopts an all-digital form. Wherein the inherent jitter in the all-digital phase-locked unit can be controlled by presetting the gain to adjust the loop bandwidth. A random number seed is injected through a jitter control module to carry out random modulation on a control word, the bit width of random modulation is controlled by a preset jitter bit width, controllable external superposition jitter is superposed on the basis of internal inherent jitter, a complex and controllable mixed entropy source is formed, and a jitter enhanced and controllable jitter clock signal is generated. Therefore, the true random number with controllable jitter characteristic and larger information entropy is obtained according to the jitter clock signal, and the quality and the stability of the true random number are improved.

The invention provides a random number generation circuit and method of a memory and the memory. The random number generation circuit comprises a random enable signal generation module, a clock module and a random number generation module which are connected in sequence, the random enable signal generation module is used for detecting the power-on duration of the power-on process state machine and the working duration of the charge pump circuit when the write-in operation state machine controls the write-in operation so as to obtain a random enable signal and output the random enable signal to the clock module; the clock module adjusts the frequency and phase of the random enable signal to obtain a random number seed and outputs the random number seed to the random number generation module; the random number generation module generates a random number sequence corresponding to the random number seed based on the random number seed. The unique random enable signal of the memory has excellent random characteristic, the true random number can be generated at lower cost, and the address of the to-be-scanned storage area in the memory is specified through the true random number to find out the storage unit needing to be repaired, so that the scanning randomness of the to-be-scanned storage area is improved.

The present disclosure provides a SQL statement execution method and device, system, electronic equipment and storage medium; it relates to the technical field of computers. The method comprises the following steps: obtaining a target SQL statement to be executed for a target business scenario, the target SQL statement being a concurrent SQL statement under the target business scenario and a SQL statement binding a variable; generating a random number seed, and adding a distinguishing condition to the target SQL statement according to the random number seed to obtain a plurality of corresponding SQL condition statements; the execution result of each SQL condition statement is the same as the execution result of the target SQL statement; and the plurality of SQL condition statements are executed respectively to realize concurrent execution of the target SQL statement. The present disclosure can solve the mutual exclusion lock contention problem caused by parallel execution of the same SQL statement by multiple sessions, and the access delay and system lag problem caused thereby.

The invention relates to the technical field of inverter pulse width modulation, and particularly provides a random pulse width modulation method and system and a storage medium, and the method comprises the steps: collecting an initial bus voltage and a three-phase current based on an ADC module; solving a difference between the initial bus voltage after filtering and the initial bus voltage before filtering to obtain a random number seed; calculating to obtain a random number; calculating a three-phase voltage based on the three-phase current, and performing sector judgment to obtain a maximum voltage value and a minimum voltage value in different sectors; calculating an overmodulation phase voltage based on the voltage maximum value and the voltage minimum value; calculating a common-mode voltage based on the initial bus voltage, the random number, the maximum voltage value and the minimum voltage value; injecting the common-mode voltage into the over-modulation phase voltage to obtain a three-phase modulation voltage, and performing normalization processing to obtain a modulation voltage; and carrying out carrier period distribution based on the modulation voltage to obtain conduction time of upper and lower bridge arms. The random number calculation is simplified, the carrier frequency does not need to be changed when the electromagnetic noise is reduced, and the consumed calculation resources are less.

The application discloses a kind of based on inter-information active learning's watershed monitoring network layout optimization method, comprising: data processing;Integrated learning prediction model is built by multiple LSTMs, each LSTM model parameter is initialized by setting different random seed, and model parameter is iteratively updated;The covariance matrix of the prediction result of potential monitoring point is constructed using the integrated learning prediction model trained;Based on the active learning algorithm of inter-information maximization, the inter-information gain of all candidate sites joining the current monitoring network is calculated, the candidate point with the maximum inter-information gain is selected, and the information value of potential monitoring point is sorted according to the size of inter-information gain;Output monitoring site optimization suggestion list.The application uses integrated learning method to scientifically quantify uncertainty, uses inter-information maximization strategy to evaluate global information value, so that site selection decision is no longer dependent on subjective experience, but based on data and rigorous calculation of information theory, significantly improve the scientific nature of decision-making.

This invention discloses a method and system for symmetric encrypted communication between a UAV and its storage compartment based on environmental entropy. The communication method includes: S1, before the UAV leaves the compartment, the storage compartment collects environmental entropy sources to form an initial data set; S2, the initial data set is hashed and truncated to output a symmetric encryption key; S3, after the storage compartment transmits the symmetric encryption key, the UAV generates an initial message and obtains the current timestamp during its exit operation; S4, the UAV processes the initial message in blocks to obtain standard data blocks; S5, a random number seed is generated using the key and timestamp as input; S6, a permutation vector is obtained based on the random number seed, the data block sequence is rearranged, an encrypted message is generated, and transmitted to the storage compartment; S7, the storage compartment performs inverse permutation restoration processing on the encrypted message and parses it to obtain the initial message. The technical solution disclosed in this invention can generate an unpredictable, truly random key, ensuring the communication security between the UAV and the storage compartment.

The application discloses a kind of security methods for improving trusted outsourcing cloud auditing based on blockchain, method includes: combining cloud auditing with blockchain, using verifiable random function, secure multi-party computation technology, propose the security method of trusted outsourcing cloud auditing based on blockchain;Allow anyone to access and verify the data stored therein, through smart contract technology to realize the automatic auditing and reward mechanism;Adopt PoS consensus mechanism, cryptography tool VRF that can produce verifiable random number, secure multi-party computation, encryption random number seed technology, propose a kind of unbiased random selection algorithm for auditor, to ensure the randomness of selecting independent auditor from blockchain.The application aims to solve the problems of security, scalability and fairness existing in the existing blockchain-based cloud auditing system, to realize the efficient blockchain consensus mechanism, to ensure the rights and interests and randomness of participants, so as to improve the overall performance and credibility of blockchain system.

The application discloses a kind of pseudo-random number generation methods based on UDS 27 diagnostic authentication service, comprising: receiving diagnostic service 27 request;Read physical variable and mixed entropy source;Check whether the last random seed exists, if yes, enter S4, if not, enter S5;S4, load history seed, and enter S6;S5, use default initial seed, and enter S6;S6, seed is combined processing;Based on the seed value after mixed processing generates new random number;New seed is saved to nonvolatile memory;Format response data;Complete 27 authentication.The pseudo-random number generation method based on UDS 27 diagnostic authentication service of the application, improve the entropy value of random number by multi-factor combination, ensure high-entropy randomness;While superimposed history seed to ensure seed randomness, ensure random sequence continuity, to avoid being illegally simulated through 27 security access authentication service, attack ability is strong.

The invention discloses a frequency hopping table generation method and system of a wireless BMS (Battery Management System) and a storage medium. The method comprises the following steps: a main node obtains a unique identifier of a battery module; performing Hash algorithm operation on the unique identifier to obtain a network ID; performing bitwise XOR operation on the network ID and a preset salt value to obtain a random number seed; and inputting the random number seed into a pseudo-random number generator to obtain a random number sequence, and generating a frequency hopping table according to the random number sequence. According to the invention, the network ID is generated through the unique identifier of the battery module, so that each battery module has the unique network ID, and the frequency hopping table is generated through the unique network ID and the preset salt value, so that each battery module has the own frequency hopping table, the situation that a plurality of communication networks transmit and receive data at the same frequency point is avoided, and the communication efficiency is improved. Therefore, the problem that the signal collision and the signal-to-noise ratio of the channel corresponding to the frequency point are reduced is solved, and the communication quality is improved.

The application provides a pseudo-random number seed generation method and device based on a blockchain, the method comprising: a transaction request node creating a high-quality pseudo-random number, hashing the pseudo-random number to obtain a random number seed, and attaching the random number seed to transaction data and signing the random number seed; the transaction request node publishing the transaction data signed by the digital signature to the blockchain, each transaction response node in the blockchain searching the random number seed to obtain a final random number seed, and verifying the validity of the final random number seed; and using the final random seed to generate a random number to participate in a block proposal process or as an input of a smart contract, and completing a complete transaction. Compared with related technologies, the pseudo-random number seed generation method and device based on the blockchain provided by the application can realize the generation of a secret, tamper-resistant, unpredictable, anti-collision and publicly verifiable random seed.

The invention discloses a pseudo-random number generation method, computing equipment and a pseudo-random number generator, and the method comprises the steps: determining an initial state according to a random number seed; multiple rounds of iteration generation are conducted through the initial state, multiple pseudo-random numbers are generated, any round of iteration generation specifically comprises the steps that a preset permutation function is used for conducting permutation processing on the generation state corresponding to the current round, a processing result is determined, the generation state of the first round is the initial state, and the generation state of the second round is the initial state; the permutation function has pseudo-randomness; and intercepting the processing result, determining the pseudo-random number corresponding to the current round, and taking the processing result as the generation state corresponding to the next round of the current round, so that the pseudo-random number stream with the side channel attack defense capability can be quickly and efficiently generated under the condition of not replacing the random number seed. The method can cope with a large amount of requirements of pseudo-random numbers meeting security in scenes such as privacy calculation and the like.

This invention discloses a fast search method for the optimal embedding position of a large model. The method includes: selecting K representative samples from the task's dataset to construct a representative set for searching the optimal embedding position, where K should not be too large; simultaneously inputting all samples from the representative set into a multimodal large model (MLLM) and a task-specific small model, and introducing a feature fusion module to fuse the features of the i-th Transformer module of the MLLM and the features of the final layer of the small model; freezing the parameters of the small model and performing k-step efficient parameter fine-tuning on all parameters of the MLLM, where k should not be too large; in each step of fine-tuning, calculating the L2 norm of the gradient vector of the parameters with respect to the loss function to obtain k gradient L2 norms, and then weighted summing the k gradient L2 norms to obtain the fitness; changing the random number seed and repeating steps 2 to 4 for P repeated experiments, where P should not be too large or too small, and 3 to 30 iterations are preferable; iterating over i within the range of 1 to T, repeating each iteration 2 to 5 times to obtain the average fitness of different embedding positions, and finally selecting i with the largest average fitness. * The optimal embedding position is defined as T, where T is the total number of MLLM layers. The fast search method for the optimal embedding position of a large model provided by this invention features high search speed and strong task scalability, making it particularly suitable for scenarios requiring rapid knowledge fusion between small models and large models.

The application relates to a chip security mechanism self-checking system, method, computer device and storage medium. The system comprises a self-checking control module, a pseudo-random number generation module, a security mechanism unit and a cyclic redundancy check module. The self-checking control module is used for sending a control signal to the security mechanism unit needing self-checking according to configuration, enabling the self-checking function of the security mechanism unit, sending a pseudo-random number seed to the pseudo-random number generation module, comparing a check code sent by the cyclic redundancy check module with a check value, and outputting a self-checking result according to a comparison result. The pseudo-random number generation module is used for generating a plurality of pseudo-random numbers according to the pseudo-random number seed. The security mechanism unit is used for starting self-checking according to the control signal, receiving the pseudo-random number excitation and generating output data. The cyclic redundancy check module is used for performing cyclic redundancy check operation on the output data of each security mechanism unit, generating a corresponding check code, and sending the current check code to the self-checking control module when the pseudo-random number is traversed. The system can improve the self-checking coverage, reduce the test time and power consumption.

The invention relates to the technical field of network and information security, in particular to a network information security software system with a dynamic encryption mechanism and an intelligent risk assessment function. Comprising a dynamic encryption module, an intelligent risk assessment module and a security policy automatic adjustment module. Wherein the dynamic encryption module is used for intelligently selecting a built-in encryption algorithm to encrypt data according to the type of network transmission data, the importance degree of the data and the security condition of a current network environment, and dynamically generating and updating an encryption key based on system time, a random number seed and user behavior characteristics; the dynamic encryption module comprises an encryption algorithm dynamic switching unit and a secret key dynamic generating and updating unit. The dynamic encryption module dynamically selects an encryption algorithm and updates a secret key according to data characteristics and a network environment, various encryption cracking attacks are effectively resisted, and confidentiality and integrity of data in the transmission and storage process are guaranteed.

The embodiment of the invention provides an abnormal network page positioning method and device, equipment and a storage medium, and the method comprises the steps: carrying out the preset region interception processing of a page screenshot to obtain a target image, and dividing the target image into a plurality of image sub-blocks according to a preset size; performing pixel brightness and darkness division on each image sub-block based on a preset brightness threshold to obtain a brightness identification sequence, generating an embedded index sequence according to the random number seed and the pixel index sequence corresponding to each image sub-block, and performing bit extraction of the corresponding brightness identification sequence on each image sub-block based on the embedded index sequence to obtain a bit sequence, combining the bit sequence corresponding to each image sub-block in sequence to obtain dark watermark information; and decoding the dark watermark information to obtain page attribute information, and positioning the corresponding page attribute information of a plurality of published network pages to obtain an abnormal network page. According to the scheme, the pre-embedded dark watermark information can be accurately extracted, and the abnormal network page can be accurately positioned.

A first processing sends a validation request is sent, which includes a random number seed and a nonce to a second processing device. The first processing device generates a first set of random numbers using the random number seed according to an algorithm for random number generation, and maps the generated first set of random numbers to memory blocks that include a copy of software for the second processing device. A first hash is calculated from contents of the mapped memory blocks concatenated to the nonce. A second hash, calculated using information included in the validation request, the mapping model, the software for the second processing device, and the nonce is received. An action is automatically performed when the hashes do not match.

The invention discloses a tampering perception traceability watermark marking method and device for a generative image, and the method comprises the steps: firstly building a copyright watermark and a positioning watermark in a watermark embedding process, employing a double-mark combined sampling strategy, carrying out the coding of the copyright watermark and the positioning watermark, and generating an image with double marks; in the watermark extraction process, watermark latent representation is reconstructed through inverse sampling, and a reconstructed copyright watermark and a reconstructed positioning watermark are obtained; generating an original positioning watermark by using the reconstructed positioning watermark and the shared random number seed, and performing tampering positioning by detecting a high-frequency change region; and the tampered region is selectively eliminated by using the tampered information, so that the decoding accuracy of the copyright information is improved. According to the method and the device, the tampering positioning strategy is designed according to the sensitivity of pixel modification, and the tampering positioning result is used as confidence guidance of watermark decoding, so that the accuracy of watermark extraction under the tampering distortion condition is improved.

The present invention relates to the technical field of engineering equipment, specifically disclosing an equipment-integrated bus data acquisition system for secure data storage. The system addresses the problem of improving existing LCGs to obtain higher-quality seeds and improve the security and reliability of data storage. The system comprises a data layer and an acquisition layer. The data layer includes a chassis sensor module for acquiring real-time equipment chassis data and an equipment mount sensor module for acquiring equipment mount data. The acquired data is transmitted to the acquisition layer via a CAN bus interface. The acquisition layer is a data recorder comprising a data acquisition module and a data storage module. The system also improves the LCG algorithm by increasing an entropy source. The system combines a linear congruential generator with an environmental thermal noise entropy source generator to enhance the security and reliability of data acquisition and storage. High-quality random number seeds and the AES encryption algorithm are used to ensure data confidentiality and integrity, optimizing the performance of engineering equipment and operational efficiency measurement.

The embodiment of the invention provides a verification code verification method and device, equipment, a storage medium and a program product. The method comprises the following steps: receiving a verification sequence and a random number seed sent by a verification center, and generating a vibration button corresponding to each verification number on a front-end page of a target terminal; in the process that the user manipulates the at least two vibration buttons one by one, the target terminal vibrates according to the vibration mode corresponding to the manipulated vibration buttons; generating a control input sequence of the user based on the control of the user on the at least two vibration buttons; and disrupting the control input sequence based on the random number seed to obtain a disrupted to-be-verified sequence, and sending the to-be-verified sequence to a verification center, so that the verification center determines a verification result based on the random number seed, the to-be-verified sequence and the verification sequence. According to the verification scheme of the verification code, the convenience of inputting the verification code by the user can be improved, and meanwhile, the security of the verification code is improved.