77results about How to "Reduce storage overhead" patented technology

The invention provides a multi-process-mode intelligent switching method and system for vehicle-mounted sludge treatment, and the method comprises the steps: carrying out the time-space alignment and standardization of multi-dimensional sensing data such as sludge characteristics, environmental parameters and equipment states, carrying out the real-time reasoning of a process switching critical value through combining with a lightweight incremental regression model, generating a self-adaptive threshold value band with a confidence interval, and carrying out the real-time reasoning of the process switching critical value. Dynamic monitoring and mode switching pre-judgment of actual operation parameters are realized; in the switching process, a weighted fusion progressive transition control strategy is adopted, the transfer rate is adaptively adjusted according to real-time feedback, the operation stability is guaranteed, and the intelligence, reliability and switching smoothness of process switching of the vehicle-mounted sludge treatment system are remarkably improved.

The embodiment of the invention discloses a harmonic compensation method for an angle error of a magnetic encoder, and the method comprises the steps: enabling the magnetic encoder to rotate at a constant speed in a setting mode, collecting sine and cosine signals outputted by the magnetic encoder, and carrying out the calculation through CORDIC, thereby obtaining an initial angle sequence containing a fundamental wave error and a harmonic wave error; extracting each harmonic component through an adaptive wave trap, constructing a harmonic compensation polynomial, and storing each coefficient; in an application mode, a magnetic encoder signal is collected in real time, a current initial angle value is solved, a harmonic compensation amount is calculated according to a current fundamental wave angular frequency and a stored coefficient, and an accurate angle value is obtained by subtracting the compensation amount from the initial angle value. According to the method, external reference signals and a large amount of storage space are not needed, each harmonic error is directly eliminated through a harmonic compensation polynomial, and approximate errors caused by traditional linear interpolation are avoided; and the compensation precision can be improved only by increasing polynomial terms, and the storage requirement is not obviously increased.

This application relates to the field of monitoring data processing technology, and particularly to a method, device, and storage medium for preventing monitoring data from being tampered with. The method includes: during the recording of monitoring data, simultaneously acquiring a thermal noise segment from an imaging sensor for each video frame, reducing equipment costs and deployment difficulty; calculating the image activity metric of the video frame and the noise energy metric of the thermal noise segment, requiring minimal computation and meeting the simple real-time needs of ordinary monitoring equipment; performing an arithmetic combination of the image activity metric and the noise energy metric to obtain a frame verification feature value, which is then associated with and stored with the video frame, making it difficult to forge and having low storage overhead; acquiring the video frame to be verified and generating a verification feature value based on the video frame; determining that the monitoring data has been tampered with when the deviation between the verification feature value and the stored verification feature value exceeds a preset range, thus achieving low-cost, low-complexity, and high-real-time monitoring data integrity protection.

This invention discloses a method, apparatus, medium, and product for calculating bonding interactions in molecules. The method includes: allocating bonding tasks using a three-dimensional interconnected computing architecture, and constructing a bonding atom set and a bonding task index table matching the allocated bonding tasks within each computing module. This ensures that identical atoms do not need to be repeatedly stored when the computing module executes all bonding tasks, reducing the storage overhead of atomic information; dynamically acquiring and filling atomic information from the local machine or other computing modules into the corresponding storage locations of the bonding atom set, extracting atomic information according to the index table, calculating bonding components in parallel, and accumulating the bonding components of identical atoms to obtain the resultant bonding force; and updating the resultant bonding force and atomic information through inter-module interaction, and re-storing the atomic information according to the updated spatial arrangement. This reduces the amount of data transmission between modules while ensuring the accuracy of molecular bonding interaction simulation, thereby improving the stability and efficiency of the overall molecular dynamics simulation.

This disclosure provides a data processing method, apparatus, and computer device based on butterfly networks, relating to the fields of artificial intelligence and hardware acceleration technology. The solution is as follows: In a switchable data processing architecture including an inverse butterfly network, a preprocessing module, a post-processing module, and a control signal generation module, when switching to aggregation operation, the preprocessing module inputs the first input data into the inverse butterfly network, the control signal generation module outputs the first control signal to complete data processing, and the post-processing module directly outputs the aggregation result. When switching to distribution operation, the preprocessing module performs input position mapping and filling on the second input data to obtain the third input data, the control signal generation module outputs the adapted second control signal to complete data processing, and the post-processing module performs rearrangement on the intermediate results to output the distribution result. This reduces hardware resource consumption, lowers control complexity and processing latency, and improves the hardware deployment adaptability of neural network sparse computing while ensuring functional integrity.

The invention relates to a trusted security traffic monitoring system based on an anti-quantum key management framework, and relates to the technical field of smart city traffic, and the system comprises an initialization module which is used for carrying out the initialization processing of a cloud platform, an edge gateway and a camera node; the data acquisition and transmission module is used for acquiring and encrypting monitoring data based on the initialized camera node, and transmitting the encrypted monitoring data to the cloud platform through the edge gateway in a preset data transmission mode; wherein the preset data transmission mode comprises a cross-gateway area transmission mode and a same-gateway area transmission mode; and the data detection module is used for deleting or replacing the scene of the corresponding node when the edge gateway or the camera node shows a malicious behavior or the equipment is aged. According to the method, a key system is constructed through an encryption and decryption algorithm, so that the key management framework oriented to edge computing has quantum aggression resistance, and the security of a city monitoring system in a traditional scheme is improved.

The embodiment of the invention provides a lightweight lattice-based ring signature method suitable for a satellite network. The method is applied to the technical field of information security and comprises the steps that a ground control center generates system public parameters; the user generates a complete signature key pair based on the NTRU grid, and a compression verification identifier for the identity identifier is obtained through calculation of a complete signature public key; the ground control center constructs a gridding user set by using a space-time fragmentation technology, calculates Merkle tree roots as ring fingerprints and distributes the ring fingerprints to satellites; the user generates an anti-replay ring signature and member proof in combination with ephemeris information and spatio-temporal context; the satellite node executes lightweight rapid pre-verification according to the time, the position and the ring fingerprint; and the ground control center executes complete ring signature and membership verification. The method effectively solves the contradiction between limited satellite network resources and high-security requirements, and has the capability of resisting quantum computing attack.

The application discloses a student programming answer prediction method combining semantic understanding and structured modeling, collects student programming homework data and pre-processes the data to obtain data samples, translates the data samples to obtain an English feature set; uses low-rank adaptation to fine-tune a pre-trained semantic understanding model, constructs input features, and predicts an answer correctness probability; performs graph neural network structured modeling based on the association relationship between problems and concepts, performs message propagation and feature updating to obtain an embedding set, inputs the embedding set and student embedding into a prediction layer to obtain a probability of answering a question correctly; and inputs the probability of answering a question correctly and the probability of answering a question correctly after fusion into a multilayer perceptron for nonlinear mapping to predict the probability of answering a question correctly. The application solves the problems of difficulty in effectively processing non-standardized codes submitted by students, insufficient semantic understanding of question texts and low prediction accuracy in programming knowledge tracking, and provides a scientific basis for personalized teaching and learning resource scheduling.

The application relates to the field of artificial intelligence and power equipment operation and maintenance data processing technology, and particularly discloses a power energy storage operation and maintenance health evaluation method based on deep learning, which comprises the following steps: a pre-processing unit first performs caliber unification according to a historical caliber change set and extracts single data in real time, eliminates equipment differences, makes the algorithm identify only one set of caliber, and reduces special cases and error sources; an edge gateway performs clock consistency test, aligns the time axis, and avoids wrong time misjudgment; if the test fails, the data is improved and data reliability is output, and the caliber change set is rewritten and corrected; then, the data and the reliability are sent to a model to perform health evaluation and physical consistency verification, and the correlation is improved to be explainable physical reliability; finally, the evaluation result is mapped to a standardized operation and maintenance disposal suggestion, and the operation and maintenance plan is synchronized to a platform; the standardized action and plan are synchronized to realize a closed loop of diagnosis, disposal and recovery, and the whole process is auditable and reproducible.

The invention relates to a one-dimensional function fitting method and system based on input dimension expansion, and the method comprises the steps: firstly obtaining a target function and a training sample thereof, carrying out the dimension expansion of each input scalar according to an odd-dimension central symmetry rule: placing the scalar at the center of an expansion vector, and filling the two sides with the same constant related to a domain of definition; and constructing a neural network taking the extended vector as input, and adjusting the width of the hidden layer to enable the deviation between the total parameter number and the reference model not to exceed a set value, thereby realizing parameter redistribution. Then, a second-order optimization algorithm is used to train the network. During prediction, new input is input into the trained model after dimension expansion according to the same rule, and a predicted value is obtained. According to the method, parameter arrangement symmetry is actively broken, and the loss function landscape is remarkably improved.

The application provides a road surface state recognition method and device, and solves the problems of single road surface state recognition category, coarse result granularity and inability to distinguish specific road surface abnormal types caused by dependence on simple threshold judgment and shallow features in the prior art. The method comprises the following steps: acquiring a vertical vibration acceleration data segment of a vehicle in operation; extracting a characteristic value according to a preset optimal feature subset, wherein the subset comprises a frequency domain peak value and a spectrum energy; inputting the characteristic value into a classification model to obtain a recognition result, wherein the model is obtained by training based on data comprising at least three road surface state labels. Through fixed feature optimization, the application realizes fine recognition of multiple states such as well frame difference, expansion joint and deceleration strip, and significantly improves the calculation efficiency while ensuring the accuracy.

This invention provides a multi-task learning resource optimization method based on the lottery hypothesis, belonging to the field of multi-task learning and neural network optimization technology. This method employs a sparse expert hybrid approach based on the lottery hypothesis to effectively prune neural network model parameters, thereby significantly reducing the computational cost and storage overhead of the neural network model. The sparse expert network structure retains high sensitivity to key task information. The introduction of a softmax-based router and mask matrix allows each task to select and activate suitable expert network components based on the features of its input data. Cross-training and iterative amplitude pruning optimize the omnipotent expert matrix, further enhancing the model's generalization ability in handling complex tasks and different data distributions. Task-specific layers independently process and optimize the features of each task, reducing mutual interference between tasks and improving the accuracy of feature extraction and the efficiency of task processing.

The application provides a broadcast multi-board parallel test method, system, medium and product, relates to the electronic manufacturing board test technical field, and the method comprises the following steps: a broadcast test instruction makes a plurality of to-be-tested boards start executing target test items at the same time. An initial token frame address list to be collected establishes an ordered mechanism for board access, and a first data segment provides a unified data collection space. In a token frame transmission operation, the first to-be-tested board executes differentiated token frame processing operations according to its own test state: in an unfinished state, the token frame is directly forwarded; in a completed state, test completion result data is written into the first data segment and the address list to be collected is updated. This makes the boards that have completed the test automatically exit the loop, and the boards that have not completed the test continue to participate in token transmission until the address list to be collected is empty, and the test results of all the boards are collected into the first token frame. In this way, the technical problem of low time utilization efficiency in the multi-board parallel test process in the prior art is solved.

The application discloses a kind of zero-knowledge identity recognition method and system based on biological characteristics, by the biological characteristic data of the example collected and the biological characteristic data of user are preprocessed, and using elliptic curve and bilinear mapping method, according to the biological characteristic corresponding to example, construct zero-knowledge proof system, and then using zero-knowledge proof system, according to the biological characteristic corresponding to user, construct common data set CRS, and according to common data set CRS, calculate proof data and verification data, obtain proof data set and verification data set, and compare proof data set and verification data set, obtain identity recognition result;The application utilizes elliptic curve and bilinear mapping method, and constructs zero-knowledge identity recognition system based on biological characteristics, reduces the storage overhead of biological characteristic data, while ensuring the uniqueness of zero-knowledge proof input, solves the technical problems of biological characteristic data leakage, data abuse, information forgery and other risks.

Apparatus, methods, and systems for lidar data processing, as well as lidar and vehicles are provided. The apparatus includes a storage circuit, a prefetch circuit, and a first processing circuit. The storage circuit is configured to store first calibration data of the lidar. The prefetch circuit is configured to: determine first index data; indexing a first subset of the first calibration data based on the first index data; the first subset is transmitted to a first processing circuit. The first processing circuit is configured to perform a first data calibration on the echo data based on the first subset. The device for laser radar data processing can reduce the overhead of storage resources of calibration data by the laser radar, and reduces the circuit cost of the laser radar while meeting the calibration requirements of the laser radar.

The invention relates to a mail system single copy storage system and method based on object storage. The system comprises a mail index service module, an object storage service module, a single copy storage management service module and a mail client. When a mail is sent, the hash value of a mail body file is calculated, only a unique copy is stored in the object storage, and the reference count of the unique copy is managed through the metadata table; when the mail is read, the file in the object storage is directly accessed through the mail index; when the mail is deleted, the reference count is decreased progressively, and the file is physically deleted only when the count is zero. According to the method and the system, single copy storage of the mail attachments is realized, the storage space occupation in a mass mail sending scene is remarkably reduced, and meanwhile, the consistency and the security of data are ensured through a reference counting mechanism.

This application provides a neural network operation method, apparatus, and storage medium, including: adding a random mask layer to a first neural network according to the type of variable dimensions to obtain a second neural network; the random mask layer is used to perform random masking processing on tensors input to the random mask layer; and training the second neural network using sample data. The neural network operation method, apparatus, and storage medium provided in this application, by adding a random mask layer to the neural network, only require training and deployment of one neural network to adapt to different input and output dimension requirements. This method has low training complexity, low storage overhead, and is easy to deploy and continuously evolve through online training.

This application provides a method and related equipment for detecting small target vehicles from a low-altitude perspective, belonging to the fields of computer vision and intelligent transportation technology. The method includes: acquiring low-altitude images and performing multi-level feature extraction, performing feature preservation operations during the extraction process to retain high-resolution detail information; adaptively enhancing the extracted features to improve target-related feature responses; constructing a high-resolution feature response path, fusing shallow high-resolution features with deep semantic features to form enhanced features; and performing target detection based on the enhanced features. This application also optimizes feature representation through feature compression and path pruning to reduce computational complexity. Through a collaborative design of feature information preservation and reconstruction during multi-scale propagation, this application achieves a simultaneous improvement in the performance and computational efficiency of small target vehicle detection in complex low-altitude scenarios.

The invention relates to an SPAD ranging method and system based on dynamic histogram storage and multi-frame fusion, and the method comprises the steps: splitting a single-frame measurement period into a plurality of subframes, and configuring corresponding phase delay amounts, so that each subframe corresponds to a different equivalent sampling time reference; for each sub-frame, in each exposure period, screening photon trigger events of a plurality of SPAD in a macro pixel, and outputting effective photon response time, time calibration information and trigger intensity; dynamically tracking the return moment of the echo photons based on the output in the current exposure period, and generating a processing result; integrating the processing results of all the exposure periods in the current sub-frame to generate a candidate histogram; and reconstructing a fusion histogram through data fusion based on the candidate histograms corresponding to all the subframes and the respective phase delay amounts, analyzing the fusion histogram by using a flight time extraction algorithm, and outputting a distance measurement result. According to the method, high-resolution three-dimensional imaging can be realized under the condition of limited hardware resources.

The application discloses a multi-agent dynamic real-time path planning method and device based on signal timing logic, and relates to the technical field of control methods. The method comprises the following steps: acquiring multi-robot path planning task information and a preset space-time constraint task demand; constructing a multi-robot task specification based on signal timing logic, and defining a corresponding robustness evaluation function; constructing a comprehensive path cost function based on the robustness evaluation function; introducing the path cost function into a path expansion and reconnection process based on a real-time rapid expansion random tree algorithm, updating a search tree structure, and generating a candidate path meeting environment constraints and timing constraints; planning paths for robots in sequence according to a preset multi-robot cooperative planning strategy, taking the path information of the robots that have been planned as dynamic obstacles and performing progressive cooperative obstacle avoidance, and outputting a conflict-free trajectory adaptive to a dynamic environment. The application can solve the path planning problem in a multi-agent scene.

This invention proposes a brain-inspired navigation method based on DSI decoupling representation and composite potential field path optimization. The method includes: acquiring a state sequence and calculating a successor representation matrix; obtaining a DSI representation from the successor representation matrix; generating a navigation path sequence based on the DSI representation; constructing a total potential function from the navigation path sequence; generating an optimal path sequence based on the total potential function and simultaneously introducing obstacle information; using the optimal state sequence extracted from the optimal path sequence as a supervision signal to update the DSI representation using gradient descent, and storing the obtained corrected DSI representation in a calibration experience buffer; repeatedly executing online navigation, path optimization, and calibration steps using the corrected DSI representation in the calibration experience buffer, and obtaining the optimal brain-inspired navigation sequence after a preset number of iterations. This invention applies composite potential energy optimization and obstacle constraints to the initial path, resulting in a geometrically more compact and safer path.

This invention discloses a distributed, block-based, multi-dimensional storage method and apparatus for seismic data volumes. The method includes: separating the header information from the seismic traces in the acquired seismic volume file and storing the header information in a preset storage space; determining multiple dimensions based on the keywords of the seismic traces to construct a multi-dimensional model; dividing the multi-dimensional model into multiple multi-dimensional data cubes and determining the name of each multi-dimensional data cube, as well as the extreme values ​​of operational values, the total number of location numbers, and the total number of multi-dimensional data cubes in each dimension; storing the multi-dimensional data cubes on a distributed storage system using a mapping algorithm; calculating the physical storage location of the target data based on the acquired target data information, and extracting the target data from the distributed storage system and / or the preset storage space. This method improves data storage flexibility and query efficiency by optimizing the seismic data storage structure and index calculation.

The application relates to a non-interactive and time-based MAC address changing method applied to a wireless communication system, wherein the wireless communication system comprises network equipment and terminal equipment; the method is based on a two-layer MAC address mechanism, an OTA MAC address and a permanent MAC address; a synchronous random change mechanism of the OTA MAC address is provided; an OTA MAC address synchronous updating mode based on current system time is introduced; the terminal equipment can update the OTA MAC address at any time within a time period selected by the terminal equipment; the network equipment associated with the terminal equipment can calculate the corresponding OTA MAC address of the terminal equipment in each time period in advance; under the premise of ensuring that the privacy of a user in a WLAN is not leaked, the method has low calculation cost, air interface communication cost and storage cost.

The invention discloses a point cloud data processing method based on a point cloud Transform network, and the method comprises the steps: carrying out the local sampling and normalization processing of point cloud data, and generating an enhanced point cloud feature sequence; then, the data are input into a point cloud Transform network into which an adapter module is embedded; gating vectors generated by geometric conditions are introduced into an adapter module, and cross-layer collaborative updating is executed among different Transform coding layers by using cross-layer memory vectors, so that deep modulation of features and continuous transmission of interlayer information are realized; and finally, decoding the final point feature sequence through a downstream task head to output a classification or segmentation result. According to the method, the fixed trunk parameter set does not participate in updating in the training stage, only the adapter and task head parameters are updated, the training cost is reduced, and meanwhile the precision of the point cloud task is effectively improved.

This invention discloses a 3D target detection method based on sparse dynamic attention and star-shaped interaction. The method obtains basic voxel features from the original LiDAR point cloud through voxelization and sparse convolution, then introduces a sparse dynamic parallel attention module. This module achieves efficient enhancement of global context and channel dimensions through dynamic attention branches and parallel channel interaction branches. A sparse star-shaped interaction module is then used to construct a star-shaped neighborhood interaction structure with a central voxel, completing local geometric modeling and nonlinear feature interaction only on non-empty voxels. Finally, keypoint sampling, RoI pooling, and a detection head output the 3D detection box, category, and confidence score. This invention, through the synergistic complementarity of SDPA and SSB, significantly improves the detection accuracy of long-distance, small-scale, and occluded targets while maintaining linear growth in computational complexity and meeting real-time requirements. It achieves balanced performance optimization across multiple categories, including vehicles, pedestrians, and cyclists, and is suitable for 3D perception scenarios with high precision and real-time requirements, such as autonomous driving.

This invention proposes a hardware implementation method, apparatus, electronic device, and computer-readable storage medium for the Sigmoid function, relating to the field of data processing technology. The method constructs a function mapping table for the Sigmoid function, containing discrete non-negative floating-point values ​​within the unsaturated interval, based on the symmetry and saturation of the Sigmoid function, thereby substantially reducing the scope of the lookup table. On this basis, the absolute value of the floating-point values ​​belonging to the unsaturated interval is taken to obtain the value to be processed. This value is then converted to the target fixed-point format, and integer operations are used for table lookup, interpolation, and symmetry restoration to obtain the output value of the Sigmoid function based on the target floating-point value, improving computational efficiency. While ensuring the function output capability across the entire range, the size and storage overhead of the function mapping table are effectively reduced, while linear interpolation under the fixed-point format improves output accuracy.

The application provides a kind of robust laser SLAM repositioning and mapping method for embodied intelligent robot in the technical field of robot autonomous navigation, comprising: receiving original laser radar point cloud and IMU data, pre-processing point cloud to extract feature points;IMU pre-integration provides initial pose constraint;Loop closure detection is through the two-stage cascade of ScanContext global search and LinK3D local matching, obtains repositioning constraint;Front-end odometry matches feature points with local voxel map, combines IMU optimization to output pose estimation, filters key frame to generate laser odometry factor;With repositioning constraint as initial value, use improved ICP to accurately register to obtain repositioning factor;Finally, construct factor graph, jointly optimize IMU, laser odometry and repositioning factor, solve global consistent pose sequence and map.The application has the advantages that: it greatly improves the system robustness, repositioning accuracy and global map consistency of quadruped robot in severe motion and large-scale scene, while meeting the real-time constraint of embedded platform.

This invention discloses a method and system for detecting high round-trip time (RTT) flows in high-speed networks, belonging to the field of traffic detection technology. The method includes: acquiring network data flows and generating RTT samples of the network data flows; judging the validity of the RTT samples and filtering out valid samples; statistically analyzing the valid samples within a statistical period using limited storage resources and obtaining statistical results; and based on the statistical results, identifying and outputting data flows corresponding to records whose cumulative RTT values ​​exceed a preset abnormal threshold as abnormal high RTT data flows. This invention employs a two-stage processing approach. The first stage generates reliable RTT samples and removes abnormally large RTTs caused by retransmissions, while the second stage continuously tracks abnormal RTT flows based on the cleaned RTT data. This achieves efficient and accurate detection of abnormal RTT flows while meeting the processing capabilities of high-speed networks.

This application relates to a method for generating character consistency images based on multi-level LoRA training, comprising: performing low-rank adaptation training on multiple preset low-level generation models based on facial feature images and their corresponding structured facial feature semantic labels, generating facial feature LoRA weights that match each low-level generation model; loading the corresponding facial feature LoRA weights through a routing mechanism to generate a reference face image of the target character and extracting the target character's identity identifier; obtaining a multi-view extended map generated based on the reference face image, performing data augmentation on the reference face image and the multi-view extended map to construct a character training set; using the character training set to perform low-rank adaptation training on the low-level generation models to obtain character LoRA weights; and generating a consistency image of the target character based on the character LoRA weights. This method addresses the problems of character identity being easily affected by the inference environment and multi-view identity degradation in existing technologies, improving the stability of cross-view character consistency image generation.

The present application relates to a kind of contract anonymous certificate generation and verification method and system based on event triggering.The method includes: certificate issuer generates system public parameter and issuer key pair, and defines double-layer authorization strategy;Each oracle generates signature key pair;User initiates certificate application, and certificate issuer generates the encrypted pre-certificate in lock state, and generates the zero-knowledge proof of encrypted pre-certificate correct construction and sends to user, and user verifies in off-chain;When real event occurs, corresponding oracle signs event attribute and label, generates event signature and broadcasts to user;User collects to meet the event signature of first layer oracle authorization strategy, corresponding encrypted pre-certificate is verified after decryption zero-knowledge proof, if meet second layer event authorization strategy, then generate final valid certificate;User generates zero-knowledge display token, and submit to verifier only to verify validity;Privacy can be guaranteed, and running cost is reduced.