22 results about "Incremental encoding" patented technology

The present application relates to the technical field of audio codec, and particularly relates to a fast coding and decoding method and system for real-time audio.The method comprises the following steps: obtaining real-time sampling data of an input audio stream, performing multi-time point sliding window division and coding preprocessing, and generating coding preprocessing enhanced audio; performing adaptive quantization strategy analysis on the coding preprocessing enhanced audio, and performing multi-level residual quantization modeling to construct a real-time incremental coding mapping table; performing multi-module decomposition and execution based on the real-time incremental coding mapping table, and performing multi-thread parallel coding optimization to construct a real-time audio parallel coding strategy; performing real-time coding processing according to the real-time audio parallel coding strategy, and performing task-by-task rate-distortion optimization to obtain optimal coding parameters of each coding task.The present application realizes high-fidelity rate and low-delay audio coding and decoding, and improves user auditory experience.

The invention provides a multi-agent reasoning optimization method and system based on Token calculation acceleration, and relates to the field of intelligent decision making, and the method comprises the steps: generating and maintaining a Token-level directed acyclic graph on line; constructing a reasoning path consistency sharing strategy, an approximate semantic sharing strategy and a forced sharing strategy to perform KV cache sharing; local change features are processed in an incremental coding mode, and context difference detection, incremental coding calculation and KV cache splicing are carried out; defining a contribution degree function for each Token so as to carry out Token contribution degree evaluation, and introducing a dynamic pruning mechanism based on the contribution degree; based on the Token-level directed acyclic graph, the parallel execution of the independent Token is realized through topology analysis and dynamic scheduling. According to the method, generation of redundant Token can be avoided, and the communication bandwidth and computing resources of the spacecraft are saved; the delay of the critical path is shortened, the decision-making efficiency of a multi-agent cooperation system can be improved, and the strict requirements of spacecraft tasks on the decision-making efficiency and the decision-making safety are met.

The application relates to the technical field of hydrogeological monitoring, in particular to a hydrogeological dynamic monitoring and analysis system and method based on big data; specifically, hydrogeological data is collected in real time, data denoising, space-time calibration and error correction are carried out in combination with edge computing, noise and drift error are eliminated by adopting variational mode decomposition, geological model constraint and wavelet transform, and the storage and transmission efficiency is optimized by increment encoding and Huffman encoding, and a quadratic form check code is generated to ensure data reliability; a space-time deep learning model and a self-attention mechanism are fused to predict a hydrological evolution trend, adaptive anomaly detection is realized by combining DBSCAN clustering and prediction error analysis; water resource allocation strategies are optimized through reinforcement learning, risk levels are evaluated in layers, and management schemes are generated, and finally, monitoring data, abnormal scores and decision results can be visually displayed. The application integrates multi-source data fusion, space-time modeling and intelligent decision-making, and improves hydrological monitoring precision and response efficiency.

The invention provides a double-track absolute position detection method and system, computer equipment and a storage medium, and the method comprises the steps: detecting a to-be-detected object through a double-track magnetic railing ruler, enabling the to-be-detected object to be connected with a reading head, enabling the reading head to move along with the to-be-detected object, and enabling the reading head to read the to-be-detected object when the to-be-detected object linearly moves relative to the double-track magnetic railing ruler. Incremental magnetic track information of the incremental coding magnetic track is read through an incremental coding reading head; a level signal is output through a plurality of switch position detection chips in an absolute coding reading head based on the polarity of an absolute coding magnetic track position closest to the switch position detection chips, and absolute magnetic track information is determined according to the level signal; and performing combined calculation according to the incremental magnetic track information and the level signal to obtain a double-magnetic-track absolute position detection result. According to the method, the defect that an existing incremental displacement sensor needs to make a zero position again or search the zero position after being powered off or restarted is overcome, and the application scene of non-contact displacement detection is expanded.

A method of database operations includes receiving a user query, generating a query vector embedding representative of the user query, querying a vector database using the query vector embedding, retrieving a first database vector of the plurality of database vectors based on the query and representative of a first data file corresponding to a first time and belonging to a first time-series data set, receiving a first plurality of delta encodings describing differences between vector representations of temporally-adjacent data files of the first time-series data set, identifying a second data file of the first time-series data set having a second vector representation that differs from the first database vector and corresponds to a second time, and retrieving the second data file from a database.

This invention proposes an integer-order differential analog-to-digital converter (ADC) for intracranial neural potential acquisition. First, an integer-order differential operator is designed to calculate arbitrary integer-order incremental codes, which more effectively utilizes the sparsity of intracranial neural potentials, significantly improving the compression rate of intracranial neural potentials. It also offers greater flexibility, with adjustable differential order, adapting to a wider range of intracranial neural potential acquisition applications. Second, a multi-level buffer matrix module is designed to generate arbitrary integer-order incremental code values ​​in a concise and efficient manner. Finally, a serialized bitstream generator is designed to serialize and package integer-order incremental data acquired from multiple parallel channels. This not only offers high scalability but also allows for reliable interfacing with subsequent synchronous clock wireless transmission systems, avoiding the overhead and latency of synchronous-asynchronous interface coordination, further improving signal acquisition efficiency.

Provided in the present invention is a fractional-order difference analog-to-digital converter for collecting a neural spike potential. Firstly, a fractional-order difference operator is designed to calculate arbitrary fractional-order delta encoding, such that the sparsity of a neural spike potential can be more effectively utilized, thereby significantly improving the compression ratio of the neural spike potential; moreover, the fractional-order difference analog-to-digital converter has higher flexibility by means of tunable differential orders, and can adapt to a wider range of neural spike potential collection scenarios. Secondly, a fractional-order buffer module is designed, such that an arbitrary fractional-order delta encoding value can be generated in a simple and efficient manner. Finally, a serialized bitstream generator is designed, such that serialized packaging can be performed on fractional-order delta data collected by a plurality of parallel channels, which not only exhibits high scalability and is applicable to multi-channel collection, but also enables reliable interfacing with a subsequent synchronous clock wireless transmission system, thus avoiding overheads and latency of synchronous-asynchronous interface coordination, and further improving the efficiency of neural spike potential collection.

A mechanism for processing video data is disclosed. The mechanism includes determining that a supplemental enhancement information (SEI) message is included in a bitstream, in which the SEI message carries danmaku information. A conversion is performed between a visual media data and the bitstream based on the SEI message carrying danmaku information.

Provided are a decoder unit, a decompressor unit, and method for a decoder of multiple stages of adders to decode delta encoded data. A decoder unit implemented in a cache memory having a cache memory cell array comprises a first stage of adder circuits to add, in parallel, pairs of encoded items transformed using a delta encoding, wherein the encoded items include a plurality of deltas of neighbors of sequential source items. The decoder unit further comprises at least one successive stage of adder circuits to add, in parallel in each stage, pairs of outputs from a previous stage of adder circuits, wherein each successive stage has fewer adder circuits than the previous stage, and wherein output from a last of the at least one successive stage comprises the sequential source items.

The present application relates to the technical field of positioning and speed measurement of super-speed maglev train, in particular to a super-speed positioning and speed measurement method and system based on laser coding, mainly including a plurality of laser code readers with unique numbers which are uniformly distributed along the track of the train, and the laser signals read each time are coded; the activation state of the laser code reader is determined according to the signal vector, and the state information of the laser code reader is transmitted to the positioning and speed measurement algorithm model. The laser propagation speed is fast, the photoelectric response speed is fast, and the positioning and speed measurement algorithm is simple, so that the delay of the speed positioning method can reach microseconds; through the combination of absolute coding and incremental coding, the problem of initial positioning loss is solved, the design of the ID code plate can make the laser coding not affected by the vertical floating of the train, and the redundant setting of the response of the two laser code readers increases the reliability and safety of the positioning and speed measurement algorithm results.

This invention provides a novel, reversible data transformation method that, while employing a simple algorithm, offers the potential for data scrambling while simultaneously producing output of the same length as the original data. [Solution] The method calculates the Hamming weight of the input symbol sequence and reverses the symbol at the Hamming weight position from the start or end until the symbol sequence reaches a predetermined convergence state (e.g., all zeros). The history information obtained during this reversal process is encoded using delta coding and Huffman transform based on occurrence intervals, etc. Finally, padding is applied to the encoded data based on the original data length, and converted data with the same length as the original input symbol sequence is output. This makes it possible to achieve data transformation that eliminates statistical bias in the data and has both confidentiality and diffusion capabilities, despite being a simple algorithm.

This invention provides a method and apparatus for analyzing incremental encoder output signals. The method includes: acquiring a first set of pulse signals and a second set of pulse signals from the incremental encoder output signal, wherein the first set of pulse signals includes amplitude, phase difference, and clutter; simulating and reproducing a Lissajous image fitted by the first and second sets of pulse signals to obtain the roundness of a standard circle; calculating a corrected Lissajous image to obtain the corrected roundness; comparing the corrected roundness with the roundness of the standard circle to confirm the failure weight of the Lissajous image; the corrected roundness includes a first corrected roundness, a second corrected roundness, and a third corrected roundness, which are calculated using clutter, amplitude, and phase difference, respectively. The incremental encoder output signal analysis method and apparatus provided by this invention improves the accuracy of the detection signal by comparing the corrected roundness with the roundness of the standard circle to confirm the failure weight of the Lissajous image.

The application provides a fractional order difference analog-to-digital conversion device for neural spike acquisition. First, a fractional order difference operator is designed to calculate arbitrary fractional order delta coding, which can more effectively utilize the sparsity of neural spikes, significantly improve the compression rate of neural spikes, and have higher flexibility, adjustable difference order, and can adapt to more neural spike acquisition application scenarios. Secondly, a fractional order buffer module is designed, which can generate arbitrary fractional order delta coding values in a simple and efficient way. Finally, a serialization bit stream generator is designed, which can serialize and pack the fractional order delta data collected by multiple parallel channels. It not only has high scalability and can be applied to multi-channel acquisition, but also can be reliably connected with the subsequent synchronous clock wireless transmission system, avoiding the overhead and delay of synchronous- asynchronous interface coordination, and further improving the neural spike acquisition efficiency.

Provided in the present invention is a total variation analog-to-digital converter for collecting an electroencephalogram. Firstly, a total variation operator is used to calculate second-order delta encoding, such that the sparsity of an electroencephalogram signal can be more effectively utilized, thereby significantly improving the compression ratio of the electroencephalogram signal. Secondly, a non-switching double-buffer module is designed, such that a second-order delta encoding value can be generated in a simple and efficient manner. Finally, a serialized bitstream generator is designed, such that serialized packaging can be performed on second-order delta data collected by a plurality of parallel channels, which not only exhibits high scalability and is applicable to multi-channel collection, but also enables reliable interfacing with a subsequent synchronous clock wireless transmission system, thus avoiding overheads and latency of synchronous-asynchronous interface coordination, and further improving the signal collection efficiency.

The application discloses a sepsis organ dysfunction early warning method based on sparse data incremental coding, comprising the following steps: obtaining event sequence data and non-event sequence data of a target object, wherein the event sequence data comprises at least one of clinical diagnosis data, treatment record data and laboratory history detection data, and the non-event sequence data comprises at least past medical history data and target sign data; performing incremental coding on the event sequence data to obtain a sparse mode of the event sequence data; processing the sparse mode, the event sequence data and the non-event sequence data according to a pre-constructed early warning model to obtain a sepsis risk index of the target object; and outputting target early warning information in the case that the sepsis risk index meets preset early warning conditions.

A mechanism for processing video data is disclosed. The mechanism includes determining that a number of syntax elements representing motion vectors for vertices is not present in a bitstream and that the syntax elements have a value of 0. A conversion is performed between a visual media data and a bitstream based on the syntax elements.

The invention provides a traceability log storage and query method based on deep embedding and single-layer RNN prediction, and relates to the technical field of traceability graph compression, and the method comprises the steps: converting a system bottom layer audit log into a traceability graph, carrying out the coding mapping of path class and identification class attribute fields through deep learning embedding representation, and obtaining a traceability graph; coding prefixes are distributed for the path type attribute fields and the identification type attribute fields through self-adaptive dictionary coding, and numerical attribute fields are recorded through incremental coding; preprocessing and learning the traceability graph through a single-layer recurrent neural network (RNN) unit; performing local adjustment on the traceability graph during data updating based on incremental coding and adaptive mapping; and mapping the query condition into an attribute coding prefix, reconstructing an attribute sequence based on the attribute coding prefix, and performing correction through the adjustment table. For redundant information and structural similarity, a deep learning embedding representation technology and an adaptive dictionary coding strategy are adopted, and shared codes of similar attribute fields are effectively learned and extracted, so that redundant data storage is reduced.

A method of encoding (e.g. compressing) data is disclosed. The data comprises a sequence of values. It is encoded by delta-encoding the sequence of values to generate a sequence of delta values, and encoding each delta value as an offset into a lookup table of delta values, thereby generating a sequence of offsets. The sequence of offsets, or data representative thereof, is stored or output as encoded data.

A method of vector database compression includes receiving first time-series vector data comprising a first plurality of vectors representative of data corresponding to a first time point, receiving second time-series vector data comprising a second plurality of vectors representative of data corresponding to a second time point, generating a plurality of delta encodings, discarding the second time-series vector data after generating the plurality of delta encodings, and modifying database data to store the plurality of delta encodings. Each vector of the second plurality of vectors corresponds to one vector of the first plurality of vectors. The plurality of delta encodings is generated by, for all corresponding vectors of the first plurality of vectors and the second plurality of vectors, generating one delta encoding that describes differences between values of corresponding elements of the vector of the second plurality of vectors and the corresponding vector of the first plurality of vectors.

The application provides an absolute value encoding method, system and program product of an optical magnetic encoder. The method is applied to an encoder system, and the method comprises the following steps: in the case that a magnetic absolute value encoder and an optical incremental encoder are both in a non-fault state, obtaining a fused absolute position value based on a magnetic absolute position value of the magnetic absolute value encoder and an optical incremental value of the optical incremental encoder, the fused absolute position value serving as an output absolute position value of the encoder system, wherein the magnetic absolute position value of the magnetic absolute value encoder is used to determine an optical reference position of the optical incremental encoder; the magnetic absolute position value has a first precision, and the fused absolute position value has a second precision, and the second precision is higher than the first precision. Therefore, the absolute position of the movement of a mechanical component can be reliably and accurately obtained.

Provided in the present invention is an integer-order difference analog-to-digital converter for collecting an intracranial neural field potential. Firstly, an integer-order difference operator is designed to calculate arbitrary integer-order delta encoding, such that the sparsity of an intracranial neural field potential can be more effectively utilized, thereby significantly improving the compression ratio of the intracranial neural field potential; moreover, the integer-order difference analog-to-digital converter has higher flexibility by means of tunable differential orders, and can adapt to a wider range of intracranial neural field potential collection scenarios. Secondly, a multi-stage buffer matrix module is designed, such that an arbitrary integer-order delta encoding value can be generated in a simple and efficient manner. Finally, a serialized bitstream generator is designed, such that serialized packaging can be performed on integer-order delta data collected by a plurality of parallel channels, which not only exhibits high scalability, but also enables reliable interfacing with a subsequent synchronous clock wireless transmission system, thus avoiding overheads and latency of synchronous-asynchronous interface coordination, and further improving the signal collection efficiency.

A mechanism for processing video data is disclosed. The mechanism includes determining that a supplemental enhancement information (SEI) message is included in a bitstream, in which the SEI message carries danmaku information. A conversion is performed between a visual media data and the bitstream based on the SEI message carrying danmaku information.