54 results about "Sequential computation" patented technology

A system is described for managing interaction between an untimed HAL portion and a timed HDL portion of the testbench, wherein the timed portion is embodied on an emulator and the un-timed portion executes on a workstation. Repeatability of verification results may be achieved even though the HAL portion and the HDL portion run in parallel with each other. A communication interface is also described for synchronizing and passing data between multiple HDL threads on the emulator domain and simultaneously-running multiple HAL threads on the workstation domain. In addition, a remote procedural-call-based communication link, transparent to the user, is generated between the workstation and the emulator. A technique provides for repeatability for blocking and non-blocking procedure calls. FSMs and synchronization logic are automatically inferred to implement remote procedural calls. A subset of behavioral language is identified that combines the power of conventional modeling paradigms with RTL performance.

A method for multi-platform parallel or real-time machine programming is provided. The method comprises specifying a parallel machine definition code defining a plurality of signal or data processing components and a communication of data or signals between the components, automatically converting the code into computer instructions for execution on an essentially sequential, non-parallel computer processor, a parallel execution of the instructions on the sequential processor and automatically converting the code into hardware programming data for providing parallel hardware operation according to the code, by including, in the hardware programming data, event control circuitry specification, priority control circuitry specification and buffering control circuitry specification to ensure that the hardware operation matches the execution on the sequential computer processor. The parallel machine code is implemented using the hardware programming data on a parallel hardware machine having parallel processing capability and using the computer instructions on the sequential processor in a manner that is essentially exchangeable.

A method and architecture for calculating input values for use in automated analysis and design during the various stages of the design process includes smart defaults. Specific parameter types are defined, some of which are adapted to be specified, or if not specified, calculated by the smart defaults. A tree structure is provided for determining parameter relationships in order to calculate parameter values. The tree structure uses a specific routine to call a smart default each time a parameter value is to be determined, and values are thereby calculated independent of the order of the data inputs.

A method for multi-platform parallel or real-time machine programming is provided. The method comprises specifying a parallel machine definition code defining a plurality of signal or data processing components and a communication of data or signals between the components, automatically converting the code into computer instructions for execution on an essentially sequential, non-parallel computer processor, a parallel execution of the instructions on the sequential processor and automatically converting the code into hardware programming data for providing parallel hardware operation according to the code, by including, in the hardware programming data, event control circuitry specification, priority control circuitry specification and buffering control circuitry specification to ensure that the hardware operation matches the execution on the sequential computer processor. The parallel machine code is implemented using the hardware programming data on a parallel hardware machine having parallel processing capability and using the computer instructions on the sequential processor in a manner that is essentially exchangeable.

A system and method for an on-line matching of job seekers with job openings and for score-based access to contact information is disclosed. Both the job seekers and the job openings are identified by their profiles and each profile has multiple parameters. The profile parameters are divided into two distinct categories, fundamental capabilities to perform a job and personal preferences regarding a job, to ensure that both the objective and subjective pictures of the job market are preserved. Correspondingly, two matching scores are calculated sequentially and the results presented simultaneously to job seekers and employers. A user's contact information is released only if the matching score meets the user's specified limit. The two score approach as well as the method for score-based access to contact information can work either together or separately in a variety of profile matching models.

The invention discloses a method for compiling the ladder diagram language into the instruction list language in accordance with the IEC 61131-3 standard. A ladder diagram applicable to the method can process logic quantity and non-logic quantity, and the obtained instruction list language accords with the IEC 61131-3 instruction set completely. The method specifically comprises: the ladder diagram is abstracted to a directed graph, and the varieties of the nodes in the directed graph are defined on the basis of the varieties of the elements in the ladder diagram; if the ladder diagram has a plurality of outputs, the directed graph will be decomposed into a plurality of subgraphs; the node stipulation method is used to judge whether the directed graph is a serial-parallel graph. The method currently only supports the compilation for the ladder diagram of a serial-parallel structure; the node stipulation method is used to build a binary decomposition tree, the result of post order traversal for the binary decomposition tree accords with the ladder diagram analysis method from top to bottom and left to right; the instruction of stipulation node of the binary decomposition tree is calculated in accordance with the sequence of post order traversal until the root node is calculated. The obtained instruction list of the root node is the final instruction list language.

The invention discloses a computation method and a path computation element (PCE) of a cross-domain disjoint path, and a communication system. The method comprises the following steps that: in the process of cross-domain disjoint path sequential computation, if computation of the disjoint path of own domain fails, the PCE of own domain regulates a path in the original path domain to a sub-optimal path under the condition that an enter boundary node and an exit boundary node in the original path domain are not changed; and the disjoint path corresponding to the sub-optimal path in a regulated original path domain is recomputed. By the method, the PCE and the communication system, a path segment which comprises nodes and links in the original path domain is subjected to sub-optimal regulation based on the conventional cross-domain path sequential computation method, so that a success rate for computing the disjoint path during cross-domain disjoint path computation is greatly improved.

The invention provides a differential new user selection access method applicable to a non-orthogonal multi-access system, and relates to differential new user selection access technology applicable to non-orthogonal multi-access systems. The differential new user selection access method is used for enabling new users in the non-orthogonal multi-access system to better access the network. The method comprises the following steps: a first step, establishing an information resource pool of a full-band candidate user set; a second step, selecting sub-bands which satisfy bandwidth requirements; a third step, sequentially calculating differences of signal to interference plus noise power ratios; a fourth step, calculating a ratio A<l> of the number of users, satisfying preset standards, of each sub-band to the total number of candidate users of the sub-band, and sorting the ratios A<l> from small to large; a fifth step, carrying out selection according to the expression described in the specification; and a sixth step, carrying out non-orthogonal coding on the new access user information and other candidate user information through superposition coding, and conducting signal transmission. The differential new user selection access method provided by the invention is applicable to the differential new user selection access of the non-orthogonal multi-access system, and can satisfy the demand of a 5G new user access solution in the future.

The present invention is an integral parallel machine for performing intensive computations. By combining data parallelism, time parallelism and speculative parallelism where data parallelism and time parallelism are segregated, efficient computations can be performed. Specifically, for sequential functions, the time parallel system in conjunction with an implementation for speculative parallelism is able to handle the sequential computations in a parallel manner. Each processing element in the time parallel system is able to perform a function and receives data from a prior processing element in the pipeline. Thus, after a latency period for filling the pipeline, a result is produced after clock cycle or other desired time period.

The invention discloses a method for identifying and estimating the water level of a hydrological ruler based on image analysis, which comprises the following steps: firstly, preliminary positioning of the hydrological ruler is realized through an HSV color space, and then an adaptive component map is constructed to realize accurate positioning of the hydrological ruler; The sample is expanded bydata enhancement technology, and the character segmentation is realized by fuzzy C clustering method for the positioned hydrological ruler image, then the training data set of convolution neural network is constructed and the whole character is recognized. The pixel heights of a plurality of complete characters E are calculated sequentially, and the mapping relationship between the pixel heights and the actual heights in the image is established based on the quadratic equation, and then the actual heights of the incomplete characters E are calculated. The invention can estimate the water depthby recognizing an image containing a hydrographic ruler, and is suitable for being integrated at a terminal, such as a front end of a camera, a camera and the like, and automatically estimating the water level of the water ruler.

A mobile station is capable of detecting a signal in the wireless communication systems using the Multiple Input Multiple Output (MIMO). The mobile station includes an apparatus that determines a vector of signals received by several receiving antennas. The apparatus estimates a channel between transmitting antenna and receiving antenna; forms a channel matrix; establishes, based on the channel state data, an order for detecting symbols transmitted by different transmitting antennas; calculates weight coefficients for detecting the symbols in the MIMO system; detects the symbols serially in the established order on the basis of the received signal vector; calculates the Euclidean distance between the detected symbols and the symbol constellation points; determines values of the Logarithmic Likelihood Ratio (LLR) for estimating the soft output bit probability, and forms a group of the most probable candidate symbols from the points of the symbol constellation.

The presented dynamic changing method for the maximal value of bandwidth request comprises: 1. during set period, BS counts current residual bandwidth, and calculates the maximal sending data length in a bandwidth allocation mapping figure according to the micro time slot number, bit, weight and terminal value; 2. BS sends the maximal length to UE; 3. the latter notifies received value to a group frame module; 4. the latter frames as the length; and 5. UE requests bandwidth as the length of frame. This invention improves bandwidth utility.

A time-series model learning unit reads in the time-series data sequentially, learns the parameters of a time-series model and stores the learned parameters in a complexity and parameter storage device. A loss function calculating unit reads in sequentially from the storage device with regard to each item of the input time-series data one at a time and calculates the values of a loss function. A complexity calculating unit sums the loss values to sequentially calculate complexity as fitting error that results when a time-series model is fit to the input time-series data. Complexity is stored in the storage device. A change-point searching unit reads in complexity regarding time-series data before and after change-point candidates from the storage device with respect to all change-point candidates, compares the sum of the complexities with complexity regarding all time-series data in a case where a change point is assumed to be absent, calculates the change-point score based upon the difference between the compared values, detects a change point in accordance with the magnitude of the change-point score and outputs the change point.

A method and system for parallelization of sequential computer program code are described. In one embodiment, an automatic parallelization system includes a syntactic analyser to analyze the structure of the sequential computer program code to identify the positions to insert SPI to the sequential computer code; a profiler for profiling the sequential computer program code by preparing call graph to determine dependency of each line of the sequential computer program code and the time required for the execution of each function of the sequential computer program code; an analyzer to determine parallelizability of the sequential computer program code from the information obtained by analysing and profiling of the sequential computer program code; and a code generator to insert SPI to the sequential computer program code upon determination of parallelizability to obtain parallel computer program code, which is further outputted to a parallel computing environment for execution and the method thereof.

According to the plant control device of this invention, based on a modified target value candidate of a control output, a reference governor uses a prediction model in which a closed loop system including a plant and a feedback controller are modeled to sequentially calculate, across a finite prediction horizon, a predicted value of state quantities of a plant including a specific state quantity on which a constraint is imposed. At such time, if a predicted value of a specific state quantity relating to a certain modified target value candidate conflicts with a constraint, the reference governor excludes the modified target value candidate from candidates for a final modified target value. Thus, the computational load required to modify a target value of a control output is decreased while ensuring the satisfiability of a constraint.

The invention discloses an entity relationship joint extraction model construction method based on a reinforcement learning algorithm, and the method comprises the steps: inputting a text, and carrying out the entity relationship marking of words of each statement in an original corpus through a joint extraction marking strategy; converting the text into a word2vec vector; pre-training an LSTM joint extractor; initializing a trainer network and disturbing a bag sequence; calculating a reward value of the current statement; calculating a total expected reward value; updating model parameters byusing the optimization function; if the model is converged, performing hyper-parameter tuning training on the model; and if the hyper-parameter is the optimal solution, generating a final entity relationship joint extraction scheme. According to the method, the reinforcement learning model is constructed based on the strategy gradient optimization algorithm, the statement entity relationship joint extraction problem of the complex natural language type original corpus can be effectively solved, and the entity relationship extraction accuracy and the F1 value can be effectively improved.

A method for the selective utilization of graphics processing unit (GPU) acceleration of database queries in database management is provided. The method includes receiving a database query in a database management system executing in memory of a host computing system. The method also includes estimating a time to complete processing of one or more operations of the database query using GPU accelerated computing in a GPU and also a time to complete processing of the operations using central processor unit (CPU) sequential computing of a CPU. Finally, the method includes routing the operations for processing using GPU accelerated computing if the estimated time to complete processing of the operations using GPU accelerated computing is less than an estimated time to complete processing of the operations using CPU sequential computing, but otherwise routing the operations for processing using CPU sequential computing.

The invention discloses a task planning method for an indoor visual-guidance AGV. The method comprises: firstly, according to a practical factory environment, a tag and environment database is established; secondly, a coordinate of each fixed AVG task point is stored in the database obtained at the first step; thirdly, an AGV task point is selected at a mobile terminal and factory modeling is carried out by using the selected AGV task point as a unit; fourthly, an optimal execution sequence is calculated by using a Floyd algorithm; fifthly, the mobile terminal sends the optimal execution sequence to a master station server and the master station server calculates a final path of vehicle operation according to the execution sequence; and sixthly, the master station server sends the final pat to an AGV and the AGV realizes task planning of the AGV based on the path. Therefore, the AGV can change the operation tasks in real time during the operation process and a task planning way of using the task point as a unit is employed, so that the influence on the whole system operation by the local environment change can be reduced.

The invention provides a behavior sequence anomaly detection method based on an unsupervised algorithm. The method comprises the steps: calculating the time interval of two operations based on the operation data of an enterprise web system through the sequence of user operations, and segmenting a user behavior sequence according to whether the time interval of the two operations is greater than a preset threshold or not, and training a probability suffix tree model, outputting a probability value corresponding to the user behavior sequence according to the probability suffix tree model, taking the probability value corresponding to the user as a feature, i.e., input of an isolated forest model, and judging whether the user behavior is abnormal or not according to a model output result.

A computer system execute summation processing even if the computing sequence is not adhered to in a system for computing a sum of floating point data of a plurality of nodes. Each node sends floating point data to a reduction mechanism, and the reduction mechanism computes the sums only for a group of which exponent sections have a highest value and a group of which exponent sections have a second highest value, and adds the sum of the group of which the exponent sections have a highest value and the sum of the group of which the exponent sections have a second highest value. By this, the consistency of the computation result can be guaranteed even if the sum is computed regardless the computing sequence of the values.

A time-series model learning unit reads in time-series data sequentially, learns the parameters of a time-series model and stores the parameters in a storage device. A loss function calculating unit reads in sequentially from storage each item of the input time-series data one at a time and calculates values of a loss function. A complexity calculating unit sums the loss values to sequentially calculate complexity as fitting error resulting when a time-series model is fit to the input data. Complexity is stored. A change-point searching unit reads in complexity regarding time-series data before and after change-point candidates from storage with respect to all change-point candidates, compares the sum of the complexities with complexity regarding all time-series data in a case where a change point is assumed to be absent, calculates the change-point score based upon the difference between the compared values, and detects and outputs the change point.

A method for the selective utilization of graphics processing unit (GPU) acceleration of database queries in database management is provided. The method includes receiving a database query in a database management system executing in memory of a host computing system. The method also includes estimating a time to complete processing of one or more operations of the database query using GPU accelerated computing in a GPU and also a time to complete processing of the operations using central processor unit (CPU) sequential computing of a CPU. Finally, the method includes routing the operations for processing using GPU accelerated computing if the estimated time to complete processing of the operations using GPU accelerated computing is less than an estimated time to complete processing of the operations using CPU sequential computing, but otherwise routing the operations for processing using CPU sequential computing.

The present invention proposes a method for dynamically changing the maximum length of a bandwidth request, including the following steps: 1. Within a set time period, the base station side counts the remaining bandwidth of the current system, according to the number of mini-slots, the number of bytes, and the weight 1. The sequence of the corresponding values ​​of the terminal calculates the maximum transmission data length of each terminal in a bandwidth allocation map; 2. The base station notifies the terminal of the maximum length of the transmitted data; 3. The terminal notifies the framing module of the received data length; 4. The framing module performs framing according to the data length; 5. The terminal makes a bandwidth request according to the length of the data frame. The method of the invention enables a single user to dynamically adjust the maximum length of the data frame requested by the bandwidth according to the current bandwidth of the system, dynamically changes the current flow according to the occupation of the system bandwidth, and effectively utilizes the system bandwidth.

The invention belongs to the technical field related to workshop scheduling, and discloses a job shop scheduling optimization method with a sufficient necessary condition neighborhood structure. The method comprises the following steps: obtaining a scheduled critical path according to a current solution of a job shop scheduling problem, and adjusting the processing sequence of procedures in a critical path block according to sufficient necessary conditions to obtain multiple different processing sequences; and calculating the total processing time of each processing sequence, and taking the processing sequence with the shortest total processing time as the current scheduling sequence. The sufficient necessary conditions are as follows: (1) a path from js (u) to v does not exist, and a neighborhood solution generated by processing after u moves to v is a sufficient necessary condition of a feasible solution; and (2) a path from u to jp (v) does not exist and is a sufficient necessary condition that a neighborhood solution generated by processing before v moves to u is a feasible solution. By means of the invention, when the job shop scheduling problem is solved, the solving qualityis effectively improved, and the solving time is shortened.

An interleaving method employing symbol interleaving, tone interleaving, and cyclic interleaving for transmitting data includes storing data at write address values in a memory which are sequentially calculated according to a predetermined process, and reading data stored at read address values of the memory which are sequentially calculated according to a predetermined process, wherein the memory has N data banks, each data bank has M storage spaces, and there are D interleaving target data having data numbers A, and the storing data operation includes storing the data at storage spaces of the data banks, the storage spaces corresponding to integer values and the data banks corresponding to remainders obtained by dividing the data numbers A by results of dividing a total number of data D by the number of storage spaces M.