1141 results about "Time step" patented technology

A system and method are disclosed for detecting intrusions in a host system on a network. The intrusion detection system comprises an analysis engine configured to use continuations and apply forward- and backward-chaining using rules. Also provided are sensors, which communicate with the analysis engine using a meta-protocol in which the data packet comprises a 4-tuple. A configuration discovery mechanism locates host system files and communicates the locations to the analysis engine. A file processing mechanism matches contents of a deleted file to a directory or filename, and a directory processing mechanism extracts deallocated directory entries from a directory, creating a partial ordering of the entries. A signature checking mechanism computes the signature of a file and compares it to previously computed signatures. A buffer overflow attack detector compares access times of commands and their associated files. The intrusion detection system further includes a mechanism for checking timestamps to identify and analyze forward and backward time steps in a log file.

A system and method are disclosed for detecting intrusions in a host system on a network. The intrusion detection system comprises an analysis engine configured to use continuations and apply forward- and backward-chaining using rules. Also provided are sensors, which communicate with the analysis engine using a meta-protocol in which the data packet comprises a 4-tuple. A configuration discovery mechanism locates host system files and communicates the locations to the analysis engine. A file processing mechanism matches contents of a deleted file to a directory or filename, and a directory processing mechanism extracts deallocated directory entries from a directory, creating a partial ordering of the entries. A signature checking mechanism computes the signature of a file and compares it to previously computed signatures. A buffer overflow attack detector compares access times of commands and their associated files. The intrusion detection system further includes a mechanism for checking timestamps to identify and analyze forward and backward time steps in a log file.

A system and method are disclosed for detecting intrusions in a host system on a network. The intrusion detection system comprises an analysis engine configured to use continuations and apply forward- and backward-chaining using rules. Also provided are sensors, which communicate with the analysis engine using a meta-protocol in which the data packet comprises a 4-tuple. A configuration discovery mechanism locates host system files and communicates the locations to the analysis engine. A file processing mechanism matches contents of a deleted file to a directory or filename, and a directory processing mechanism extracts deallocated directory entries from a directory, creating a partial ordering of the entries. A signature checking mechanism computes the signature of a file and compares it to previously computed signatures. A buffer overflow attack detector compares access times of commands and their associated files. The intrusion detection system further includes a mechanism for checking timestamps to identify and analyze forward and backward time steps in a log file.

A method, system and apparatus for real-time reservoir model updating using ensemble Kalman filters is described. The method includes a conforming step for bring bringing static and dynamic state variables into conformance with one another during a time step of the updating. Also, an iterative damping method is used in conjunction with the conformance step to account for nonGaussian and nonlinear features in a system. Also, a re-sampling method is described which reduces the ensemble size of reservoir models which are to be updated.

The invention provides an autonomous underwater vehicle (AUV) trajectory tracking control method based on deep reinforcement learning, belonging to the field of deep reinforcement learning and intelligent control. The autonomous underwater vehicle trajectory tracking control method based on deep reinforcement learning includes the steps: defining an AUV trajectory tracking control problem; establishing a Markov decision-making process model of the AUV trajectory tracking problem; constructing a hybrid policy-evaluation network which consists of multiple policy networks and evaluation networks;and finally, solving the target policy of AUV trajectory tracking control by the constructed hybrid policy-evaluation network, for the multiple evaluation networks, evaluating the performance of eachevaluation network by defining an expected Bellman absolute error and updating only one evaluation network with the lowest performance at each time step, and for the multiple policy networks, randomly selecting one policy network at each time step and using a deterministic policy gradient to update, so that the finally learned policy is the mean value of all the policy networks. The autonomous underwater vehicle trajectory tracking control method based on deep reinforcement learning is not easy to be influenced by the bad AUV historical tracking trajectory, and has high precision.

The present invention provides a phase shift circuit that supports multiple frequency ranges. The phase shift circuit receives a plurality of control bits and causes a phase shift in a received signal, the phase shift corresponding to a number of time steps, the number depending on the control bits, and the time step is selected from a plurality of different time steps based on a frequency range associated with the received signal.

A method for tracking pedestrians in a video sequence, where each image frame of the video sequence corresponds to a time step, includes using marginal space learning to sample a prior probability distribution p(x_t|Z^t−1) of multi-person identity assignments given a set of feature measurements from all previous image frames, using marginal space learning to estimate an observation likelihood distribution p(z_t|x_t) of the set of features given a set of multi-person identity assignments sampled from the prior probability distribution, calculating a posterior probability distribution p(x_t|Z^t) from the observation likelihood distribution p(z_t|x_t) and the prior probability distribution p(x_t|Z^t−1), and using marginal space learning to estimate the prior probability distribution p(x_t+1|Z^t) for a next image frame given the posterior probability distribution p(x_t|Z^t) and a probability p(x_t+1|x_t), where the posterior probability distribution of multi-person identity assignments corresponds to a set of pedestrian detection hypotheses for the video sequence.

The invention provides a vehicle autonomous parking path programming method used for multiple parking scenes. The method is used for automatically parking a vehicle in a parking space through an autonomous parking system when the autonomous parking system detects the available parking space. The method includes the steps that target parking space information is detected, and a parking scene is determined; the initial state and target state of the to-be-parked vehicle are determined; a vehicle kinematics differential equation is established; state variables and control variables of the vehicle are segmented, equidistance sampling is performed on each segment according to certain time step, and to-be-optimized variables are obtained; an equality constraint, boundary constraints and inequality constraints of the to-be-optimized variables are formed; motion range constraints of the to-be-parked vehicle are formed according to the motion range limit in the parking process of the vehicle; an optimization objective is determined, and an objective function is established; and by means of a nonlinear programming solver, an optimal solution of a parking path is obtained. The vehicle autonomous parking path programming method is suitable for the multiple parking scenes, the design is reasonable, abundant information can be provided so as to control autonomous parking of the vehicle, and the security coefficient is high.

A system for correctly evaluating a price distribution and a risk distribution for a financial product or its derivatives introduces a probability density function generated with a Boltzmann model at a higher accuracy than the Gaussian distribution for a probability density. The system has an initial value setup unit and an evaluation condition setup unit. Initial values include at least one of price, price change rate, and the price change direction of a financial product. The evaluation conditions include at least time steps and the number of trials. The Boltzmann model analysis unit receives the initial values and the evaluation conditions, and repeats simulations of price fluctuation, based on the Boltzmann model using a Monte Carlo method. A velocity/direction distribution setup unit supplies the probability distributions of the price, price change rate, and the price change direction for the financial product to the Boltzmann model analysis unit. A random number generator for a Monte Carlo method employed in the analysis by the Boltzmann model, and an output unit displays the analysis result. A dealing system applies the financial Boltzmann model to option pricing, and reproduces the characteristics of Leptokurcity and Fat-tail by linear Boltzmann equation in order to define risk-neutral and unique probability measures. Consequently, option prices can be evaluated in a risk-neutral and unique manner, taking into account Leptokurcity and Fat-tail of a price change distribution.

A system and method are disclosed for detecting intrusions in a host system on a network. The intrusion detection system comprises an analysis engine configured to use continuations and apply forward- and backward-chaining using rules. Also provided are sensors, which communicate with the analysis engine using a meta-protocol in which the data packet comprises a 4-tuple. A configuration discovery mechanism locates host system files and communicates the locations to the analysis engine. A file processing mechanism matches contents of a deleted file to a directory or filename, and a directory processing mechanism extracts deallocated directory entries from a directory, creating a partial ordering of the entries. A signature checking mechanism computes the signature of a file and compares it to previously computed signatures. A buffer overflow attack detector compares access times of commands and their associated files. The intrusion detection system further includes a mechanism for checking timestamps to identify and analyze forward and backward time steps in a log file.

A recursive algorithm is provided for adaptive multi-parameter regression enhanced with forgetting factors unique to each regressed parameter. Applications of this algorithm can include lead acid batteries, nickel-metal hydride batteries, and lithium-ion batteries, among others. A control algorithm is presented, having an arbitrary number of model parameters, each having its own time-weighting factor. A method to determine optimal values for the time-weighting factors is included, to give greater effect to recently obtained data for the determination of a system's state. A methodology of weighted recursive least squares is employed, wherein the time weighting corresponds to the exponential-forgetting formalism. The derived mathematical result does not involve matrix inversion, and the method is iterative, i.e. each parameter is regressed individually at every time step.

Methods and systems for performing symbolic simulation, including techniques for translating a conventional simulation into a symbolic simulation, for handling wait and delay states, and for performing temporally out-of-order simulations. Additional techniques for extracting a signal graph from an HDL representation of a device, for representing signal values as functions of time using binary decision diagrams, and for computing minimal signal sets for accurate simulation. Techniques and methods for improving waveform dumping, reducing the waveform database, and for combining out-of-order simulation or reduced time steps with conventional time-based simulation.

An intelligent system for collecting readings from electric meters, comprising the following components: several electric meters, a data concentrator connected to the electric meters, and a system main station connected to the concentrator wirelessly, in which: the electric meter for recording the consumed quantity of electricity by user, metering the quantity of electricity, measuring demand and making a fault recording, and supporting a time-sharing measurement and communication with the data concentrator by electric power carrier; a concentrator for providing intelligent relay management, implementing the communication with the electric meters and the system main station, the management of communication route of electric meters and the storage function of data; the system main station for realizing a stable bi-directional communication with the concentrator and several electric meters, reading data in the concentrator, collecting information on electric energy, synchronizing the time of the concentrator and the electric meters associated therewith, renewing time step tariff rate and meter parameters of the electric meters. Bi-directional stable communication between the main station and the intelligent electric meters and collection of electrical energy information and the adjustment of parameters of electric meters can be realized by transmission medium in the present invention. The system of the present invention features complete functions, stable user-friendly operation interface, and has good expansion potential. Therefore this system is an important technical means for user-side management. The present invention greatly improves management efficiency and economic benefits in the electrical power department.

An output of a photovoltaic (PV) device is predicted by applying Fourier analysis to historical data to obtain frequencies and a mean of the frequencies in the data. Regression analysis is applied to the data to obtain a regression coefficient. Then, the prediction is a sum of the mean at the time step and a deviation from the mean at a previous time step, wherein the means are represented and approximated by selected frequencies, and the deviation for the previous time step is weighted by the regression coefficient.

Disclosed is a method for detecting additional relevant resources in relation to a predefined initial number of resources, consisting of the following steps: a) a set of referencing resources consisting of all resources being linked to at least one of the initial resources is detected; b) a set of potential resources consisting of all resources referenced by the referencing resources is established; c) for each potential resource, a potential-resource relevance score is calculated between said potential resource and all initial resources on the basis of existing links in the referencing resources to the potential resource and the initial resources, and also on the basis of the relevance scores of referencing resources assigned to each referencing resource; d) for each referencing resource, a referencing-resource relevance score is recalculated on the basis of links occurring in said referencing resource to the potential resources and also on the basis of the potential-resource relevance score assigned to the potential resources in step c); e) step c), and if necessary several times step d) followed by step c), is repeated if necessary; f) said additional relevant resources are identified as the potential resources having the highest potential-resource relevance scores.

To increase a feeling of competitiveness between players of a music-oriented game. Based on evaluation of performance of the first and second players on their steps, the number of times step guidance is presented is varied. Specifically, while reference arrows 46, 48 are displayed, a plurality of timing guidance arrows 50, 62 approaching the reference arrows 46, 48 are also displayed on a game screen image 40, whereby a state of step timing gradually arriving is shown to the player. In addition, the number of timing guidance arrows 50 62, namely the number of times step guidance is presented to each player, is changed based on the result of evaluation on steps by each player.

Disclosed is a digital neural processor comprising at least one neural processing element. The neural processing elements including at least one simulated dendrite and a simulated axon. Each of the simulated dendrites may include: a dendrite input capable of receiving at least one dendrite input signal and a dendrite signal propagation function capable of calculating a dendrite output signal in discrete time steps from each dendrite input signal. The signal propagation function may further include a delay parameter; a duration parameter; and an amplitude parameter. The simulated axon includes an axon input capable of receiving dendrite output signals, an axon function, capable of calculating an axon output signal from dendrite output signal(s) and an axon output capable of outputting the axon output signal.

The invention relates to a reservoir prediction method based on a bidirectional recurrent neural network. The reservoir prediction method includes: generating well seismic data, generating well markeddata, generating a marked dataset, building and training the bidirectional recurrent neural network, storing a trained model, extracting seismic data corresponding to all geographic positions in a region, and performing prediction on the seismic data to acquire probability distribution of reservoirs below the whole region. A one-to-one corresponding relation between each sampling point and reservoirs and non-reservoirs is established, wherein input of each time step is n-dimension seismic data of each sampling point while output of the same is corresponding reservoir or non-reservoir marks. An optimal model is acquired by training marked samples and adjusting hyper-parameters, in this way, unmarked data can be predicted to finally generate a predicted seismic body of same dimension, and value of each sampling point is prediction probability between 0 and 1, so that the method has good effect on reservoir prediction.