79 results about "Iterative approximation" patented technology

The invention relates to an initializing method of a mobile satellite communication antenna. According to the method, the mobile satellite communication antenna provided with a low-cost inertia navigation system can fast align a satellite and accurately estimate an initial course angle of the low-cost inertia navigation system. The method comprises the steps of: firstly calculating a satellite searching instruction by utilizing the initial course and gesture information of the low-cost inertia navigation system, then controlling a pitch axis and a polarization axis of the antenna to rotate according to calculated pitch instruction angle and polarization instruction angle; controlling an azimuth axis to scan between 0 and 360 degrees at a constant speed, and finding out an accurate course instruction angle through a satellite beacon recognition technology; accurately estimating the initial course angle error of the low-cost inertia navigation system by utilizing three-coordinate calculation and an iterative algorithm; finally calculating the satellite searching instruction of the antenna again according to the estimated inertia navigation course and gesture information, and controlling the antenna to rotate according to the instruction angle and accurately align the satellite. After finishing initialization of the antenna by using the method, the antenna can track the satellite in real time with high precision.

A system and method are provided for designing RF pulses which have improved magnetization profiles. By utilizing an optimal control approach as an alternative to, or in combination with, non-iterative approximations, RF pulses generated by the system and method described herein will exhibit less deviation from that of “ideal” Bloch solutions. Consequently, the magnetization profiles produced by the RF pulses generated by the system and method described herein will be closer to the desired profiles. In addition, limitations of non-iterative approximations, such as maximum tip angle limits and linearity constraints, can be avoided.

The invention discloses a braking point determining method of a rail transit train operation path. The braking point determining method comprises the following steps: determining a train operation path; confirming speed limiting sections for which braking points need to be set in sequence; determining highest operation speed and traction distance of the train; if the highest operation speed is equal to a limiting speed, obtaining a train braking distance; if the distance difference is more than an error threshold, correspondingly changing the traction or constant speed operation distance into half of the difference; repeating the steps; if the highest operation speed is less than the limiting speed, reducing the highest operation speed by half of the difference; figuring out train traction and braking distances; if the distance difference is more than the error threshold, repeating the calculation by using a processing method that the highest operation speed is equal to the limiting speed; and finally, detecting whether braking points are obtained in all processing or not. According to the invention, by using an iterative approximation method, a braking curve meeting any accuracy is obtained by finite operation and the braking point is confirmed.

The invention relates to a kind of method and device to calculate synthesized focusing parameter of supersonic wave beam, and it sets the initial parameter, and adds 2*c2*n+2*tD*n to first intermediate variable sum to calculate the new value of first intermediate variable, and calculate the second intermediate variable sum_a: sum_a=sum-2*T-2*tD+1, and compares the absolute value of first intermediate variable sum and second intermediate variable sum_a, if |sum_a|<|sum|, it will iterative compute the focus matrix delay inequality tD, and the value of first intermediate variable sum is set to be the second intermediate variable sum_a, and the focus bite is 0 or 1 according to the comparing result and output. It just needs to load the limited initial parameter to obtain the ideal focusing parameters by the iterative computation. It fits for linear array, phased array and protrusion array. And it fits for multiple-beam, and supports the multiple-beam to receive the focusing calibration. It can realize in FPGA, and can be used in system with random channels and focuses.

The invention provides a design method for PNKF-VB and relates to the field of a filter. According to the method, variational distribution approximate to a posterior probability density function is constructed under a Gaussian assumption condition; iteration approximation of state estimation is realized by taking KL (Kullback-Leibler) divergence as a penalty function; and the PNKF-VB is derived bytaking confidence lower bound maximization as a target function according to a variational Bayesian frame. According to the method, a relatively tight approximation form of the nonlinear state posterior probability density function can be obtained, so the state estimation precision is improved. The difficult solution problem of the integral of the posterior probability density function in a system state estimation process can be transformed into the problem of optimizing ELBO lower bound. The adaptive weighted KL divergence is taken as the penalty function, so the optimization flexibility isimproved and the state estimation precision is improved. The method has very high significance to nonlinear state estimation theory and practical engineering application.

Image processing utilizing numerical calculation of fractional exponential powers of a diagonalizable numerical transform operator for use in an iterative or other larger computational environments. In one implementation, a computation involving a similarity transformation is partitioned so that one part remains fixed and may be reused in subsequent iterations. The numerical transform operator may be a discrete Fourier transform operator, discrete fractional Fourier transform operator, centered discrete fractional Fourier transform operator, and other operators, modeling propagation through physical media. Such iterative environments for these types of numerical calculations are useful in correcting the focus of misfocused images which may originate from optical processes involving light (for example, with a lens or lens system) or from particle beams (for example, in electron microscopy or ion lithography).

Minimum cost networks, such as cellular networks, able to supply k-fold coverage to users, are obtained by defining potential network elements having cost and coverage parameters, and selecting from these available network elements using an iterative approximation method or system that first constructs a k-set coverage formulation incorporating the cost and coverage parameters of the available network elements, and then finds an optimal solution to the k-set coverage formulation using applications of GRASP (greedy randomized adaptive search procedure) and path-relinking heuristics. In one version, GRASP and path-relinking start from a random choice of a network element and build feasible solutions using greedy construction, local search and path-relinking heuristics that are iteratively applied to find an approximate solution. In a second version, a Lagrangian relaxation formulation of the k-set coverage formulation is solved with iterative subgradient optimization methods, and a Lagrangian heuristic that finds a feasible solution at each iteration of the subgradient optimization method, to modify the subgradient optimization and find an approximate feasible solution. The Lagrangian heuristic may be a greedy construction heuristic applied at each iteration of the subgradient optimization method, or may be a GRASP with path-relinking heuristic applied at each iteration of the subgradient optimization method to find an approximate feasible solution. The approximate solutions to the k-set coverage formulation are then used to select at least some of the potential network elements for use in a minimum cost network. The method and system may be software or hardware implementations of conventional computer equipment enabling good approximate solutions to be obtained in reasonable times.

The invention discloses a video streaming business code rate self-adaption method based on online study. The method comprises the steps of firstly establishing a wireless video streaming business system frame, and generating different code rate levels to be selected in a way of using scalable video coding as a source end coding mode; then establishing a utility function, and using a discounted overall MOS value to represent QoE, so as to serve as an optimization goal of a code rate self-adaption mechanism; and establishing an action value function space, using an online study algorithm to perform iterative operation, updating the action value function space and a confidence coefficient distribution function according to the information of immediate MOS value and packet loss probability fed back by a terminal in each period, and accordingly selecting the proper video code rate for transmission. According to the method, the QoE is used as the optimization goal of self-adaption of the video streaming code rate, the optimal solution of code rate adjustment is iteratively approached by the online study algorithm, the link congestion can be effectively relieved, the user QoE is improved, and the computation complexity is relatively low.

In order to provide an image reconstruction device with which images having the quality required by the operator can be produced when cross-sectional images are reconstructed using iterative approximation, the invention is an image reconstruction device that performs iterative approximation based on the difference between measured projection data and calculated projection data obtained by forward projection of a reconstructed image reconstructed on the basis of the measured projection data. The image reconstruction device is characterized in being provided with: an input unit where image quality parameters that the operator considers to be important are input; a weighting computation unit that computes weighting coefficients based on the image quality parameter; and an iterative approximation unit that performs iterative approximation based on values obtained from weighted addition of pixel values that are about to be determined using the weighting coefficient and pixel values that were determined in the previous iteration.

The invention discloses a method for detecting a working point of a current transformer under a direct current magnetic biasing condition. The method comprises the following steps of determining alternating current magnetic flux of the current transformer by a numerical integration algorithm; superposing one direct current magnetic flux on the alternating current magnetic flux, calculating a corresponding exciting current instantaneous value and a direct current exciting current average value according to a magnetizing curve of an iron core, and calculating the direct current magnetic flux of the iron core of the current transformer by utilizing an iterative approximation method to determine the characteristic quantity associated with the working point of the current transformer under the direct current magnetic biasing condition. The method disclosed by the invention can be used for calculating and determining the characteristic quantities of the sizes of the alternating current magnetic flux and the direct current magnetic flux of the iron core of the current transformer under the direct current magnetic biasing condition, the value of exciting current, the values of alternating current exciting current and direct current exciting current, the current transformer secondary current and the like associated with the working point of the current transformer.

A CORDIC unit for the iterative approximation of a vector rotation through a rotary angle θ by a number of elementary rotations through elementary angles αi, including elementary rotation stages for respectively affecting an elementary rotation through an elementary angle αi as an iteration step in the iterative approximation. After such an elementary rotation there remains a residual angle through which rotation is still to be affected. The elementary rotation stages of the CORDIC unit are adapted for rotation through elementary angles αi given by powers of two with a negative integral exponent. The CORDIC unit can also include a triggering device for triggering the elementary rotations, a triggering device which is adapted prior to each iteration step to compare the residual angle to at least one of the elementary angles and to omit those elementary rotation stages whose elementary angles are greater than the residual angle.

A system and method are provided for designing RF pulses which have improved magnetization profiles. By utilizing an optimal control approach as an alternative to, or in combination with, non-iterative approximations, RF pulses generated by the system and method described herein will exhibit less deviation from that of “ideal” Bloch solutions. Consequently, the magnetization profiles produced by the RF pulses generated by the system and method described herein will be closer to the desired profiles. In addition, limitations of non-iterative approximations, such as maximum tip angle limits and linearity constraints, can be avoided.

The invention relates to a quick and active shimming method for a MRI (Magnetic Resonance Imaging) system. The method is characterized by comprising the following steps: 1, scanning parameters of the MRI system are set; 2, data are collected; 3, a magnetic field nonuniformity distribution field of the MRI system is calculated; 4, a complex iterative approximation method is adopted to obtain magnetization vector phase information; and 5, shimming current is calculated. Compared with the prior art, the quick and active shimming method for the MRI (Magnetic Resonance Imaging) system has the advantages that a double-echo sequence is used for collecting echo data; a complex fitting synthesis method is adopted to indirectly obtain the magnetic field nonuniformity distribution field of the MRI system; trouble on phase fitting caused by phase winding can be avoided; and thus, the method provided by the invention can realize accurate and stable shimming for the MRI system, phase unwinding is not needed, the calculation speed is fast, and accuracy is high.

The invention discloses a simplified design method for highway easement curves. The method comprises the steps of confirming a highway easement curve starting point, a highway easement curve terminal point and a highway easement curve starting point radius, a highway easement curve terminal point radius and highway easement curve parameters; setting a search line starting point as the highway easement curve starting point and a search line terminal point as the highway easement curve terminal point; calculating a maximum offset between a search line and a highway easement curve; and comparing the maximum offset with a preset accuracy control value to perform curve fitting. According to the simplified design method for highway easement curves, the design accuracy can be controlled accurately; fitting straight line or arc segments are few; control errors are calculated directly instead of being approximated iteratively, the speed is fast, and the efficiency is high while easement curves are drawn in mass; problems of difficulties in controlling calculation accuracy and calculating directly of parallel line drawing, relevant lengths and intersection and the like of easement curves are solved; and the problem that differences of chord arcs arranged on bridges on easement curves are difficult to accurately calculate is solved.

The invention relates to a per-survivor processing based iterative equalization method applied to direct-sequence spread-spectrum communication, belongs to a joint equalization and decoding method adaptable to quick time-varying channels and can effectively track the time-varying channels without insertion of extra training data. Compared with the conventional direct-sequence spread-spectrum communication signal processing method, the method can realize flexible regulation of the spread spectrum ratio and can track varying of the channels adaptively without extra training data overhead. In the method, direct-sequence spread-spectrum symbols are taken as processing units of a receiver, a 'per-survivor channel estimation method' is utilized in an iterative equalization detector, and the channel estimation is updated adaptively according to size of the 'channel estimation matching index'; the detector adopts a soft-input and soft-output detection algorithm, a subsequent decoder adopts a soft-input and soft-output iterative decoding algorithm, and soft information is exchanged between the detector and the decoder, thus, the optional solution is approached by means iteration.

The present invention provides a united dynamic resource allocation method of the energy efficiency optimization in an LTE system, relating to the united dynamic resource allocation method of the energy efficiency optimization in an LTE system. According to the method, stepwise resource scheduling is used, and the convex function of LTE system energy efficiency about spectral efficiency is used, which means that the energy efficiency is the convex function of transmission rate. When user dispatch is finished, an energy efficiency function is a section derivative continuous function, an optimal solution under an unconstrained condition is obtained, and thus an optimal rate closed-form solution under a satisfied constraint condition is obtained. Finaly the transmission power allocation is carried out according to the allocated transmission rate, the calculation complexity is effectivvey reduced through a stepwise resource scheduling strategy, a lot of iterative approximations are avoided, in the condition of ensuring user QoS requirements and spectrum efficiency, the energy efficiency is optimal, the energy utilization efficiency is improved, and thus the mobile communications is more environmentally friendly.

The invention provides a vascular remodeling method based on parameter deformation model energy optimization. A convenient tool is provided for directed treatment of vascular diseases. The method comprises the following steps of: 1. establishing a parameter deformation model; 2. performing generalized gradient vector current (GGVF) space back projection synthesis, namely performing back projection synthesis on a GGVF energy field vector in two contrastographic pictures in a three-dimensional space according to an adaptive weighting coefficient, and taking the synthesized vector as an external force field of the parameter deformation model; 3. performing parameter deformation curve iterative approximation, namely allowing the deformation curve to continuously and iteratively approximate to a real vascular center line due to the action of the external force and self-elasticity bending internal force of the GGVF synthesis model, and obtaining the result with minimum energy under the constraint conditions with minimum distance potential energy; 4. introducing a motion estimation matrix to represent the relative displacement of blood vessels in two contrastographic pictures due to the breathing movement and heartbeat; and 5. optimizing the reconstruction result.

The invention belongs to the fields of optics and automatic correction, and particularly relates to an infinite approximation automatic correction method and system based on optical axis consistency measurement. According to the method, a deviation angle between a laser optical axis and a white light axis is calculated by combining a real-time image of a visible light view field according to a position of a center spot of a laser beam in a three-light sensing imaging target, and a deviation angle between the laser optical axis and an infrared axis is also calculated in the same way; with the laser optical axis as a reference, a compensation error is sent through an automatic control protocol of a corrected photoelectric device, so as to perform first correction, a combined variance of first correction is calculated, the above steps are repeated till the variance is infinitely close to zero or a preset maximum error, or preset times for automatic correction is reached. Compared with traditional detection correction manners, the method based on image processing and automatic control technologies greatly reduces manual operation, saves time, and improves correction accuracy due to loop iterative approximation, thereby having high practical value.

In an aspect, a pipelined execution resource can produce an intermediate result for use in an iterative approximation algorithm in an odd number of clock cycles. The pipelined execution resource executes SIMD requests by staggering commencement of execution of the requests from a SIMD instruction. When executing one or more operations for a SIMD iterative approximation algorithm, and an operation for another SIMD iterative approximation algorithm is ready to begin execution, control logic causes intermediate results completed by the pipelined execution resource to pass through a wait state, before being used in a subsequent computation. This wait state presents two open scheduling cycles in which both parts of the next SIMD instruction can begin execution. Although the wait state increases latency to complete an in-progress algorithm, a total throughput of execution on the pipeline increases.

The invention discloses a mixed-region-based moving contour model image segmentation method. The method comprises: enhancement processing is carried out on an original image based on an image enhancement algorithm; a mixed-region-based moving contour model energy functional is built by using a local gray scale fitting item of the original image and a global gray scale fitting item of an enhanced image; and then the constructed energy functional is solved by a variational method, mathematical calculation is carried out by using solution framework of a variational level set and a contour curve is expressed in a hidden level set manner, the constructed energy functional is transformed into a partial differential equation by using a gradient descending flow and an Euler-Lagrangian equation, an optimal solution of the partial differential equation is obtained based on an iterative approximation method, and then a final target boundary contour is obtained. The method is not sensitive to the position of the initial contour curve and gray scale nonuniformity of the image; the segmentation efficiency and precision are high; and the influence of noises is low; and the robustness of the segmentation result is high.

The invention discloses an adaptive estimation method of Mars atmosphere entry based on model perturbation and belongs to the technical field of deep-space exploration. A coupling relation between uncertain parameters and an aerodynamic model of a Mars entry section is analyzed and perturbation on a dynamic system, caused by the uncertain parameters, is converted into aerodynamic model deviation so as to reduce the number of the uncertain parameters; according to a change rule of the perturbation of the aerodynamic model, filtering models aiming at perturbation amount deviation in the dynamic system are established; the perturbation amount in each filtering model dynamic system corresponds to a deviation value; a measurement residual error is predicated through each filtering module and a weight value of each model is self-adaptively updated based on measurement residual error information; and iterative approximation is close to real model perturbation so that influences of the aerodynamic model deviation on state estimation precision are inhibited. By adopting the self-adaptive estimation method disclosed by the invention, influences of the aerodynamic model perturbation on the state estimation precision in a Mars entry section combined navigation method can be reduced, and the state estimation precision of a detector in an entering process and the stability of a navigation system are guaranteed.

This X-ray CT device uses iterative reconstruction to obtain a CT image with the desired noise reduction or X-ray reduction ratio. An iterative approximation reconstruction unit (136) is provided which, from measured projection data obtained by an X-ray detection unit of the X-ray CT device, iteratively reconstructs a CT image in a reconstruction range of a subject, and iteratively corrects the CT image such that calculated projection data, calculated through forward projection of a CT image, is the same as the difference between measured projection data detected by the X-raw detection unit and calculated projection data. The iterative approximation reconstruction unit is provided with a parameter determination unit (151), an iterative correction unit (152), and a table unit (153) which calculates in advance the relation between each parameter used in the iterative reconstruction, and noise reduction or X-ray reduction ratio in the CT image. The parameter determination unit (151) determines the parameters from a calculation table in the table unit (153) depending on the desired reduction ratio.

Method and Apparatus for eliminating motion artifacts in magnetic resonance imaging are disclosed according to the present invention. The present invention relates to magnetic resonance imaging field. The method for eliminating motion artifacts in magnetic resonance imaging according to the present invention utilizes the concept of iterative approximation to control the difference between the data lines in the K-space caused by motions and allow the common features between the data lines to be remained, such that the motion artifacts in the reconstructed image are restrained and the motion artifacts caused under various circumstances are well restrained. Accordingly, the quality of the magnetic resonance imaging is improved.

The invention discloses a cloud resource scheduling method for heterogeneous costs under random demands. The method includes the following steps: adopting a random demand model to describe a fine-grained resource requirement of a user application program, adopting a general cost function to describe a heterogeneous cost model, establishing a nonlinear programming problem model with universality, quickly obtaining the lower bound of a solution based on dynamic programming, and then performing iterative approximation to obtain a near-optimal solution. A two-stage algorithm is adopted for scheduling, in the first stage, a quick acquisition algorithm is adopted to obtain a feasible solution of the scheduling problem, and in the second stage, a near-optimal solution acquisition algorithm is adopted to obtain a final solution of the scheduling problem. The scheme of the invention conforms to the actual situation of cloud platform scheduling, and fully utilizes the advantages of cloud resource scheduling.

The invention discloses an allocation design method for stiffness on different position of fixed joint surface of machine, comprising the steps of (1) confirming every parameter of machine to be analyzed; (2) extracting all degree of freedom of bed and column of machine to be analyzed; (3) building dynamic model of bed and column system; (4) confirming the design space of variable and building equation of motion for non-damping vibration of system with multi degree of freedom; (5) constructing two-phase response surface model and conducting allocation design for the stiffness on different position of fixed joint surface of machine by using genetic algorithm iterative approximation optimum seeking technology; and (6) using finite element analysis software ANSYS to verify the optimize results. The method processes estimation and allocation design for the stiffness on different position fixed joint surface of bed and column and thus improves dynamic character of machine.

The invention discloses a multi-target design method of a steering system structure. The multi-target design method comprises the following steps: while multi-target design is performed on the steering system structure, generating initial points through an optimal Latin-square experimental method by utilizing the thickness and material types of reinforcing plates of each partial structure in a steering system as design variables, and respectively establishing a Kriging agent model, a radial primary function agent model and a second-order polynomial response surface agent model of the steering system; by utilizing a local important region updating method and a stage sampling strategy, selecting sample points and updating an iterative approximation model; by aiming at different design targets, sequencing the sample points; and coupling the sample points so as to generate different design solution sets, performing selection, and performing design solution set selection and verification by applying a satisfaction degree function. By virtue of reasonable design on the approximation model and a sample point group, the structural design efficiency and structural design precision of the steering system structural design can be obviously improved, so that the performance of the steering system structure is effectively improved and the production cost is lowered.

The invention discloses an intelligent traffic signal control method. The method comprises the steps of obtaining road network traffic flow operation data and evaluating a current traffic jam value; constructing and training a traffic signal control model; and using the trained traffic signal control model to determine the road network state and obtain an optimal traffic signal control strategy capable of reducing road network traffic congestion. According to the method, traffic road network state data is processed through the convolutional neural network, the space-time dependence characteristics of traffic flow are effectively extracted, and meanwhile, the problem of huge signal phase combination space is solved in a value function iterative approximation mode by adopting an Actor-Critic reinforcement learning method, so that an intelligent traffic signal control model is established. The model is high in portability and can be suitable for various types of road traffic conditions.

This X-ray CT device uses iterative reconstruction to obtain a CT image with the desired noise reduction or X-ray reduction ratio. An iterative approximation reconstruction unit (136) is provided which, from measured projection data obtained by an X-ray detection unit of the X-ray CT device, iteratively reconstructs a CT image in a reconstruction range of a subject, and iteratively corrects the CT image such that calculated projection data, calculated through forward projection of a CT image, is the same as the difference between measured projection data detected by the X-raw detection unit and calculated projection data. The iterative approximation reconstruction unit is provided with a parameter determination unit (151), an iterative correction unit (152), and a table unit (153) which calculates in advance the relation between each parameter used in the iterative reconstruction, and noise reduction or X-ray reduction ratio in the CT image. The parameter determination unit (151) determines the parameters from a calculation table in the table unit (153) depending on the desired reduction ratio.

The invention provides an iterative approximation follow-up leveling control method for a beam-pumping unit, and the method comprises the steps: firstly analyzing the movement rule and energy circulation mechanism of a pumping unit system, and building an incidence relation mathematic model between the change of an input electric power integral value of a motor and the balance degree of the pumping unit; secondly, collecting real-time operation electrical parameters of the beam-pumping unit, and calculating the power integral ratio of unit stroke or multi-stroke up-stroke and down-stroke to detect the balance degree of a pumping unit system on line; and finally, according to the calculated balance degree, judging the balance state of the oil pumping unit operation system, and adopting a servo method to control a motor to drive a lead screw to adjust the position of the balance box body to change the balance torque so as to adjust the optimal balance state of the oil pumping unit. Compared with the prior art, the iterative approximation follow-up leveling control method and the iterative approximation follow-up leveling control device for the beam-pumping unit have the advantages that the working efficiency and the production safety coefficient of balance adjustment of the pumping unit are improved, the energy saving rate can reach 3 percent, and the engineering applicability ishigh.