72 results about "Round-off error" patented technology

When processing a signal having a sequence of discrete values, wherein there is a first frequency range, in which the signal has a high energy, and wherein there is a second frequency range, in which the signal has a low energy, the sequence of discrete values is first manipulated to obtain a sequence of manipulated values, so that at least one of the manipulated values is non-integer. Then the sequence of manipulated values is rounded to obtain a sequence of manipulated values. The rounding is formed to effect a spectral shaping of a generated rounding error so that a spectrally shaped rounding error has a higher energy in the first frequency range than in the second frequency range. By spectrally shaping the rounding error so that the rounding error does not have any energy either in the storage areas where there is no signal energy, an especially efficient coding is obtained particularly in connection with a lossless coding context.

The present invention provides a method for avoiding rounding errors during rounding of values after performing an inverse discrete orthogonal transformation. In a first step, a block of coefficients (F′[u][v]) is transformed into a block of image pixel values (f′[y][x]) by means of an inverse discrete orthogonal transformation, wherein each image pixel comprises an image pixel value. In a second step, a product of the block of image pixel values (f′[y][x]) with a first pixel pattern (A[y][x]) is calculated. In a third step, the product values of the second step are summed up to a first sum (Sum1′). In a fourth step, it is determined whether the first sum (Sum1′) is even or odd. In a fifth step, if the first sum (Sum1′) is even, a block of manipulation values (C[y][x]) is added to or subtracted from the block of image pixel values (f′[y][x]) for generating a manipulated block of image pixel values (f[y][x]), wherein the block of manipulation values (C[y][x]) is formed in such a way that rounding errors are avoided in a subsequent rounding operation.

Systems and methods in accordance with various embodiments of the present invention provide for the rendering of data visualizations with minimal round-off errors. The data visualization method represents a dataset as graphical elements within a graphical layout area by performing layout of a data value within the dataset on the graphical layout area. The data value is positioned on the graphical layout area by determining a floating point position of the data value. A round-off error is calculated by casting the floating point position to an integer position on the graphical layout area. The round-off error is distributed to one or more subsequent data values that are yet to be positioned on the graphical layout area.

The invention discloses a neutral buoyancy robot pose and trajectory control method based on active disturbance rejection theory, the method aims at a round-off error problem caused by poor robot anti-interference ability and finite word length in a neutral buoyancy system, and delta operator theory is utilized to construct a neutral buoyancy system delta domain dynamic state equation. A delta domain tracking differentiator is designed, and a input signal transient process is arranged. A delta domain extended state observer is designed for coupling terms between water resistance and control forces in the neutral buoyancy system. The output state of the tracking differentiator and the estimation state of the extended state observer are utilized to design a delta domain compound anti-interference controller, and robot pose and trajectory control in the neutral buoyancy system is completed. According to the invention, the extended state observer is utilized to estimate total interference in the neutral buoyancy system and intra-system non-linearly coupled terms, the system anti-interference ability is improved, and the round-off error of the finite word length of system parameters is effectively reduced, and the method can be used in robot pose and trajectory high precision control under microgravity environment.

One embodiment of the present invention sets forth a technique for consistently evaluating geometric patches with shared boundaries using barycentric coordinates. A barycentric parameter is generated and represented using a fixed-point fraction. The barycentric parameter is then used to generate a fixed-point barycentric coordinate. The fixed-point barycentric coordinate is then converted to a floating-point representation for evaluating the geometric patches. Computing shared boundary splits using fixed-point fractions eliminates inconsistencies in associated barycentric coordinates due to round-off errors. Evaluating geometric patch equations using consistent barycentric coordinates facilitates precise, consistent computation of vertices along shared boundaries.

The invention relates to a satellite attitude determining method based on an improved self-adaptive square root UKF (Unscented Kalman Filter) algorithm, which belongs to the technical field of satellite attitude determination and solves the problems that a satellite attitude determining system is unstable, the satellite attitude precision is low, and the traceability to an virtual condition of a satellite is weak due to overlarge round-off errors numerically calculated by virtue of existing EKF (Extended Kalman Filter), UKF (Unscented Kalman Filter) and SRUKF (Square Root Unscented Kalman Filter) algorithms when the satellite attitude determining system suffers from uncertain interferences and is influenced by noise. The satellite attitude determining method comprises the main realization processes: estimating an error quaternion and a gyroscopic drift error by virtue of the improved self-adaptive square root UKF; substituting gyroscopic measurement value and the estimated gyroscopic drift error into an attitude kinematical equation to calculate an attitude quaternion; correcting the calculated attitude quaternion by virtue of the estimated error quaternion; carrying out attitude resolving by virtue of the corrected attitude quaternion so as to determine the attitude of the satellite. The satellite attitude determining method is suitable for the technical field of the satellite attitude determination.

Methods and systems for performing arithmetic functions. In accordance with a first aspect of the invention, methods and apparatus are provided, working in conjunction of software algorithms and hardware implementation of class network routing, to achieve a very significant reduction in the time required for global arithmetic operation on the torus. Therefore, it leads to greater scalability of applications running on large parallel machines. The invention involves three steps in improving the efficiency and accuracy of global operations: (1) Ensuring, when necessary, that all the nodes do the global operation on the data in the same order and so obtain a unique answer, independent of roundoff error; (2) Using the topology of the torus to minimize the number of hops and the bidirectional capabilities of the network to reduce the number of time steps in the data transfer operation to an absolute minimum; and (3) Using class function routing to reduce latency in the data transfer. With the method of this invention, every single element is injected into the network only once and it will be stored and forwarded without any further software overhead. In accordance with a second aspect of the invention, methods and systems are provided to efficiently implement global arithmetic operations on a network that supports the global combining operations. The latency of doing such global operations are greatly reduced by using these methods.

An apparatus and method are disclosed for minimizing accumulated rounding errors in coefficient values in a lookup table for interpolating polynomials. Unlike prior art methods that individually round each polynomial coefficient of a function, the method of the present invention use a “ripple carry” rounding method to round each coefficient using information from the previously rounded coefficient. The “ripple carry” method generates rounded coefficients that significantly improve the total rounding error for the function.

A tetrahedralization and triangulation method used with the proximity based rounding method to satisfy topological consistency of tetrahedralization with the bounded precision of a digital computer is described. Tetrahedralization is applied to a VLSI design, and more specifically for solving Maxwell's equation to extract parasitic capacitances and 3-D optical proximity correction applications. The exactness of solving Maxwell's equation and finite element analysis depends on the correctness of the topological properties of the tetrahedralization. Among the important aspects of the correctness of the topological properties is the absence of spurious intersection of two or more tetrahedra. In a typical digital computer, numbers are represented using finite sized words. Round-off errors occur when a long number is represented using the finite word size. As a result, tetrahedralization loses its topological consistency. The proximity based rounding method finds potential locations of spurious intersections and pre-corrects these locations to avoid the generation of any topological inconsistencies.

A sample rate converter (200) is described for converting an input data stream including a plurality of input samples (X) at one sample rate to an output data stream including a plurality of output samples (Y) at another sample rate. The converter uses an interpolation approach that utilizes an integer accumulator (220) to track the timing relation between input samples and output samples. Based on the value of the accumulator, the method determines if the correct input samples are being used to calculate the current output sample. If so, the output sample is calculated as a function of the input samples and the accumulator value. The converter provides the robustness of a table based conversion approach without the need to precalculate and store a table, simplifies the calculations involved, and is less sensitive to numeric round off errors.

The invention discloses a programmable IIR filter analog hardware implementation method based on a memristor, which comprises the following steps of: (1) sampling an analog input signal by utilizing a switched capacitor circuit to obtain a digital sampled signal of the analog input signal; (2) carrying out delay on the sampled digital signal by using a single-cycle delay circuit to obtain a delayed signal of the intermediate digital signal; (3) according to a direct type structure of an IIR filter, establishing two parts of circuits by utilizing a cross matrix structure of the memristor and an adder, and respectively using two parts of circuits as a feedback circuit and weighted summation output; and (4) selecting and setting a resistance value of the memristor, completing setting of filter parameters, inputting a to-be-filtered analog signal, and obtaining a filter signal at an output end of an operational amplifier. According to the programmable IIR filter analog hardware implementation method disclosed by the invention, the circuits are programmable, a round-off error can be reduced, a filtering speed is also not limited to an analog-digital converter, the complex multiplication calculation is avoided, all the calculations are in parallel, and the programmable IIR filter analog hardware implementation method can be used for implementing a high-order IIR filter.

The invention provides an image color correction method and device based on an FPGA. The method comprises the following steps: carrying out color space conversion on input image data to obtain converted multiple paths of A bit data; extracting high B bit data of each path of A bit data in the multiple paths of A bit data to carry out table look-up to obtain at least two data with adjacent addresses and bit widths of C bits; and carrying out an interpolation operation on the low D bit data of each path of A bit data and the at least two data with adjacent addresses and bit widths of C bits of each path of A bit data to obtain output image data. The image color correction method provided by the invention is used for solving the problem in related technology of relatively large error and lost image details caused by GAMMA correction and color space conversion in the case of limited FPGA resources, so as to reduce the rounding errors in an image color correction process and perfect the dark image details after the color space conversion.

Provided is an interval prediction control modeling and optimizing method based on soft constraints. The control method comprises the following steps: (1) a quadratic performance index including a constraint item, a control item and an economic item is established based on a process prediction model; (2) whether an overall optimization method is feasible is judged by solving a slack variable; (3) a method for solving a soft constraint slack variable when a control model output constraint is not feasible is provided, and adjustment of the range of a feasible region when an interval prediction control model output constraint is not feasible is realized; and (4) a boundary feasible sequence quadratic programming method is adopted to solve the problem that poor initial point selection causes calculation amount increase of the method or difficulty in finding an optimal solution, the problem that the positive definiteness of a Hessian matrix is destroyed due to the influence of round-off error in calculation, and the like, and to figure out the optimal control input. A complicated multivariable system control model can be established, the control law can be solved accurately and quickly based on soft constraint adjustment, and good control on a multivariable system can be achieved.

The invention relates to a sine-cosine function IP core capable of reconfiguring a spaceborne computer and a control method thereof, relating to the technical field of aerospace electronics, and solving the problems that the result of initial value assignment is largely influenced due to rounding errors, and the larger the angle is, the more the iteration times is needed. An initializing module is connected with a right-shift N-bit module and a module 1, the module 1 is connected with a cosine value storing device, the right-shift N-bit module is connected with a sine value storing device, the cosine value storing device and the sine value storing device are connected with a subtracter-adder and a first multiplier, the subtracter-adder is connected with a second multiplier and then is connected with the cosine value storing device, the first multiplier is connected with a left-shift 1-bit module and then is connected with the sine value storing device, and the a controller is connected with the sine value storing device and the cosine value storing device. The control method comprises the following steps of: mapping eta into [0, pi/2] to generate theta; judging whether theta is less than and equal to beta; if theta is less than and equal to beta, directly assigning the value and carrying out the last step; otherwise, carrying out the iteration operation, and obtaining an assigned initial value according to an approximate rule; calculating by utilizing a double-angle formula, judging whether n is less than N; if n is less than N, calculating again; otherwise, finishing the operation; finally assigning a sine/cosine function value according to the positive/negative result obtained by the sign bit judgment. The invention is applied to attitude control.

This invention provides a linear iterative multi-user test device of low complexity and a method, which avoids computation of matrix inversion by a series of circulated iteration, since the method reduces the round-off error in the computation, the computation accuracy is increased greatly at the same time when reducing computation volume and reduces noises resulted from MAI to resume multi-user information very well.