RU95104370A - Device which calculates scalar value of m-dimensional vector - Google Patents

Device which calculates scalar value of m-dimensional vector

Info

Publication number
RU95104370A
RU95104370A RU95104370/09A RU95104370A RU95104370A RU 95104370 A RU95104370 A RU 95104370A RU 95104370/09 A RU95104370/09 A RU 95104370/09A RU 95104370 A RU95104370 A RU 95104370A RU 95104370 A RU95104370 A RU 95104370A
Authority
RU
Russia
Prior art keywords
group
adders
subtracters
additions
sign
Prior art date
Application number
RU95104370/09A
Other languages
Russian (ru)
Other versions
RU2080650C1 (en
Inventor
Е.И. Лухнич
В.А. Егунов
Original Assignee
Е.И. Лухнич
В.А. Егунов
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Е.И. Лухнич, В.А. Егунов filed Critical Е.И. Лухнич
Priority to RU95104370A priority Critical patent/RU2080650C1/en
Publication of RU95104370A publication Critical patent/RU95104370A/en
Application granted granted Critical
Publication of RU2080650C1 publication Critical patent/RU2080650C1/en

Links

Landscapes

  • Complex Calculations (AREA)

Abstract

FIELD: computer engineering. SUBSTANCE: device has sign decoders, first group of adders-subtracters, second group of adders-subtracters, shift registers, unit which alters sign, registers, group of adders, group of shift registers. Device provides conversion of vector coordinates in m-dimensional space as single macro by means of orthogonal reflection transform without long multiplication operation. This results in possibility to increase speed by O(n) additions for each iteration and by O($$$) additions for piped processing which uses n processors which are connected in series. EFFECT: increased speed. 3 dwg

Claims (1)

Изобретение относится к вычислительной технике и предназначено для построения на его основе специальных ЭВМ. Устройство содержит дешифраторы знака, первую группу сумматоров-вычитателей, вторую группу сумматоров- вычитателей, сдвигатели, блок изменения знака числа, регистры, группу сумматоров, группу сдвигателей. В устройстве производится преобразование компонент вектора в m-мерном пространстве как одна макрооперация путем ортогональных преобразований отражения без использования "длинной" операции умножения. Это приводит к увеличению быстродействия на время порядка n сложений на одной итерации и на время порядка n2 сложений при конвейерной обработке с использованием n последовательно включенных устройств.The invention relates to computer technology and is intended to build on its basis special computers. The device comprises sign decoders, a first group of adders-subtracters, a second group of adders-subtracters, shifters, a block for changing the sign of a number, registers, a group of adders, a group of shifters. The device transforms the vector components in m-dimensional space as one macro operation by means of orthogonal reflection transformations without using the "long" multiplication operation. This leads to an increase in performance by a time of the order of n additions in one iteration and by a time of the order of n 2 additions during pipeline processing using n series-connected devices.
RU95104370A 1995-03-01 1995-03-01 Device for calculation of absolute value of m- dimensional vector RU2080650C1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
RU95104370A RU2080650C1 (en) 1995-03-01 1995-03-01 Device for calculation of absolute value of m- dimensional vector

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
RU95104370A RU2080650C1 (en) 1995-03-01 1995-03-01 Device for calculation of absolute value of m- dimensional vector

Publications (2)

Publication Number Publication Date
RU95104370A true RU95104370A (en) 1997-01-27
RU2080650C1 RU2080650C1 (en) 1997-05-27

Family

ID=20166014

Family Applications (1)

Application Number Title Priority Date Filing Date
RU95104370A RU2080650C1 (en) 1995-03-01 1995-03-01 Device for calculation of absolute value of m- dimensional vector

Country Status (1)

Country Link
RU (1) RU2080650C1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2708501C1 (en) * 2019-05-21 2019-12-09 Федеральное государственное бюджетное образовательное учреждение высшего образования "Юго-Западный государственный университет" (ЮЗГУ) Parallel-series adder-subtractor by higher positions forward on neurons

Also Published As

Publication number Publication date
RU2080650C1 (en) 1997-05-27

Similar Documents

Publication Publication Date Title
US9015354B2 (en) Efficient complex multiplication and fast fourier transform (FFT) implementation on the ManArray architecture
Bronson et al. Experimental application-driven architecture analysis of an SIMD/MIMD parallel processing system
AU2193599A (en) Pipelined fast fourier transform processor
Cheung et al. High performance computing on a cluster of workstations
WO2002091221A3 (en) Address generator for fast fourier transform processor
McAulay Optical crossbar interconnected digital signal processor with basic algorithms
RU95104370A (en) Device which calculates scalar value of m-dimensional vector
DE69113639T2 (en) REAL-TIME INPUT / OUTPUT METHOD FOR A VECTOR PROCESSOR SYSTEM.
Harvey et al. Low cost scaleable parallel image processing system
GEORGE et al. Parallel algorithms for split-aperture conventional beamforming
Pechanek et al. MFAST: a single chip highly parallel image processing architecture
Shenhav The decomposition of long FFT's for high throughput implementation
Bershader et al. The giant-Fourier-transform
Reddaway et al. Ultra-high performance, low-power, data parallel radar implementations
Calvin et al. Overlapping techniques of communications
RU94030340A (en) Scalar processor
Lu Implementation of'multiply-add'FFT algorithms for complex and real data sequences
Adams et al. Experiments with parallel fast Fourier transforms
Gudvangen et al. Computation of 2-dimensional fast transforms and image filtering on transputer networks
RU2024931C1 (en) Device for performing discrete orthogonal conversions
Marsh et al. TM-66 swiFFT DSP chip: architecture, algorithms, and applications
Sokolov et al. Speed-Independent Fused Multiply-Add Unit of Gigaflops Rating: Methodological Aspects
CN117437113A (en) System, method and storage medium for accelerating image data
Pechanek et al. TENSOR PRODUCT FFT’s ON M. fast.: A HIGHLY PARALLEL SINGLE CHIP DSP
Kung An algorithm basis for systolic/wavefront array software