Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
JPH0748206B2 - Integrated circuit device - Google Patents
[go: Go Back, main page]

JPH0748206B2 - Integrated circuit device - Google Patents

Integrated circuit device

Info

Publication number
JPH0748206B2
JPH0748206B2 JP1052904A JP5290489A JPH0748206B2 JP H0748206 B2 JPH0748206 B2 JP H0748206B2 JP 1052904 A JP1052904 A JP 1052904A JP 5290489 A JP5290489 A JP 5290489A JP H0748206 B2 JPH0748206 B2 JP H0748206B2
Authority
JP
Japan
Prior art keywords
storage
storage means
elements
matrix
data
Prior art date
Legal status (The legal status 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 status listed.)
Expired - Lifetime
Application number
JP1052904A
Other languages
Japanese (ja)
Other versions
JPH02232765A (en
Inventor
雅夫 中屋
秀樹 安藤
Original Assignee
工業技術院長
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 工業技術院長 filed Critical 工業技術院長
Priority to JP1052904A priority Critical patent/JPH0748206B2/en
Publication of JPH02232765A publication Critical patent/JPH02232765A/en
Publication of JPH0748206B2 publication Critical patent/JPH0748206B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Landscapes

  • Complex Calculations (AREA)
  • Semiconductor Integrated Circuits (AREA)

Description

【発明の詳細な説明】 〔産業上の利用分野〕 この発明は正方行列の転置行列を容易に得るための集積
回路装置に関するものである。
Description: TECHNICAL FIELD The present invention relates to an integrated circuit device for easily obtaining a transposed matrix of a square matrix.

〔従来の技術〕[Conventional technology]

従来、n×nの行列に対し、その転置行列を得るために
汎用のメモリを用いた場合は、n2回のメモリに対するリ
ード/ライトが必要であり、行列の要素の増大とともに
その処理時間が非常に長くなってしまうという問題点が
あった。
Conventionally, when a general-purpose memory is used to obtain a transposed matrix for an n × n matrix, it is necessary to read / write the memory n 2 times, and the processing time increases as the number of matrix elements increases. There was a problem that it would be very long.

即ち、これはある行列に対し、汎用のメモリを用いて転
置行列を生成する場合には、該行列の各要素を一時スト
ア用のメモリに待避させ、順番を入れ換え、再びメモリ
に書き込むという作業を行うためである。
That is, for a certain matrix, when a transposed matrix is generated using a general-purpose memory, the work of saving each element of the matrix in the memory for temporary store, changing the order, and writing again in the memory This is to do it.

〔発明が解決しようとする課題〕[Problems to be Solved by the Invention]

以上のように従来の汎用のメモリを用いて転置行列を生
成する場合、その作業が煩雑であるため、行列要素が多
くなるとその処理時間が長くなってしまうという問題点
があった。
As described above, when the transposed matrix is generated using the conventional general-purpose memory, the work is complicated, and therefore, there is a problem that the processing time becomes longer as the number of matrix elements increases.

この発明は上記のような問題点を解消するためになされ
たもので、転置行列を容易に短時間で生成することがで
きる集積回路装置を得ることを目的とする。
The present invention has been made to solve the above problems, and an object thereof is to obtain an integrated circuit device that can easily generate a transposed matrix in a short time.

〔課題を解決するための手段〕[Means for Solving the Problems]

この発明に係る集積回路装置は、n行n列(n:自然数)
の二次元に行列配列されたデータの各要素を記憶する複
数の記憶素子からなり、行方向,列方向ともデータのシ
フトが可能な構造を有する第1の記憶手段と、複数の記
憶素子が上記第1の記憶手段の複数の記憶素子と同じ行
列状態に配列されてなる第2の記憶手段と、上記第1の
記憶手段と上記第2の記憶手段の同一の配列位置にある
記憶素子間をつなぐデータ転送線と、上記データ転送線
による上記第1の記憶手段の記憶素子から上記第2の記
憶手段の記憶素子へのデータの転送を制御する転送制御
手段とを備えたものである。
The integrated circuit device according to the present invention has n rows and n columns (n: natural number).
A first storage unit having a structure capable of shifting data in both the row direction and the column direction, and a plurality of storage elements. The second storage means arranged in the same matrix as the plurality of storage elements of the first storage means and the storage elements at the same array position of the first storage means and the second storage means are arranged. A data transfer line to be connected, and a transfer control means for controlling the transfer of data from the storage element of the first storage means to the storage element of the second storage means by the data transfer line.

〔作用〕[Action]

この発明においては、上記第1の記憶手段においてデー
タの行シフトと列シフトを順次行った後、上記第1の記
憶手段と第2の記憶手段間におけるデータの転送を行う
動作サイクルを繰り返し実行することにより、転置行列
を容易に生成することができる。
In the present invention, after the row shift and the column shift of the data are sequentially performed in the first storage means, the operation cycle for transferring the data between the first storage means and the second storage means is repeatedly executed. Thus, the transposed matrix can be easily generated.

〔実施例〕〔Example〕

第1図はこの発明の一実施例による、4行4列の正方行
列の転置行列を生成する半導体集積回路装置を示す。
FIG. 1 shows a semiconductor integrated circuit device for generating a transposed matrix of a square matrix of 4 rows and 4 columns according to an embodiment of the present invention.

今、行列〔A〕が と表わされるとすると、転置行列〔A〕Tとなる。本実施例は1)から2)を求めるものであり、
そのハードウェア構成は以下の通りである。
Now the matrix [A] And the transposed matrix [A] T is Becomes In this embodiment, 1) to 2) are obtained,
The hardware configuration is as follows.

第1図において、1は〔A〕の各要素を記憶する部分で
ある第1面、2は転置行列の結果を記憶する部分である
第2面である。3は第1面の各要素a1 ijと第2面の各要
素a2 ijをつなぐデータ転送線であり、4はデータ転送線
3によるデータの転送を制御するスイッチである。
In FIG. 1, 1 is a first surface that is a portion that stores each element of [A], and 2 is a second surface that is a portion that stores a result of a transposed matrix. Reference numeral 3 is a data transfer line connecting each element a 1 ij on the first surface and each element a 2 ij on the second surface, and reference numeral 4 is a switch for controlling data transfer by the data transfer line 3.

次に、本実施例の動作について説明する。 Next, the operation of this embodiment will be described.

1.最初の状態において、第1面1の要素のうち上記3)
式において丸印をつけた要素のみスイッチ3を“オン”
として第2面2へデータを転送し、他は“オフ”として
データを転送しない。
1. In the initial state, the above 3) among the elements on the 1st surface 1)
Only the elements marked with a circle in the formula turn on the switch 3.
, The data is transferred to the second surface 2, and the others are set to “off” and the data is not transferred.

2.次に第1面の要素を行方向へシフトし次式の行列を得
る。
2. Next, shift the elements on the first surface in the row direction to obtain the matrix of the following equation.

3.次にこれを列方向へシフトしさらに丸印をつけた要素
のみ第2面2へデータを転送し、次式の行列を得る。
3. Next, this is shifted in the column direction, and only the elements marked with circles are transferred to the second surface 2 to obtain the matrix of the following equation.

4.さらに行方向シフト,列方向シフトを行い所要の要素
を転送し、次式の行列を得る。
4. Perform row-wise shift and column-wise shift to transfer the required elements, and obtain the matrix of the following equation.

5.行方向シフト,列方向シフトをさらに繰り返し、所要
の要素を転送し、次式の行列を得る。
5. Repeat row-wise shift and column-wise shift to transfer the required elements and obtain the matrix of the following equation.

以上のように、行シフト,列シフト,及び転送を繰り返
すことにより、転置行列が得られる。
As described above, the transposed matrix is obtained by repeating the row shift, the column shift, and the transfer.

第2図は上記実施例装置の第1記憶面1または第2記憶
面2の記憶素子の一部を示す図であり、図中21はインバ
ータ、22は各記憶面内での行列要素の転送のためのスイ
ッチ、φx1,φx2,φy1,φy2は各面内での転送のための
クロック、23は2つのインバータ21と2つのスイッチ22
とからなる一要素分の記憶素子である。
FIG. 2 is a diagram showing a part of the storage elements of the first storage surface 1 or the second storage surface 2 of the apparatus of the above embodiment, in which 21 is an inverter and 22 is transfer of matrix elements in each storage surface. , Switches for φx1, φx2, φy1, φy2 are clocks for in-plane transfer, 23 is two inverters 21 and two switches 22
It is a memory element for one element consisting of.

なお、第1面,第2面の物理的配置については特に限定
されるものではないが、活性層(トランジスタ)を積層
した三次元構造,即ち三次元回路素子を用い、第1図に
示されるように、第1面の下に第2面を配置することに
より、各要素間を接続する線を短くでき、装置の小型化
を図ることができる。
The physical arrangement of the first surface and the second surface is not particularly limited, but a three-dimensional structure in which active layers (transistors) are laminated, that is, a three-dimensional circuit element is used and is shown in FIG. By arranging the second surface under the first surface as described above, the line connecting the elements can be shortened, and the device can be downsized.

〔発明の効果〕 以上のようにこの発明にかかる集積回路装置によれば、
n行n列(n:自然数)の二次元に行列配列されたデータ
の各要素を記憶する複数の記憶素子からなり、行方向,
列方向ともデータのシフトが可能な構造を有する第1の
記憶手段と、複数の記憶素子が上記第1の記憶手段の複
数の記憶素子と同じ行列状態に配列されてなる第2の記
憶手段と、上記第1の記憶手段と上記第2の記憶手段の
同一の配列位置にある記憶素子間をつなぐデータ転送線
と、上記データ転送線による上記第1の記憶手段の記憶
素子から上記第2の記憶手段の記憶素子へのデータの転
送を制御する転送制御手段とを設けたので、転置行列を
生成する処理速度を従来に比して大幅に向上することが
でき、短時間で転置行列を生成することができる効果が
ある。
As described above, according to the integrated circuit device of the present invention,
It is composed of a plurality of storage elements for storing each element of data arranged in a two-dimensional matrix of n rows and n columns (n: natural number), in the row direction,
First storage means having a structure capable of shifting data in the column direction, and second storage means having a plurality of storage elements arranged in the same matrix as the plurality of storage elements of the first storage means A data transfer line connecting between the storage elements of the first storage means and the storage elements of the second storage means at the same array position, and the storage element of the first storage means by the data transfer line to the second storage means. Since the transfer control means for controlling the transfer of data to the storage element of the storage means is provided, the processing speed for generating the transposed matrix can be significantly improved compared to the conventional one, and the transposed matrix can be generated in a short time. There is an effect that can be.

【図面の簡単な説明】 第1図は本発明の一実施例による集積回路装置を示す斜
視図、第2図は上記実施例の第1記憶面または第2記憶
面の記憶素子の一部を示す図である。 図において、1は第1記憶面、2は第2記憶面、11,21
は行列要素、3は両記憶面の要素間をつなぐデータ転送
線、4はデータの転送を制御するスイッチ(転送制御手
段)、11はインバータ、12はスイッチ、13は一要素分の
記憶素子である。
BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a perspective view showing an integrated circuit device according to an embodiment of the present invention, and FIG. 2 shows a part of a memory element on a first memory surface or a second memory surface of the above embodiment. FIG. In the figure, 1 is a first storage surface, 2 is a second storage surface, 11, 21
Is a matrix element, 3 is a data transfer line connecting elements on both storage surfaces, 4 is a switch (transfer control means) for controlling data transfer, 11 is an inverter, 12 is a switch, and 13 is a storage element for one element. is there.

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】n行n列(n:自然数)の二次元に行列配列
されたデータの各要素を記憶する複数の記憶素子からな
り、行方向,列方向ともデータのシフトが可能な構造を
有する第1の記憶手段と、 複数の記憶素子が上記第1の記憶手段の複数の記憶素子
と同じ行列状態に配列されてなる第2の記憶手段と、 上記第1の記憶手段と上記第2の記憶手段の同一の配列
位置にある記憶素子間をつなぐデータ転送線と、 上記データ転送線による上記第1の記憶手段の記憶素子
から上記第2の記憶手段の記憶素子へのデータの転送を
制御する転送制御手段とを備えたことを特徴とする集積
回路装置。
1. A structure comprising a plurality of storage elements for storing respective elements of data arranged in a two-dimensional matrix of n rows and n columns (n: natural number), and having a structure capable of shifting data in both the row and column directions. A first storage means having the same; a second storage means having a plurality of storage elements arranged in the same matrix state as the plurality of storage elements of the first storage means; the first storage means and the second storage means. And a data transfer line connecting the storage elements at the same array position of the storage means, and the transfer of data from the storage element of the first storage means to the storage element of the second storage means by the data transfer line. An integrated circuit device comprising: transfer control means for controlling.
JP1052904A 1989-03-07 1989-03-07 Integrated circuit device Expired - Lifetime JPH0748206B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP1052904A JPH0748206B2 (en) 1989-03-07 1989-03-07 Integrated circuit device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP1052904A JPH0748206B2 (en) 1989-03-07 1989-03-07 Integrated circuit device

Publications (2)

Publication Number Publication Date
JPH02232765A JPH02232765A (en) 1990-09-14
JPH0748206B2 true JPH0748206B2 (en) 1995-05-24

Family

ID=12927831

Family Applications (1)

Application Number Title Priority Date Filing Date
JP1052904A Expired - Lifetime JPH0748206B2 (en) 1989-03-07 1989-03-07 Integrated circuit device

Country Status (1)

Country Link
JP (1) JPH0748206B2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108145865B (en) * 2017-11-16 2019-11-15 华中科技大学 Chamfering processing method and corresponding device

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6074053A (en) * 1983-09-30 1985-04-26 Fujitsu Ltd Array memory
JPS613450A (en) * 1984-06-18 1986-01-09 Hiroshima Daigaku Shared memory integrated device of three-dimensional photo coupling
JPS62267168A (en) * 1986-05-15 1987-11-19 Fuji Xerox Co Ltd Apparatus for transposition processing of matrix data

Also Published As

Publication number Publication date
JPH02232765A (en) 1990-09-14

Similar Documents

Publication Publication Date Title
EP0122048A2 (en) Data processing cells and parallel data processors incorporating such cells
JP2003308224A (en) Multi-port scan chain register apparatus and method
JPH01195554A (en) Serial access memory device
JPH0748206B2 (en) Integrated circuit device
JPS6250870B2 (en)
JP3124852B2 (en) Data transfer circuit and address array
JP2824976B2 (en) 2D array data rotation device
JPH02148237A (en) Method of storing and operating block of data
JPH05204744A (en) Two-dimensional array data access device
JPH04188243A (en) Storage device
JP2695266B2 (en) Arbitrary image rotation method and device
JPS63293787A (en) Semiconductor memory device
JP2855899B2 (en) Function memory
JP2580501B2 (en) Parallel data processing device
JPH11103028A5 (en)
WO2007099581A1 (en) Method manufacturing of semiconductor device, and semiconductor memory device
JPH0236008B2 (en)
JPS62151987A (en) Multi-port memory for image processing
JP2642970B2 (en) Shift register circuit
CN121999825A (en) Writing/reading methods for magnetic storage cells, peripheral devices, and magnetic storage arrays
SU374598A1 (en) FUNCTIONAL CONVERTER TO DETERMINE THE DISTANCE BETWEEN POLES OF TWO-DIMENSIONAL GEOMETRIC FIGURES
JPH061449B2 (en) Image memory for image editing
JPS58185089A (en) semiconductor equipment
KR950009773B1 (en) Method of arranging unit circuit in a mesh type parallel computer
JPH04206092A (en) Serial access memory

Legal Events

Date Code Title Description
EXPY Cancellation because of completion of term