JP2629737B2 - accumulator - Google Patents

accumulator

Info

Publication number
JP2629737B2
JP2629737B2 JP24348787A JP24348787A JP2629737B2 JP 2629737 B2 JP2629737 B2 JP 2629737B2 JP 24348787 A JP24348787 A JP 24348787A JP 24348787 A JP24348787 A JP 24348787A JP 2629737 B2 JP2629737 B2 JP 2629737B2
Authority
JP
Japan
Prior art keywords
addition
supplied
output
carry
circuit
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
JP24348787A
Other languages
Japanese (ja)
Other versions
JPS6486271A (en
Inventor
孝雄 山崎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sony Corp
Original Assignee
Sony Corp
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 Sony Corp filed Critical Sony Corp
Priority to JP24348787A priority Critical patent/JP2629737B2/en
Publication of JPS6486271A publication Critical patent/JPS6486271A/en
Application granted granted Critical
Publication of JP2629737B2 publication Critical patent/JP2629737B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Landscapes

  • Complex Calculations (AREA)

Description

【発明の詳細な説明】 〔産業上の利用分野〕 本発明はアキユムレータに関し、特に長語長のデイジ
タル値の累加算に好適なものである。
Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an accumulator, and is particularly suitable for cumulative addition of digital values having a long word length.

〔発明の概要〕[Summary of the Invention]

本発明はアキユムレータに関し、供給される値を全加
算回路に供給し、この全加算回路からの加算出力と桁上
出力とをレジスタを介して帰還して累加算を行うと共
に、この加算出力と桁上出力とを桁上加算回路に入力し
て累加算出力を得ることにより、特に累加算時の演算速
度を高められるようにしたものである。
The present invention relates to an accumulator, in which a supplied value is supplied to a full adder circuit, and an addition output and a carry output from the full adder circuit are fed back via a register to perform cumulative addition. By inputting the upper output to the carry adder circuit to obtain a cumulative output, it is possible to increase the operation speed particularly at the time of cumulative addition.

〔従来の技術〕[Conventional technology]

従来のアキユムレータは例えば第3図に示すように構
成されている。図において、順次供給される値Xはnビ
ツトの桁上加算回路(7)に入力され、この加算出力が
同じくnビツトのレジスタ(8)に入力され、このレジ
スタ(8)の出力が加算回路(7)に帰還される。
A conventional accumulator is configured, for example, as shown in FIG. In the figure, a sequentially supplied value X is input to an n-bit carry adder (7), and the added output is also input to an n-bit register (8), and the output of the register (8) is added to the adder. Returned to (7).

すなわちこの装置において、レジスタ(8)に格納さ
れた前回の加算値が加算回路(7)に帰還されて新たに
供給された値Xと加算され、この加算出力が再度レジス
タ(8)に格納され、これが繰り返されることによつて
累加算が行われる。そして所望の累加算期間の開始の直
前にクリア信号CLRがレジスタ(8)に供給されて、レ
ジスタ(8)の格納内容が“0"リセツトされた時点から
累加算が行われ、累加算期間の終了時にレジスタ(8)
の格納内容を取出すことによつて、累加算値ΣXを得る
ことができる。
That is, in this device, the previous addition value stored in the register (8) is fed back to the addition circuit (7) and added to the newly supplied value X, and the added output is stored again in the register (8). This is repeated to perform cumulative addition. The clear signal CLR is supplied to the register (8) immediately before the start of the desired cumulative addition period, and the cumulative addition is performed from the time when the content stored in the register (8) is reset to "0". Register (8) at the end
By extracting the stored contents of, the accumulated value ΣX can be obtained.

〔発明が解決しようとする問題点〕[Problems to be solved by the invention]

ところが上述の装置において、桁上加算回路(31)及
びレジスタ(32)は具体的には第4図に示すように構成
される。なお図はn=4ビツトの場合を示す。
However, in the above device, the carry adder circuit (31) and the register (32) are specifically configured as shown in FIG. The figure shows the case where n = 4 bits.

図において、値XのLSB〜MSBの4ビツトがそれぞれ全
加算回路(FA)(71)〜(74)の一方の入力端Aに供給
され、この回路(71)〜(74)の加算出力端Sからの信
号がそれぞれレジスタ(R)(81)〜(84)に供給され
る。さらにレジスタ(81)〜(84)からの信号がそれぞ
れ回路(71)〜(74)の他方の入力端Bに供給されると
共に、各回路(71)〜(73)の桁上出力端COからの信号
がそれぞれ1ビツト上位の回路(72)〜(74)の桁上入
力端CIに供給される。
In the figure, four bits LSB to MSB of the value X are supplied to one input terminal A of the full adders (FA) (71) to (74), respectively, and the addition output terminals of the circuits (71) to (74) are provided. The signal from S is supplied to the registers (R) (81) to (84), respectively. Further, signals from the registers (81) to (84) are supplied to the other input terminals B of the circuits (71) to (74), respectively, and from the carry output terminals CO of the circuits (71) to (73). Are supplied to the carry input terminals CI of the circuits (72) to (74) one bit higher.

そしてこの装置において各全加算回路(71)〜(74)
の入力端A,Bに信号が供給された場合に、加算演算はま
ずLSB側の回路(71)で行われ、ここで形成された加算
出力が取出されると共に桁上出力が1ビツト上位の回路
(72)に供給されて次の回路(72)での加算演算が行わ
れる。従つて各桁上出力が順次伝播されて演算が行われ
るために、全体の演算時間が極めて多くかかつてしま
う。
In this device, each full adder circuit (71) to (74)
When signals are supplied to the input terminals A and B, the addition operation is first performed in the circuit (71) on the LSB side, the added output formed here is taken out, and the carry output is higher by one bit. The signal is supplied to the circuit (72) and the addition operation is performed in the next circuit (72). Therefore, since each carry output is sequentially propagated and the operation is performed, the entire operation time is extremely long.

このためnの値が大きくなるとこの演算による遅延時
間が処理のサイクルタイムを制限し、高速信号に対する
累加算を行うことができないなどの問題を生じていた。
For this reason, when the value of n becomes large, the delay time due to this operation limits the cycle time of the processing, and there has been a problem that the cumulative addition cannot be performed on the high-speed signal.

これに対して、いわゆる桁上先見加算回路(CLA)を
用いて演算速度を高めることもできるが、nの値が大き
くなるとCLAであつても演算速度には限界があり、またC
LAは一般的にハードウエアの量が極めて多くなつてしま
う問題点があつた。
On the other hand, the operation speed can be increased by using a so-called carry look-ahead addition circuit (CLA). However, as the value of n increases, the operation speed of the CLA is limited, and C
LA generally had a problem that the amount of hardware became extremely large.

〔問題点を解決するための手段〕[Means for solving the problem]

本発明は、順次供給される複数の値Xを累加算するア
キュムレータにおいて、上記供給される値を全加算回路
(1)に入力し、この全加算回路の加算出力と桁上出力
とをそれぞれ第1及び第2のレジスタ(2)(3)に入
力し、今回入力である上記供給される値に対して、前回
入力である上記第1及び第2のレジスタの出力を上記全
加算回路に帰還して累加算を行うと共に、上記加算出力
と桁上出力とを桁上加算回路(6)に入力して累加算出
力ΣXを得るようにしたアキュムレータである。
According to the present invention, in an accumulator for cumulatively adding a plurality of values X sequentially supplied, the supplied value is input to a full adder circuit (1), and an addition output and a carry output of the full adder circuit are respectively output to a first adder. Input to the first and second registers (2) and (3), and output the outputs of the first and second registers, which are the previous inputs, to the full adder circuit with respect to the supplied value, which is the current input. The accumulator is configured to perform cumulative addition, and to input the addition output and the carry output to the carry addition circuit (6) to obtain a cumulative addition output ΣX.

〔作用〕[Action]

これによれば、累加算期間は全加算回路を用いて加算
を行うと共に、累加算値を得るときのみ桁上加算が行わ
れるので、累加算時の演算速度が速くなり、処理のサイ
クルタイムが短くされて、高速信号に対する累加算を容
易に実現することができる。
According to this, in the cumulative addition period, addition is performed using the full addition circuit, and carry-over is performed only when the cumulative value is obtained. Therefore, the operation speed at the time of cumulative addition is increased, and the cycle time of processing is reduced. Since the length is shortened, the cumulative addition for the high-speed signal can be easily realized.

〔実施例〕〔Example〕

第1図は1ビツト分の構成を示す。 FIG. 1 shows a configuration for one bit.

この図において、順次供給される値Xの任意のビツト
が全加算回路(1)の一方の入力端Aに供給される。こ
の全加算回路(1)の加算出力端Sと桁上出力端COから
の信号がそれぞれレジスタ(2)(3)に供給される。
このレジスタ(2)からの信号が全加算回路(1)の他
方の入力端Bに供給されると共に、レジスタ(3)から
の信号は1ビツト上位に送られ、1ビツト下位からの信
号が全加算回路(1)の桁上入力端CIに供給される。さ
らにレジスタ(2)(3)からの信号がレジスタ(4)
(5)に供給され、これらのレジスタ(4)(5)から
の信号が桁上加算回路(6)に供給される。
In this figure, an arbitrary bit of a value X sequentially supplied is supplied to one input terminal A of a full adder (1). Signals from the addition output terminal S and the carry output terminal CO of the full addition circuit (1) are supplied to registers (2) and (3), respectively.
The signal from the register (2) is supplied to the other input terminal B of the full adder circuit (1), and the signal from the register (3) is sent one bit higher, and the signal from the one bit lower is fully input. It is supplied to the carry input terminal CI of the adder circuit (1). Further, the signals from the registers (2) and (3) are transmitted to the register (4).
(5), and the signals from these registers (4) and (5) are supplied to the carry adder circuit (6).

さらに第2図は例えばn=4ビツトとした場合の具体
回路を示す。この図において値XのLSB〜MSBの4ビツト
がそれぞれ全加算回路(FA)(11)〜(14)の一方の入
力端Aに供給され、この回路(11)〜(14)の加算出力
端Sからの信号がそれぞれレジスタ(R)(21)〜(2
4)に供給されると共に、回路(11)〜(13)の桁上出
力端COからの信号がそれぞれレジスタ(31)〜(33)に
供給される。そしてレジスタ(21)〜(24)からの信号
がそれぞれ回路(11)〜(14)の他方の入力端Bに供給
されると共に、レジスタ(31)〜(33)からの信号がそ
れぞれ1ビツト上位の回路(12)〜(14)の桁上入力端
CIに供給される。
FIG. 2 shows a specific circuit when n = 4 bits, for example. In this figure, four bits LSB to MSB of the value X are supplied to one input terminal A of the full adder circuits (FA) (11) to (14), respectively, and the addition output terminals of the circuits (11) to (14) are provided. The signals from S are respectively the registers (R) (21) to (2)
4) and the signals from the carry output terminals CO of the circuits (11) to (13) are supplied to the registers (31) to (33), respectively. The signals from the registers (21) to (24) are supplied to the other input terminals B of the circuits (11) to (14), respectively, and the signals from the registers (31) to (33) are each higher by one bit. Input terminals of circuits (12) to (14)
Supplied to CI.

さらにレジスタ(21)〜(24)及び(31)〜(33)か
らの信号がそれぞれレジスタ(41)〜(44)及び(51)
〜(53)に供給される。このレジスタ(42)〜(44)か
らの信号がそれぞれ全加算回路(62)〜(64)の一方の
入力端Aに供給されると共に、レジスタ(51)〜(53)
からの信号がそれぞれ1ビツト上位の回路(62)〜(6
4)の他方の入力端Bに供給され、この回路(62)(6
3)の桁上出力端COからの信号がそれぞれ1ビツト上位
の回路(63)(64)の桁上入力端CIに供給される。そし
てレジスタ(41)及び回路(62)〜(63)の加算出力端
Sから4ビツトの累加算値ΣXが取出される。
Further, the signals from the registers (21) to (24) and (31) to (33) are respectively applied to the registers (41) to (44) and (51).
To (53). The signals from the registers (42) to (44) are supplied to one input terminal A of the full adders (62) to (64), respectively, and the registers (51) to (53)
From the circuits (62)-(6)
4), and is supplied to the other input terminal B of this circuit (62) (6).
The signal from the carry output terminal CO of 3) is supplied to the carry input terminal CI of each of the circuits (63) and (64) one bit higher. Then, a 4-bit accumulated value ΣX is taken out from the addition output terminal S of the register (41) and the circuits (62) to (63).

すなわちこの装置において、所望の累加算期間の開始
の直前にクリア信号CLRがレジスタ(21)〜(24)及び
(31)〜(33)に供給されて、レジスタの格納内容が
“0"リセツトされた時点から累加算が行われる。そして
累加期間の終了時にレジスタ(41)〜(44)及び(51)
〜(53)にホールド信号HLDが供給されることによつ
て、このときのレジスタ(21)〜(24)及び(31)〜
(33)の格納内容がレジスタ(41)〜(44)及び(51)
〜(53)に保持され、この内容が全加算回路(62)〜
(64)で桁上加算されて累加算値ΣXが取出される。
That is, in this device, the clear signal CLR is supplied to the registers (21) to (24) and (31) to (33) immediately before the start of the desired cumulative addition period, and the contents stored in the register are reset to "0". The cumulative addition is performed from the point in time. Then, at the end of the cumulative period, the registers (41) to (44) and (51)
Since the hold signal HLD is supplied to the registers (21) to (24) and (31) to (53) at this time,
The contents stored in (33) are registers (41) to (44) and (51)
(53), and the contents are stored in the full adder circuit (62).
At (64), carry-up is performed to obtain the cumulative addition value ΣX.

従つてこの装置によれば、累加算期間は全加算回路を
用いて加算を行うと共に、累加算値を得るときのみ桁上
加算が行われるので、累加算時の演算速度が速くなり、
処理のサイクルタイムが短くされて、高速信号に対する
累加算を容易に実現することができる。
Therefore, according to this device, during the cumulative addition period, addition is performed using the full addition circuit, and carry-over is performed only when the cumulative addition value is obtained.
The cycle time of the processing is shortened, and the cumulative addition for the high-speed signal can be easily realized.

すなわちこの装置において、1ビツトの全加算回路の
演算時間+レジスタの遅延時間をサイクルタイムとする
アキユムレータを実現することができる。
That is, in this device, it is possible to realize an accumulator in which the cycle time is the operation time of the 1-bit full adder circuit plus the delay time of the register.

なお上述の装置において、演算のビツト数nは、オー
バーフローが生じないように充分に大きくする必要があ
るが、その場合にも長語長になることによつて演算速度
が遅くなることがない。
In the above-described apparatus, the number n of operation bits needs to be sufficiently large so as not to cause overflow. In this case, however, the operation speed does not decrease due to the long word length.

また上述の装置で累加算期間が充分に長いものであれ
ば、桁上加算回路(6)は極めて低速のものを用いるこ
とができる。従つてこのためのハードウエアは極めて少
量とすることができ、例えばCLAを用いる場合に比べて
回路規模を極めて小さくすることができる。
Also, if the accumulative addition period is sufficiently long in the above-described device, an extremely slow carry-up circuit (6) can be used. Therefore, the hardware for this can be made very small, and the circuit scale can be made extremely small as compared with the case where CLA is used, for example.

さらに上述の装置によれば、高速動作を行う全加算回
路(11)〜(14)等の回路部分に規則性が強く、VLSI等
に形成する場合に極めて有利である。
Further, according to the above-described apparatus, the circuit portions such as the full adder circuits (11) to (14) that operate at high speed have a strong regularity, and are extremely advantageous when formed in a VLSI or the like.

〔発明の効果〕〔The invention's effect〕

この発明によれば、累加算期間は全加算回路を用いて
加算を行うと共に、累加算値を得るときのみ桁上加算が
行われるので、累加算時の演算速度が速くなり、処理の
サイクルタイムが短くされて高速信号に対する累加算を
容易に実現することができるようになつた。
According to the present invention, in the cumulative addition period, addition is performed using the full addition circuit, and carry-over is performed only when a cumulative value is obtained, so that the operation speed at the time of cumulative addition is increased, and the cycle time of processing is increased. Has been shortened so that cumulative addition for high-speed signals can be easily realized.

【図面の簡単な説明】[Brief description of the drawings]

第1図は本発明の一例の構成図、第2図はその説明のた
めの図、第3図,第4図は従来の技術の説明のための図
である。 (1)は全加算回路、(2)〜(5)はレジスタ、
(6)は桁上加算回路である。
FIG. 1 is a diagram showing an example of the present invention, FIG. 2 is a diagram for explaining the same, and FIGS. 3 and 4 are diagrams for explaining a conventional technique. (1) is a full addition circuit, (2) to (5) are registers,
(6) is a carry addition circuit.

Claims (1)

(57)【特許請求の範囲】(57) [Claims] 【請求項1】順次供給される複数の値を累加算するアキ
ュムレータにおいて、 上記供給される値を全加算回路に入力し、 この全加算回路の加算出力と桁上出力とをそれぞれ第1
及び第2のレジスタに入力し、 今回入力である上記供給される値に対して、前回入力で
ある上記第1及び第2のレジスタの出力を上記全加算回
路に帰還して累加算を行うと共に、 上記加算出力と桁上出力とを桁上加算回路に入力して累
加算出力を得るようにしたアキュムレータ。
1. An accumulator for accumulating a plurality of values sequentially supplied, wherein said supplied value is inputted to a full adder circuit, and an addition output and a carry output of said full adder circuit are respectively outputted to a first adder circuit.
And to the second register, for the supplied value which is the current input, feeds back the outputs of the first and second registers which are the previous input to the full adder circuit to perform cumulative addition. An accumulator configured to input the addition output and the carry output to a carry addition circuit to obtain a cumulative addition output.
JP24348787A 1987-09-28 1987-09-28 accumulator Expired - Lifetime JP2629737B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP24348787A JP2629737B2 (en) 1987-09-28 1987-09-28 accumulator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP24348787A JP2629737B2 (en) 1987-09-28 1987-09-28 accumulator

Publications (2)

Publication Number Publication Date
JPS6486271A JPS6486271A (en) 1989-03-30
JP2629737B2 true JP2629737B2 (en) 1997-07-16

Family

ID=17104619

Family Applications (1)

Application Number Title Priority Date Filing Date
JP24348787A Expired - Lifetime JP2629737B2 (en) 1987-09-28 1987-09-28 accumulator

Country Status (1)

Country Link
JP (1) JP2629737B2 (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0391832A (en) * 1989-09-05 1991-04-17 Sony Corp Addition circuit
ES1045230Y (en) * 2000-02-14 2001-02-16 Salas Luis Jose Diaz COFFEE MONOUSO FILTER CONTAINER

Also Published As

Publication number Publication date
JPS6486271A (en) 1989-03-30

Similar Documents

Publication Publication Date Title
EP0127988B1 (en) A normalizing circuit
JPH0479013B2 (en)
CA1229172A (en) Logic adder circuit
EP0416869B1 (en) Digital adder/accumulator
US4122527A (en) Emitter coupled multiplier array
US4899305A (en) Manchester carry adder circuit
JP2629737B2 (en) accumulator
JP3210356B2 (en) Data zero judgment device
JP2643165B2 (en) Arithmetic circuit
JP2508041B2 (en) Increment circuit
JP2599984B2 (en) Input data peak value detection circuit
JP3166781B2 (en) Adder circuit
JPH04195423A (en) Multiplier
JPH0523013Y2 (en)
JPH0756713A (en) Full adder circuit
GB2179770A (en) Method and digital circuit for fixed coefficient serial multiplication
JPH0779247B2 (en) Decode circuit
JPH1021054A (en) Arithmetic processor
JPS62184534A (en) Arithmetic circuit
JPH07191830A (en) Arithmetic unit
JPH0445858B2 (en)
JPH05233217A (en) Three-input rounding addition circuit
JPH0375903B2 (en)
JPH0611104B2 (en) Pulse width modulator
JPS60241333A (en) Adding circuit of mod(2n-1)

Legal Events

Date Code Title Description
EXPY Cancellation because of completion of term
FPAY Renewal fee payment (prs date is renewal date of database)

Free format text: PAYMENT UNTIL: 20080418

Year of fee payment: 11