JPS59202542A - Decoder circuit - Google Patents
Decoder circuitInfo
- Publication number
- JPS59202542A JPS59202542A JP58077751A JP7775183A JPS59202542A JP S59202542 A JPS59202542 A JP S59202542A JP 58077751 A JP58077751 A JP 58077751A JP 7775183 A JP7775183 A JP 7775183A JP S59202542 A JPS59202542 A JP S59202542A
- Authority
- JP
- Japan
- Prior art keywords
- decoder circuit
- output
- gate
- partial product
- inverter
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F7/38—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation
- G06F7/48—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices
- G06F7/52—Multiplying; Dividing
- G06F7/523—Multiplying only
- G06F7/533—Reduction of the number of iteration steps or stages, e.g. using the Booth algorithm, log-sum, odd-even
- G06F7/5334—Reduction of the number of iteration steps or stages, e.g. using the Booth algorithm, log-sum, odd-even by using multiple bit scanning, i.e. by decoding groups of successive multiplier bits in order to select an appropriate precalculated multiple of the multiplicand as a partial product
- G06F7/5336—Reduction of the number of iteration steps or stages, e.g. using the Booth algorithm, log-sum, odd-even by using multiple bit scanning, i.e. by decoding groups of successive multiplier bits in order to select an appropriate precalculated multiple of the multiplicand as a partial product overlapped, i.e. with successive bitgroups sharing one or more bits being recoded into signed digit representation, e.g. using the Modified Booth Algorithm
- G06F7/5338—Reduction of the number of iteration steps or stages, e.g. using the Booth algorithm, log-sum, odd-even by using multiple bit scanning, i.e. by decoding groups of successive multiplier bits in order to select an appropriate precalculated multiple of the multiplicand as a partial product overlapped, i.e. with successive bitgroups sharing one or more bits being recoded into signed digit representation, e.g. using the Modified Booth Algorithm each bitgroup having two new bits, e.g. 2nd order MBA
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
Abstract
Description
【発明の詳細な説明】
産業上の利用分野
本発明はMOSトランジスタで構成するに適したディジ
タル乗算器、特に変形Boothのアルゴリズムを用い
た並列型ディジタル乗算器のデコーダ・回路に関するも
のである○
従来例の構成とその問題点
並列型ディジタル乗算器は最近丑すます応用範囲が拡大
ちれつつあり、そのLSI化が進みつつある。LSI化
された並列型ディジタル乗算器において時に高速な乗算
速度が要求されるものに関しては、変形Boothのア
ルゴリズムに基いて構成される。この変形Bo、oth
のアルゴリズムを説明すると、Xを被乗数、Yを乗数と
し各々2の補数表示とすると乗算結果は次の如く表わす
ことができる。DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a digital multiplier suitable for configuring with MOS transistors, and in particular to a decoder/circuit for a parallel digital multiplier using a modified Booth's algorithm. Example configuration and its problems Parallel digital multipliers are increasingly being used in an increasingly wide range of applications, and are increasingly being integrated into LSIs. Parallel digital multipliers implemented in LSI, which are sometimes required to have high multiplication speeds, are constructed based on a modified Booth algorithm. This transformation Bo, oth
To explain the algorithm, if X is a multiplicand and Y is a multiplier, and each is expressed as a two's complement number, the multiplication result can be expressed as follows.
X@Y=X−(−2n−’ayn+2n−2myn’、
+・・・→−y1)ただし、yO−Oである。X@Y=X-(-2n-'ayn+2n-2myn',
+...→-y1) However, yO-O.
ここでPP、−(y2i+y2.+1−2y2□+2)
・X は部分積であり、その値は連続するYのラビット
のパターンに応じてO2±X、±2Xのいずれかの値を
とる。部分積の数はYのピット数の約半分となり、これ
に応じてこれら部分積を加算する時間も半減し高速な乗
算が可能となる。並列型ディジタル乗算器は通常、部分
積を生成し前段から送出される部分和に加算し次段に部
分和を送出するセルと、このセルにO2±X、±2Xの
いずれかの部分積を生成されるかを制御するセルとから
構成され、後者は特にBoothのデコーダ回路と呼ば
れている。Here PP, -(y2i+y2.+1-2y2□+2)
・X is a partial product, and its value takes either O2±X or ±2X depending on the continuous Y rabbit pattern. The number of partial products is approximately half the number of pits of Y, and accordingly, the time for adding these partial products is also halved, allowing high-speed multiplication. Parallel digital multipliers usually have a cell that generates a partial product, adds it to the partial sum sent from the previous stage, and sends the partial sum to the next stage, and a partial product of either O2±X or ±2X to this cell. The latter is particularly called a Booth decoder circuit.
Boothのデコーダ回路は制御出力として○、±X、
±2Xの代シに、Xl、X2.Nの3本を持ち、Xlは
Xを部分積とすることを、X2は2Xを部分積とするこ
とを、Nは部分積を負数にすることを夫々表わしておシ
、連続するYの3ビツトのパターンとは次に示す表の如
く対応している。The Booth decoder circuit outputs ○, ±X,
In place of ±2X, Xl, X2. It has three pieces of N, and Xl represents that X is a partial product, X2 represents that 2X is a partial product, and N represents that the partial product is a negative number. The bit patterns correspond as shown in the table below.
以下余白
Xl、x2.N は夫々次のような論理式て表わすこと
ができ、
X1=72i0y2i+1(■は排他的論理和)(2)
X2=y2i @y2i+1 ′″y2i+2+y2i
@V2i+1°y2i+2(3)
N=y21+2(4)
この結果Boothのデコーダ回路は、制御出力に十分
な駆動能力を与えるだめ一段のインバータを付加した場
合、第1図のようになる0同図において1〜3は連続す
るYの3ビツトの入力信号線y2□T72□+1.y2
□+2であり4〜6は夫々X1゜X2.Nに相当する制
御出力線である。Below margin XL, x2. Each of N can be expressed as the following logical formula, X1=72i0y2i+1 (■ is exclusive OR) (2)
X2=y2i @y2i+1 ′″y2i+2+y2i
@V2i+1°y2i+2(3) N=y21+2(4) As a result, Booth's decoder circuit becomes as shown in Figure 1 when one stage of inverter is added to provide sufficient driving capability for the control output. 1 to 3 are consecutive Y 3-bit input signal lines y2□T72□+1. y2
□+2, and 4 to 6 are each X1°X2. This is a control output line corresponding to N.
しかしながら、第1図の回路を通常の0MO8論理ゲー
トで構成するとトランジスタ数が34個必要であり、ま
た6人力AND−NORゲートのような速度の面で不利
な複合ゲートを含んでしまい高集積化、高速化が困難で
あった。However, if the circuit shown in Figure 1 is configured with ordinary 0MO8 logic gates, it will require 34 transistors, and it will also include composite gates such as 6-man AND-NOR gates, which are disadvantageous in terms of speed, resulting in high integration. , it was difficult to increase the speed.
発明の目的
本発明はこのような従来の問題に鑑みてなされたもので
、その目的とするところは、回路の総素子数を低減し、
高速化、高集積化を図ったBoothのデコーダ回路を
提供するものである。OBJECTS OF THE INVENTION The present invention has been made in view of such conventional problems, and its purpose is to reduce the total number of elements in a circuit,
The present invention provides a Booth decoder circuit that achieves high speed and high integration.
発明の構成
本発明は、トランスファ・ゲートとトライステート・イ
ンバータを効果的に用いることによシ、速度の遅い多入
力複合ゲートを用いることなく総素子数の低減を図った
Eoothのデコーダ回路を実現するものである。Structure of the Invention The present invention realizes an Eooth decoder circuit that reduces the total number of elements by effectively using transfer gates and tristate inverters without using slow multi-input composite gates. It is something to do.
実施例の説明
第2図は本発明の第1の実施例におけるBoothのデ
コーダ回路を示し、同図の1〜6は第1図の1〜6に同
じく、7はトランスファ・ゲート、8はトライステート
・インバータ、9ばNANDゲート、1oは○R−NA
NDゲート、11はNORゲート、12はインバータで
あり、夫々CMO8構成のゲートである。DESCRIPTION OF THE EMBODIMENTS FIG. 2 shows a Booth decoder circuit in the first embodiment of the present invention, 1 to 6 in the same figure are the same as 1 to 6 in FIG. 1, 7 is a transfer gate, and 8 is a try. State inverter, 9 is NAND gate, 1o is ○R-NA
An ND gate, 11 is a NOR gate, and 12 is an inverter, each of which has a CMO8 configuration.
(3〕式は次式の如く変形することができる。Equation (3) can be transformed as shown in the following equation.
(5)
この式の意味するところは、72i+2が高論理レベル
の時y2iとy2i+1のNOR出力を反転して出力と
し、y2i+2が低論理レベルの時y2□と72i+1
のN A N D出力をそのまま出力とし、さらにこの
出力を反転するということである。前者に対1−てはト
ライステート・インバータ8をy2i+2で活性化する
ことによって、後者に対してはトランスファ・ゲート7
をy2i+2で活性化することによって各々の出力を得
ることができる。72iとy2□や、とのNAND出力
は第1図、址たは第2図に示した如<Xl 信号を得る
過程で得ることができるので新たにN ORゲートを付
加するたけでよい。Xl。(5) What this formula means is that when 72i+2 is at a high logic level, the NOR output of y2i and y2i+1 is inverted and output, and when y2i+2 is at a low logic level, y2□ and 72i+1
This means that the N A N D output of is used as an output, and this output is further inverted. For the former pair 1-, by activating the tri-state inverter 8 at y2i+2, for the latter pair, the transfer gate 7
Each output can be obtained by activating y2i+2. Since the NAND output of 72i, y2□, etc. can be obtained in the process of obtaining the <Xl signal as shown in FIG. Xl.
N信号は従来例と同じゲート構成で得ている。The N signal is obtained with the same gate configuration as the conventional example.
以」二のように、本実施例によれば、トランスファ・ゲ
ート7とトライステート・インバータ8を効果的に用い
、y2□+2の論理レベルによって所望の出力が得られ
るように相補的に活性にすることにより、従来34個必
要であったトランジスタを28個に減することができる
。寸た従来例ではX2信号を得るのに6人力の複合ゲー
トを用いているが、一般にこのような多入力複合ゲート
は、配線容量やドレイン及びソース領域の拡散容量が大
きく、負荷の駆動も多数のトランジスタを介して行なわ
れるため速度が遅い。また第1図の6人力後合ゲートに
おいてはR悪の場合、トランジスタ3個を介して出力負
荷を駆動しなければならない。これに対して本夫施例で
X2信号を得る構成は、最悪の場合でも2個のトランジ
スタを介して出力負荷を駆動しており、また配線容量や
拡散容量も従来例よりも小さくすることができ、従来例
よりも平均的な動作速度は早い。実際にインバータの遅
延時間を1・△とすると、同じトランジスタで構成した
従来例の6人力後合ゲートは3△〜4△程度要するのに
対し、トライステート・インバータば2△〜2.6△、
トランスファ・スイッチは0.5△〜1△程度の遅延で
ある。As described above, according to this embodiment, the transfer gate 7 and the tristate inverter 8 are effectively used and activated in a complementary manner to obtain the desired output depending on the logic level of y2□+2. By doing so, it is possible to reduce the number of transistors that were conventionally required from 34 to 28. In the conventional example, a six-person composite gate is used to obtain the X2 signal, but such multi-input composite gates generally have large wiring capacitances, large diffusion capacitances in the drain and source regions, and require many loads to be driven. The speed is slow because it is carried out through transistors. In addition, in the case of R-bad in the six-man power combination gate shown in FIG. 1, the output load must be driven through three transistors. On the other hand, in the configuration in which the X2 signal is obtained in the present embodiment, the output load is driven through two transistors even in the worst case, and the wiring capacitance and diffusion capacitance can also be made smaller than in the conventional example. The average operating speed is faster than that of the conventional example. If the delay time of an inverter is actually 1.△, a conventional 6-man power gate using the same transistors would require about 3△ to 4△, whereas a tri-state inverter would require 2△ to 2.6△. ,
The transfer switch has a delay of about 0.5△ to 1△.
本実施例においては、Xl 信号はゲート9.10より
得られるY2iとy214−1のexctusive
NOR出力をインバータ12によって反転し7て得てい
るが、NORゲートとA i”J D −N ORグー
5トを用いてY2iとy2i+1のexctus iv
e OR出力をXl と1〜で出力してもよく、この場
合は、実施例でN0ftゲートを新たに付加する代りに
NANDゲートを新たに付加することによって同様の構
成からX2信号を得ることができる。In this embodiment, the Xl signal is the exclusive signal of Y2i and y214-1 obtained from gate 9.10.
The NOR output is obtained by inverting it with the inverter 12, but the exact of Y2i and y2i+1 is obtained using the NOR gate and A
e The OR output may be output as Xl and 1~, and in this case, the X2 signal can be obtained from the same configuration by adding a new NAND gate instead of adding a new N0ft gate in the embodiment. can.
また、本実施例においては、制御出力線4〜6に十分な
駆動能力を与えるために1段のインバータ12を付加し
た例であるが、このインバータを付加しない場合、夫々
X1.X2.N信号の反転出力X1.X2.Nが得られ
ることも明らかである。Further, in this embodiment, one stage of inverter 12 is added in order to provide sufficient driving capacity to the control output lines 4 to 6, but if this inverter is not added, each X1. X2. Inverted output of N signal X1. X2. It is also clear that N can be obtained.
さらに入力信号y2□、 y2□+1.y2□+2の代
りに各々の反転信号y2□、y2□+1,2□+2が入
力される場合の構成についても本実施例から容易に考案
されることは明らかである。Furthermore, input signals y2□, y2□+1. It is clear that a configuration in which the respective inverted signals y2□, y2□+1, and 2□+2 are input instead of y2□+2 can be easily devised from this embodiment.
発明の効果
以上のように、本発明はトランスファ・ゲートとトライ
ステート・インバータを相補的に活性にして夫々N A
N D出力を正転し、NOR出力を反転して出力する
ことにより、構成トランジスタ数を低減し、信号の伝搬
も高速にすることができる優れたBoothのデコーダ
回路を実現できるものである。Effects of the Invention As described above, the present invention activates the transfer gate and the tri-state inverter in a complementary manner so that each N A
By inverting the ND output and inverting the NOR output, it is possible to realize an excellent Booth decoder circuit that can reduce the number of constituent transistors and increase signal propagation speed.
第1図は従来のBoothのデコーダ回路図、第2図は
本発明の一実施例におけるBoothのデコーダ回Ii
!8図である。
1〜3・・・・・連続するYの3ピツトの入力信号線、
4〜6・・・・・・Xl、X2.Nに相当する制御出力
線、7・・・・・トランスノァ・ゲート、8・・・・・
・トライステート・インバータ。FIG. 1 is a conventional Booth decoder circuit diagram, and FIG. 2 is a Booth decoder circuit Ii in an embodiment of the present invention.
! This is Figure 8. 1 to 3...Continuous Y 3-pit input signal line,
4-6...Xl, X2. Control output line corresponding to N, 7... Transnor gate, 8...
- Tri-state inverter.
Claims (1)
前記第1.第2の入力信号を入力とするNORゲートと
、前記NANDゲートの出力を入力とするトランスファ
・ゲートと、前記NORゲートの出力を入力とし、出力
が前記トランスファ・ゲートの出力と共通接続されたト
ライステート・インバ〜りとを具備し、前記トランスフ
ァ・ゲートと前記トライステート・インバータとが、第
3の入力信号の論理レベルに応じて相補的に活性となる
ように構成して、前記トランスファ・ゲートの出力に出
力信号を得ることを特徴とするデコーダ回路。1st. a NAND gate that receives a second input signal;
Said 1st. a NOR gate that receives a second input signal as an input; a transfer gate that receives the output of the NAND gate as an input; a state inverter, the transfer gate and the tristate inverter are configured to be activated in a complementary manner according to a logic level of a third input signal, and the transfer gate A decoder circuit characterized in that an output signal is obtained at the output of the decoder circuit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP58077751A JPS59202542A (en) | 1983-05-02 | 1983-05-02 | Decoder circuit |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP58077751A JPS59202542A (en) | 1983-05-02 | 1983-05-02 | Decoder circuit |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS59202542A true JPS59202542A (en) | 1984-11-16 |
Family
ID=13642624
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP58077751A Pending JPS59202542A (en) | 1983-05-02 | 1983-05-02 | Decoder circuit |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS59202542A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62229439A (en) * | 1986-03-31 | 1987-10-08 | Toshiba Corp | Parallel multiplier |
JPS62293343A (en) * | 1986-06-11 | 1987-12-19 | Toshiba Corp | Booth converting circuit |
JPH03176734A (en) * | 1989-12-05 | 1991-07-31 | Sharp Corp | Encoder for parallel type multiplier |
-
1983
- 1983-05-02 JP JP58077751A patent/JPS59202542A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS62229439A (en) * | 1986-03-31 | 1987-10-08 | Toshiba Corp | Parallel multiplier |
JPH0431412B2 (en) * | 1986-03-31 | 1992-05-26 | ||
JPS62293343A (en) * | 1986-06-11 | 1987-12-19 | Toshiba Corp | Booth converting circuit |
JPH0448254B2 (en) * | 1986-06-11 | 1992-08-06 | Tokyo Shibaura Electric Co | |
JPH03176734A (en) * | 1989-12-05 | 1991-07-31 | Sharp Corp | Encoder for parallel type multiplier |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5151875A (en) | MOS array multiplier cell | |
Quach et al. | High-speed addition in CMOS | |
Mahmoud et al. | A 10-transistor low-power high-speed full adder cell | |
US5841300A (en) | Semiconductor integrated circuit apparatus | |
JPH0215088B2 (en) | ||
Morgenshtein et al. | Gate-diffusion input (gdi)-a novel power efficient method for digital circuits: a design methodology | |
JPS6014321A (en) | Multi-stage shift apparatus | |
JPS6359171B2 (en) | ||
US4831578A (en) | Binary adder | |
US5568069A (en) | High speed, low power pipelined logic circuit | |
JPS63102510A (en) | Composite circuit constituting exclusive or gate and/or exclusive nor gate | |
JPS6224815B2 (en) | ||
Lee et al. | Application of dynamic pass-transistor logic to an 8-bit multiplier | |
US5732008A (en) | Low-power high performance adder | |
JPS59202542A (en) | Decoder circuit | |
JPH0225537B2 (en) | ||
Zimmermann et al. | Low-power logic styles: CMOS vs CPL | |
Jullien et al. | Dynamic computational blocks for bit-level systolic arrays | |
US6970017B2 (en) | Logic circuit | |
Roberts et al. | Design and Analysis of Improved Low Power and High-Speed N-Bit Adder | |
JPS6055439A (en) | Decoder circuit | |
JP2002312160A (en) | Binary carry arithmetic circuit, half addition circuit and incrementer using the same, binary borrow arithmetic circuit, half subtraction circuit and decrementer using the same | |
JPS59123930A (en) | Carry signal generator | |
Lai et al. | Radix-4 adder design with refined carry | |
EP0447254A2 (en) | Array multiplier |