TW419891B - Asynchronous sensing differential logic (ASDL) circuit - Google Patents

Asynchronous sensing differential logic (ASDL) circuit Download PDF

Info

Publication number
TW419891B
TW419891B TW088109736A TW88109736A TW419891B TW 419891 B TW419891 B TW 419891B TW 088109736 A TW088109736 A TW 088109736A TW 88109736 A TW88109736 A TW 88109736A TW 419891 B TW419891 B TW 419891B
Authority
TW
Taiwan
Prior art keywords
output
signal
enable signal
input
transistor
Prior art date
Application number
TW088109736A
Other languages
Chinese (zh)
Inventor
Bai-Sun Kong
Original Assignee
Lg Semicon Co Ltd
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 Lg Semicon Co Ltd filed Critical Lg Semicon Co Ltd
Application granted granted Critical
Publication of TW419891B publication Critical patent/TW419891B/en

Links

Classifications

    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03KPULSE TECHNIQUE
    • H03K19/00Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits
    • H03K19/02Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components
    • H03K19/08Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices
    • H03K19/094Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices using field-effect transistors
    • H03K19/0944Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices using field-effect transistors using MOSFET or insulated gate field-effect transistors, i.e. IGFET
    • H03K19/0948Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using semiconductor devices using field-effect transistors using MOSFET or insulated gate field-effect transistors, i.e. IGFET using CMOS or complementary insulated gate field-effect transistors
    • HELECTRICITY
    • H03ELECTRONIC CIRCUITRY
    • H03KPULSE TECHNIQUE
    • H03K19/00Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits
    • H03K19/02Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components
    • H03K19/173Logic circuits, i.e. having at least two inputs acting on one output; Inverting circuits using specified components using elementary logic circuits as components
    • H03K19/1733Controllable logic circuits
    • H03K19/1738Controllable logic circuits using cascode switch logic [CSL] or cascode emitter coupled logic [CECL]

Abstract

An asynchronous sensing differential logic circuit using a charge-recycling technique includes a control block carrying out a logical operation on a request signal from a preceding stage and a request signal for a succeeding stage, and outputting a first or second input enable signal and a first or second clock signal, a functional block carrying out an operation on an input data according to the first or second input enable signals and the first or second clock signals from the control block, and outputting a first or second output enable signal and an output data, and a latch block triggered by an acknowledge signal from the succeeding stage, and outputting a request signal for the succeeding stage and a final output data by carrying out an operation on the first or second output enable signals and the output data from the functional block.

Description

經濟部智慧財產局員工消費合作社印製 A7 -419.89j__B7_____ 五、發明說明(1-) 發明背景 1. 發明領域 本發明係關於一種非同步電路,特別係關於採用_種 電荷循環技術的非同步感測差動邏輯(A s D L)電路。 2. 背景技術之說明 一般而言,半導體電路設計時,非同步設計技術之優 點為因未採用通用時脈信號,故不會產生時脈歪斜及不會 招致時脈信號分佈的額外開銷,由於信號過渡僅於有事件 時產生故可減少功率消耗,及因處理時間並非由最惡劣情 況延遲決定而係由平均延遲決定故可減少等候時間。 為了採用非同步信號,區域功能方塊需要順利通訊。 此處主要使用2相信號交換方案或4相信號交換方案,特別 ’4相信號交換方案容易與電路實施因此被普遍採用。 第1圖為示意方塊圖說明習知非同步管線配置,包括 一功能方塊102執行各種邏輯作業,一完成偵測器1〇3通訊 功能方塊102之作業完成,一控制方塊101控制功能方塊102 之信號交換,及一閂鎖方塊104根據功能方塊102之作業輸 出資料。 一差動串接電壓開關(DCVS)邏輯或一電荷循環差動 邏輯(CRDL)其屬於容易進行得自前一階段完成偵測的差 動邏輯系列,主要係用作功能方塊1 。其電路圖分別說 明於第2及3圖。 第2圖為示意電路圖說明一差動串接電壓開關(DCVS) 邏輯電路。如該圖所示,一時脈信號CK外加至二PMOS電 本紙張Κ度適用中囷國家標準(CNS)A4規格(210 X 297公釐) : ·---------- 裝--------訂---------線 (請先^*讀背£-之注音'?事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 419891 B; 五、發明說明(2 ) 晶體PM1 ' PM2的閘體,其源極被外加電源電壓Vdd。時 脈信號CK外加至NMOS電晶體NM1之閘體,NM1之源極 係接地。PMOS電晶體PM1、PM2之汲極透過一串接邏輯 102-1共通連結至NMOS電晶體NM1之汲極,串接邏輯係 遵照輸入資料而被開/關。PMOS電晶體PM1、PM2及串接 邏輯102-1之輸出汲極端子各自分別連結至二反相器XI、 X2之輸入端子。反相器XI、X2輸出個別輸出信號OUT ' ~~OUT ο 第3圖為示意電路圖說明電荷循環差動邏輯(CRDL)電 路。如該圖所示,致能信號Ei外加至NMOS電晶體NM15 之閘體,NM15之源極接地。時脈信號CK連結至PMOS電 晶體PM14之閘體,PM14之源極被外加電源電壓Vdd。其 閘體被外加時脈信號CK之PMOS電晶體PM13,其介於 PMOS電晶體PM11、PM12之閘體間的通道連結,PM11、 PM12個別之源極被外加電源電壓Vdd。PMOS電晶體PM11 、PM13之閘體分別連結至NMOS電晶體NM11 ' NM12之 閘體、NMOS電晶體ΝΜΠ、NM14之汲極,及連結至導通 電晶體邏輯電路1〇2_2之輸出端子’因此由其共通節點輸 出信號OUT、一。NMOS電晶體NM11、NM13之源極 共通連結至NMOS電晶體NM15之汲極。NMOS電晶體 NM12、NM14之源極共通連結至PMOS電晶體PM14之汲極 ,反相器X3之輸入端子輸出次一階段之致能信號Eo。 導通電晶體邏輯102·2可根據資料輸入信號DATAIN將 輸出信號OUT、—out接地。 本紙張尺度適用中國國家標準(CNS)A4規格(_2Ι0 X 297公釐) --.-------i---ί' ^--------^--------- (請先^讀背^-之江意事項箅填^本頁) 經濟部智慧財產局員工消費合作社印製 419891 a? _________B7__ 五、發明說明(3 ) 它方面’密勒C元件經常被採用作為非同步系統之信 號交換電路,原因為其具有一延遲不敏感性質。當二輸入 值相等時,輸出值係與輸入值相同。若二輸入值不同,則 進行保留前一值的作業。 第1圖配置之控制方塊101主要係使用密勒C元件作信 號交換控制實施》 此外’閂鎖方塊104可為傳統流-閂鎖型或密勒c元件 。但流-閂鎖不具有延遲不敏感性質’因此大半採用密勒c 元件。 如第4圖之舉例說明,閂鎖方塊104中,來自次一階段 的確認信號ACK共通外加至PMOS電晶體PM22之閘體, PM22之源極被外加電源電壓vdd,以及共通外加至NMOS 電晶體NM22之閘體其源極接地,以及輸入信號(Din,主 要為OUT或 Ot/71)共通外加至PMOS電晶體PM21之閘體 ’ PM21之源極係連結至PMOS電晶體PM22之汲極,以及 外加之NMOS電晶體NM21之閘體,NM21之源極係連結至 NMOS電晶體NM22之汲極。閂鎖方塊104包括二電路附有 —閂鎖104-1 ’閂鎖104-1係由二反相器X4、X5反相並聯 連結組成,俾便閂鎖由PMOS電晶體PM21與NMOS電晶體Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 -419.89j__B7 _____ V. Description of the Invention (1-) Background of the Invention 1. Field of the Invention The present invention relates to a non-synchronous circuit, and in particular relates to a non-synchronous sense using _ charge cycle technology. Measuring differential logic (A s DL) circuit. 2. Description of background technology Generally speaking, when designing semiconductor circuits, the advantage of non-synchronous design technology is that because the universal clock signal is not used, clock skew will not occur and the overhead of clock signal distribution will not be incurred. Signal transitions are generated only when there is an event, which can reduce power consumption, and because processing time is not determined by the worst-case delay, but by average delay, waiting time can be reduced. In order to use asynchronous signals, the regional function blocks need to communicate smoothly. The two-phase signal exchange scheme or the four-phase signal exchange scheme is mainly used here. In particular, the four-phase signal exchange scheme is easy to implement with a circuit and is therefore generally used. Figure 1 is a schematic block diagram illustrating a conventional asynchronous pipeline configuration, including a function block 102 for performing various logical operations, a completion of the operation of the detector 103 communication function block 102, and a control block 101 for controlling the function block 102. Handshake and a latch block 104 output data according to the operation of the function block 102. A differential series voltage switch (DCVS) logic or a charge cycle differential logic (CRDL) belongs to the differential logic series which is easy to detect from the previous stage and is mainly used as a function block 1. The circuit diagrams are shown in Figures 2 and 3, respectively. Figure 2 is a schematic circuit diagram illustrating a differential series voltage switch (DCVS) logic circuit. As shown in the figure, the one-clock signal CK is applied to two PMOS electronic papers, and the K-degree is applicable to the China National Standard (CNS) A4 specification (210 X 297 mm): · ----------- ------- Order --------- line (please ^ * read back the phonetic of £-'before filling out this page) Printed on 419891 B by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs; 5. Description of the invention (2) The source of the gate of the crystal PM1 'PM2 is the source voltage Vdd. The clock signal CK is applied to the gate of the NMOS transistor NM1, and the source of the NM1 is grounded. The drains of the PMOS transistors PM1 and PM2 are connected to the drain of the NMOS transistor NM1 in common through a series of logic 102-1. The series logic is turned on / off according to the input data. The PMOS transistors PM1, PM2 and the output drain terminals of the serial logic 102-1 are respectively connected to the input terminals of the two inverters XI and X2. The inverters XI and X2 output individual output signals OUT '~~ OUT ο Figure 3 is a schematic circuit diagram illustrating a charge loop differential logic (CRDL) circuit. As shown in the figure, the enable signal Ei is applied to the gate of the NMOS transistor NM15, and the source of the NM15 is grounded. The clock signal CK is connected to the gate of the PMOS transistor PM14, and the source of the PM14 is applied with a power supply voltage Vdd. The gate is connected with a PMOS transistor PM13 with a clock signal CK, which is connected between the gates of the PMOS transistors PM11 and PM12. The individual source of PM11 and PM12 is applied with a power supply voltage Vdd. The gates of the PMOS transistors PM11 and PM13 are respectively connected to the gates of the NMOS transistor NM11 'NM12, the drains of the NMOS transistors NMΠ and NM14, and the output terminals connected to the conducting crystal logic circuit 10-2_2. The common node outputs signals OUT, one. The sources of the NMOS transistors NM11 and NM13 are commonly connected to the drain of the NMOS transistor NM15. The sources of the NMOS transistors NM12 and NM14 are connected to the drain of the PMOS transistor PM14 in common, and the input terminal of the inverter X3 outputs the enable signal Eo of the next stage. The conducting crystal logic 102.2 can ground the output signals OUT, —out according to the data input signal DATAIN. This paper size applies to China National Standard (CNS) A4 (_2Ι0 X 297 mm) --.------- i --- ί '^ -------- ^ ----- ---- (please ^ read me back ^-the meaning of the river 箅 箅 fill in this page) printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 419891 a? _________B7__ V. Description of the invention (3) It's Miller C element It is often used as a signal exchange circuit for asynchronous systems because of its delay insensitivity. When the two input values are equal, the output value is the same as the input value. If the two input values are different, the previous value is retained. The control block 101 configured in Figure 1 is mainly implemented by using Miller C elements for signal exchange control. In addition, the 'latch block 104 may be a conventional flow-latch type or Miller c element. However, the flow-latch does not have a delay insensitivity 'and therefore most of the Miller c elements are used. As shown in the example in FIG. 4, in the latch block 104, the confirmation signal ACK from the next stage is commonly applied to the gate of the PMOS transistor PM22, the source of the PM22 is applied with the power supply voltage vdd, and the NMOS transistor is commonly applied. The source of the NM22 gate body is grounded, and the input signal (Din, mainly OUT or Ot / 71) is commonly applied to the gate body of the PMOS transistor PM21. The source of the PM21 is connected to the drain of the PMOS transistor PM22, and In addition to the gate of the NMOS transistor NM21, the source of the NM21 is connected to the drain of the NMOS transistor NM22. The latch block 104 includes two circuits attached. The latch 104-1 is composed of two inverters X4 and X5 connected in anti-parallel and parallel. The latch is composed of a PMOS transistor PM21 and an NMOS transistor.

NM21之節點輸出之信號,及輸出被鎖存的信號DATAOUT 〇The signal output by the node of NM21, and the latched signal DATAOUT.

一般而言,第1圖配置之完成偵測器103包括一 NAND 閘或NOR閘俾便對來自功能方塊102之兩個輸出信號OUT 、進行邏輯作業以及產生作業完成信號。 本纸張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) I —r^ ίι--i n I n - I. - Γ[ * n I I ϋ 0« a n I n n η I 線丨- {請先閱讀背面之;i意事項再填寫本頁)Generally speaking, the completion detector 103 configured in FIG. 1 includes a NAND gate or a NOR gate, and performs logic operations on the two output signals OUT from the function block 102 and generates operation completion signals. This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) I —r ^ ίι--in I n-I.-Γ [* n II ϋ 0 «an I nn η I line 丨-丨(Please read the back; please fill out this page)

419891 發明說明(4) 現在說明習知非同步系統作業。 (請L閱讀背面之注意事項再填寫本頁) 當來自次一階段之確認信號ACOUT低時,若來自前 一階段之請求信號REQIN為低,則由控制方塊1〇1輸出的 時脈信號CK變高,功能方塊102進入評估期,評估輸出值 OUT及—Oi/r’以及完成偵測器103對輸出值ουτ、一0[/Γ 進行邏輯作業而產生一高輸出值β 若來自完成偵測器103之輸出值為高,則輪出值被傳 輸作為次一階段之請求信號REQOUT,以及作為前一階段 之確認信號ACKIN。 然後當來自前一階段之請求信號REQIN為高時,來自 控制方塊101之輸出信號維持其前一輸出態。當來自次一 階段之確認信號AC0UT為高時,控制方塊輸出的時脈 信號CK變低,如此’功能方塊1〇2進入預充電期。 如此來自完成偵測器103之輸出信號被產生為低,如 此使得傳輸至前一階段之確認信號ACKIN及傳輸至次一 階段之請求信號REQOUT值變低。 換言之功能方塊102之邏輯作業係藉重複前述過程進 行。 經濟部智慧財產局Μ工消費合作杜印制π 功能方塊102之組成可如第2或3圖之說明,敘述如下 若如第2圖所示執行,則當時脈信號CK為低時,PMOS 電晶體PM 1、PM2被導通,如此内部輸出端子亦即反相器 XI、X2之輸入端子被預充電於高準位。接收到來自内部 輸出端子之高信號的反相器XI、X2分別放電外部輸出端 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 經濟部智慧財產局員工消費合作祍印制衣 419891 A7 _____B7_ 五、發明說明(5 ) 子OUT、~Ί9ί/Γ至低準位。 隨後當時脈信號由低轉高時,NMOS電晶體ΝΜ1被導 通,如此串接邏輯102-1進入作業態。 此時,串接邏輯102-1根據輸入值DATAIN,放電内部 輸出端子亦即反相器XI、Χ2之輸入端子之一至低準位。 此處推定例如反相器XI之輸入端子被放電。 結果反相器XI其輸入端子被放電至低準位,反相器 之輸出端子OUT變成高,而反相器其輸入端子被充電至高 準位’反相器X2之輪出端子 〇 ^/7\维持於低準位。 然後’當時脈信號CK由高轉低時,資料作業係經由 重複前述過程於預充電態進行。 此外第3圖之CRDL電路之作業類似第2圖之電路,但 輸出端子OUT、—ΟϋΓ之預充電電壓為r 1/2vda』及其擺 動寬度為『l/2Vda』。 換言之第3圖之CRDL電路作業被歸類為預充電期及 評估期。 首先’於預充電期’當時脈信號CK為低時,PMOS電 晶體PM13被導通,如此輸出端子out ' —此連結 。結果輸出端子OUT、—ϋί/Γ之電壓準位根據電荷劃分效 應變成完全相同。 此時,輸出端子之準位為互補,換言之一者經常為『 Vdd』而另一者經常為rVss』。因此當輸出端子連結時, 電壓準位決定於『Vdd』與『Vss』$。一般而言,當輸 出端子OUT、—ϋί/Γ之寄生電容量類似時,該值決定為電 度適用中關家標準(CNS)A4規格⑵Q χ 297公楚) -- ij — 1'—11---Γ 装.I ------訂-111-111.,線丨 i -. <請先閱讀背面之江意事項再填寫本頁> A7 A7 419891 B7 _ 五、發明說明(6 ) 源電壓Vdd之半。 若由於寄生電容的不匹配導致等化電壓值係低於所需 值,則交叉耦合PMOS電晶體對PM11、PM12被導通,電 荷額外供給預充電節點俾便使電壓約等於電源電壓Vdd之 半。 隨後,具有高闉值電壓的PMOS電晶體PM11、PM12 已經被導通,如此即使獲得所需電壓值,輸出端子OUT、 Οί/Τ也不會被拉高。 時脈信號CK為低,如此PMOS電晶體ΡΜ14被導通, 反相器Χ3之輸入端子變高。 如此反相器Χ3輸出一低準位的致能信號Ε。 然後於評估期,當時脈信號CK變高時,PMOS電晶體 ΡΜ13被導通,如此輸出端子OUT、—彼此分開。 此時,導通電晶體邏輯102-2接收資料將輸出端子OUT 、—之一接地。此處例如推定輸出端子一ot/r接地。 結果,輸出端子OUT被維持於高準位,及輸出端子^ 之準位變低。 此外,時脈信號CK係於高準位,如此PMOS電晶體 PM14被關閉。然而,輸出端子OUT係於高準位,因此NMOS 電晶體NM14被導通。結果,反相器X3之輸入端子之準位 變低。 因而致能信號Eo由低轉高。 換言之,輸出端子OUT、間之電壓差越大,則 NMOS電晶體ΝΜ14被導通的速度愈快。因此,反相器Χ3 本紙張尺度適用中國國家標準(CNS)A4規格(2〗0 X 297公f ) -----------.--!產--------訂---------線 (請先'^讀背面,之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印制农 9 經濟部智慧財產局員工消費合作社印製 419891五、發明說明(7 ) 之輸入端子準位變低。此處被外加低信號之反相器X3輸 出致能信號Eo於高準位用於次一階段作動感測放大器。 隨後’當時脈信號CK由高轉低時,經由重複前述過 程於預充電態進行資料作業。 當由如第2或3圖說明之組成的功能方塊1 〇2輸出時, 資料OUT、—Ot/Γ係透過閂鎖方塊1〇4被傳輸至次一階段 ’閃鎖方塊104包括二電路係由複數電晶體pM21、PM22 、NM21、NM22及閂鎖104-1组成,如第4圖所示e 現在將以來自功能方塊1〇2之輸出資料信號〇υτ為例 說明作業。當來自次一階段之確認信號AC0UT係於低準 位時,若來自功能方塊1〇2之輸出資料信號〇^了係於低準 位,則唯有PM0S電晶體ΡΜ2ΐ、ΡΜ22被導通,電源電壓 被外加至閂鎖104-1,如此閂鎖輸出低階資料 DATAOUT。當確認信號AC0UT係於高階時,唯有NM〇s 電晶體NM21 ' NM22被導通,如此閂鎖1〇4-1之輸入端子 接地。結果’閂鎖104-1輸出高準位資料dataOUT。 當確認信號ACOUT係於低準位時,若來自功能方塊 102之輸出資料信號0UT係於高準位,或當確認信號 AC0UT係於高準位時,若來自功能方塊1〇2之輸出資料信 號OUT係於低準位,則閂鎖104·〗維持前—輸出準位直到 二輸出信號準位變相同為止。 習知技術理論上可保證低功率消耗及高速運作。實際 上’其缺點為比較同步系統無法顯著改進性能,原因為用 作功能方塊的DCVS邏輯之信號交換方案以及率消 本紙張尺度適用中關家鮮(CNS)A』ϋ咖X视公 (請先閱讀背面之注意事項再填寫本頁) ;.又* φ° --線 經濟部智慧財產局員工消費合作社印製 419991 A7 ______B7 _ 五、發明說明(8 ) 耗需要控制電路的額外開銷。 此外’當使用CRDL時,雖然功率消耗減低,但有另 —缺點為需要高電壓俾便提高交又耦合PM〇s電晶體對之 閾值電壓Vt。 發明概述 因此本發明之二目的係提供—種非同步感測之差動邏 輯(ASDL)電路,其經由應 <電荷循4術至^非同步、系統 而可減低功率消耗。 為了達到前述本發明之目的,於一種非同步系統包括 一控制方塊、一功能方塊、一完成偵測器及一閂鎖方塊, 功能方塊被設有完成偵測功能,藉克&除習完成偵測約 的需求且遠成低分率作業。 換言之一種根據本發明之非同步管線配置包括:一控 制方塊對來自前一階段之請求信號及次一階段之請求信號 進行邏輯作業,且輸出第一或第二輸入致能信號及第一或 第二時脈信號;一功能方塊其根據來自控制方塊之第一或 第二輸入致能信號及第一或第二時脈信號,對輸入資料進 行作業,且輸出第一或第二輸出致能信號及輸出資料;及 一閂鎖方塊其係由來自前一階段之確認信號觸發,對來自 功能方塊之第一或第二輸出致能信號及輸出資料進行作業 ’且輪出對次一階段請求信號及最終輸出資料。 圖式之簡單說明 經由參照附圖將更為瞭解本發明,附圖僅供舉例說明 之用’如此並非囿限本發明,附圖中: 本纸張K度適用中國國家標準(CN:S)A4規格(.210 X 297公g ) 11 --T--^--I I I I I Γ 楚:-----i —11* — — —----'線 * „ (請先閱讀背面之注意事項再填窵本頁) 413991 A7 __B7___ 五'發明說明(9) 第1圖為示意方塊圖說明習知非同步管線配置; 第2圖為示意電路圖說明第1圖配置中之功能方塊之第 —具體例; 第3圖為示意電路圖說明第1圖配置中之功能方塊之第 二具體例; 第4圖為示意電路圖說明第1圖之配置之閂鎖方塊; 第5圖為示意方塊圖說明本發明之非同步管線配置; 第6圖為示意電路圖說明根據本發明之第一具體例於 第5圖配置之功能方塊; 第7圖為示意電路圖說明根據本發明之第二具體例於 第5圖配置之功能方塊; 第8圖為示意電路圖說明根據本發明之第5圖配置之閂 鎖方塊; 第9圖為示意電路圖說明根據本發明之第三具體例於 第5圖配置之功能方塊; 第10圈為示意方塊圖說明第9圖中採用功能方塊電路 之三位元連鎖連結配置; 第11圖為示意電路圓說明根據本發明之第四具體例於 第5圖配置之功能方塊; 第12圖為示意方塊圖說明第丨丨圖中採用功能方塊電路 之五位元連鎖連結配置; 第13圖為示意電路圖說明根據本發明之第五具體例於 第5圖配置之功能方塊; 第14A至14D圖為根據本發明之非同步管線操作 本纸張尺度適用中國國家標準(CNS)A4規格(¾ 297公爱) -----I i I I l· I I [ ' 11--1 i 訂"-----I I I (請先閱讀背δ之·;i意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印?衣 A7 419391 B7 五、發明說明(ίο) 態之個別時序圖; 發明之詳細說明 現在將參照附圓說明根據本發明之非同步感測差動邏 輯(ASDL)電路細節。 第5圖示意方塊圖說明根據本發明之非同步管線配置 。如該圖所示,非同步管線配置包括:功能方塊2〇2執行 待執行的邏輯作業且其中具有完成偵測功能,且對各邏輯 作業產生完成信號;控制方塊2〇 I產生信號cKi/CKib或 Ei/Eib俾便於根據請求信號rEqIN、rEq〇uT進行作業時 控制功能方塊202之信號交換;及閂鎖方塊203其儲存由功 能方塊202送出的資料,輸出被儲存的資料,及同時輸出 次一階段之請求信號REQOUT。 此處,控制方塊201係由密勒C元件組成。 根據本發明之功能方塊202之多個具體例分別顯示於 第 6、7、9、Π及 13 圖。 第ό圖為示意電路圖說明根據本發明之第一具體例之 非同步電荷循環差動邏輯電路。如該圖所示,非同步電荷 循環差動邏輯電路包括:反相器Χ53反相時脈信號CKi, 及輸出被反相的時脈信號CKb ;反相器X54反相該被反相 的時脈信號CKib及輸出未反相的時脈信號CK ; 一對反相 器X51、X52交又耦合於反相器X53、X54之輸出端子CKb ' CK間,且分別閂鎖輸出信號q、Qb; NMOS電晶體NM53 連結於輸出端子Q、Qb間及於來自反相器X54之輸出信號 CK為高時等化輸出端子Q、Qb;串接邏輯電路202-1根據 本紙張尺度適用中國國家標準(CNS)A4規格(210 *297公i ) (請先閲讀背面之注意事項再填寫本頁) - I I I I ί I I ^ 1!]1!!11 1 經濟部智慧財產局員工消費合作社印制衣 13419891 Description of the invention (4) Now, the conventional asynchronous system operation will be described. (Please read the notes on the back and fill in this page again.) When the confirmation signal ACOUT from the next stage is low, if the request signal REQIN from the previous stage is low, the clock signal CK output by the control block 101 Becomes high, the function block 102 enters the evaluation period, and evaluates the output values OUT and -Oi / r 'and the completion detector 103 performs logic operations on the output values ουτ, 0 [/ Γ to generate a high output value β. The output value of the detector 103 is high, and the turn-out value is transmitted as the request signal REQOUT of the next stage and the acknowledgement signal ACKIN of the previous stage. Then when the request signal REQIN from the previous stage is high, the output signal from the control block 101 maintains its previous output state. When the confirmation signal ACOUT from the next stage is high, the clock signal CK output by the control block becomes low, so the function block 102 enters the precharge period. In this way, the output signal from the completion detector 103 is generated low, so that the acknowledgement signal ACKIN transmitted to the previous stage and the request signal REQOUT transmitted to the next stage become low. In other words, the logical operation of the function block 102 is performed by repeating the foregoing process. The composition of the π function block 102 produced by the Intellectual Property Bureau of the Intellectual Property Bureau of the Ministry of Economic Affairs can be described as shown in Figure 2 or 3, and described as follows. If implemented as shown in Figure 2, when the clock signal CK is low, the PMOS circuit The crystals PM1 and PM2 are turned on, so that the internal output terminals, that is, the input terminals of the inverters XI, X2, are precharged to a high level. The inverters XI and X2 that received the high signal from the internal output terminal discharge the external output respectively. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm). Clothing 419891 A7 _____B7_ V. Description of the invention (5) Child OUT, ~ Ί9ί / Γ to a low level. Subsequently, when the clock signal changes from low to high, the NMOS transistor NM1 is turned on, so that the logic 102-1 is connected in series to the operating state. At this time, the serial connection logic 102-1 discharges one of the internal output terminals, that is, one of the input terminals of the inverters XI, X2, to a low level according to the input value DATAIN. Here, it is assumed that, for example, the input terminal of the inverter XI is discharged. As a result, the input terminal of inverter XI is discharged to a low level, the output terminal OUT of the inverter becomes high, and the input terminal of the inverter is charged to a high level. \ Keep low. Then, when the clock signal CK changes from high to low, the data operation is performed in a precharged state by repeating the foregoing process. In addition, the operation of the CRDL circuit in Figure 3 is similar to the circuit in Figure 2, but the pre-charge voltage of the output terminals OUT and -0ϋΓ is r 1 / 2vda "and its swing width is" l / 2Vda ". In other words, the CRDL circuit operation in Figure 3 is classified into a precharge period and an evaluation period. First, "in the precharge period", when the clock signal CK is low, the PMOS transistor PM13 is turned on, so that the output terminal out '-this connection. As a result, the voltage levels of the output terminals OUT and -ϋί / Γ become completely the same according to the charge division effect. At this time, the levels of the output terminals are complementary. In other words, one is often "Vdd" and the other is often "rVss". Therefore, when the output terminals are connected, the voltage level is determined by "Vdd" and "Vss" $. In general, when the parasitic capacitances of the output terminals OUT and —ϋί / Γ are similar, the value is determined as the electrical energy applicable to the Zhongguanjia Standard (CNS) A4 specification ⑵Q χ 297 Chu)-ij — 1'—11 --- Γ Install.I ------ Order -111-111., Line 丨 i-. ≪ Please read the Italian matter on the back before filling this page > A7 A7 419891 B7 _ 5. Description of the invention (6) Half of the source voltage Vdd. If the equalized voltage value is lower than the required value due to the parasitic capacitance mismatch, the cross-coupled PMOS transistor is turned on to PM11 and PM12, and the charge is additionally supplied to the precharge node, so that the voltage is approximately equal to half the power supply voltage Vdd. Subsequently, the PMOS transistors PM11 and PM12 with high threshold voltages have been turned on, so that even if the required voltage value is obtained, the output terminals OUT, Οί / Τ will not be pulled high. The clock signal CK is low, so that the PMOS transistor PM14 is turned on, and the input terminal of the inverter X3 becomes high. In this way, the inverter X3 outputs a low-level enable signal E. Then during the evaluation period, when the clock signal CK goes high, the PMOS transistor PM13 is turned on, so that the output terminals OUT,-are separated from each other. At this time, the conductive crystal logic 102-2 receives the data and grounds one of the output terminals OUT and-. Here, for example, it is estimated that the output terminal ot / r is grounded. As a result, the output terminal OUT is maintained at a high level, and the level of the output terminal ^ becomes low. In addition, the clock signal CK is at a high level, so that the PMOS transistor PM14 is turned off. However, the output terminal OUT is at a high level, so the NMOS transistor NM14 is turned on. As a result, the level of the input terminal of the inverter X3 becomes low. Therefore, the enable signal Eo changes from low to high. In other words, the larger the voltage difference between the output terminals OUT, the faster the NMOS transistor NM14 is turned on. Therefore, the inverter X3 paper size applies to the Chinese National Standard (CNS) A4 specification (2〗 0 X 297 male f) -----------.--! -Order --------- line (please read the back of the page first, and pay attention to this page before filling in this page) Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 9 Printed by the Employees’ Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 419891 V. Description of the invention (7) The input terminal level becomes low. Here, the low-signal inverter X3 outputs the enable signal Eo at a high level for the next stage to actuate the sense amplifier. Subsequently, when the clock signal CK changes from high to low, the data operation is performed in the precharge state by repeating the foregoing process. When the output of the functional block 1 02 as illustrated in Figure 2 or 3 is output, the data OUT, -Ot / Γ are transmitted to the next stage through the latch block 104, and the flash block 104 includes two circuit systems. It is composed of complex transistors pM21, PM22, NM21, NM22 and latch 104-1. As shown in Figure 4, e will now take the output data signal 0υτ from function block 102 as an example to illustrate the operation. When the confirmation signal AC0UT from the next stage is at the low level, if the output data signal from the function block 102 is at the low level, only the PM0S transistors PM2 and PM22 are turned on, and the power supply voltage It is applied to the latch 104-1, so that the latch outputs the low-order data DATAOUT. When the confirmation signal AC0UT is high-order, only the NM0s transistor NM21 ′ NM22 is turned on, so that the input terminal of the latch 104-1 is grounded. As a result, the latch 104-1 outputs the high-level data dataOUT. When the confirmation signal ACOUT is at the low level, if the output data signal OUT from the function block 102 is at the high level, or when the confirmation signal AC0UT is at the high level, if the output data signal from the function block 102 OUT is at a low level, the latch 104 · is maintained before the output level until the two output signal levels become the same. The conventional technology can theoretically guarantee low power consumption and high-speed operation. Actually, its disadvantage is that the comparison system cannot significantly improve performance, because the DCVS logic signal exchange scheme used as a functional block and the paper size are applicable to Zhongguanxian (CNS) A. (Please read the notes on the back before filling this page);. And * φ °-printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Line Economy 419991 A7 ______B7 _ V. Description of the invention (8) It requires additional overhead of the control circuit. In addition, when CRDL is used, although the power consumption is reduced, there is another disadvantage: a high voltage is required, and the threshold voltage Vt of the AC coupled transistor is increased. SUMMARY OF THE INVENTION Therefore, a second object of the present invention is to provide an asynchronous sensing differential logic (ASDL) circuit, which can reduce power consumption by applying a charge cycle to an asynchronous system. In order to achieve the foregoing object of the present invention, an asynchronous system includes a control block, a function block, a completion detector, and a latch block. The function block is provided with a completion detection function. Detect the needs of appointments and become a low-scoring operation. In other words, a non-synchronous pipeline configuration according to the present invention includes: a control block performs logical operations on a request signal from a previous stage and a request signal from a next stage, and outputs a first or second input enable signal and a first or second Two clock signals; a function block that operates on the input data according to the first or second input enable signal and the first or second clock signal from the control block, and outputs the first or second output enable signal And output data; and a latch block which is triggered by a confirmation signal from the previous stage, and operates the first or second output enable signal and output data from the function block 'and rotates to the next stage request signal And final output data. Brief description of the drawings The present invention will be better understood by referring to the drawings. The drawings are for illustration purposes only. This is not a limitation on the present invention. In the drawings: The K degree of this paper is applicable to the Chinese national standard (CN: S) A4 specifications (.210 X 297 g) 11 --T-^-IIIII Γ Chu: ----- i —11 * — — — '' line * '(Please read the note on the back first Matters are refilled on this page) 413991 A7 __B7___ Five 'invention description (9) Figure 1 is a schematic block diagram illustrating the conventional asynchronous pipeline configuration; Figure 2 is a schematic circuit diagram illustrating the first functional block in the configuration of Figure 1— Specific example; Figure 3 is a schematic circuit diagram illustrating the second specific example of the functional block in the configuration of Figure 1; Figure 4 is a schematic circuit diagram illustrating the latch block of the configuration of Figure 1; Figure 5 is a schematic block diagram illustrating this Non-synchronous pipeline configuration of the invention; FIG. 6 is a schematic circuit diagram illustrating a functional block configured according to the first specific example of the present invention in FIG. 5; FIG. 7 is a schematic circuit diagram illustrating a second specific example according to the present invention in FIG. Configuration functional block; Figure 8 is a schematic circuit diagram illustrating Figure 5 according to the present invention Configuration of the latch block; Figure 9 is a schematic circuit diagram illustrating a functional block configured in Figure 5 according to a third specific example of the present invention; Circle 10 is a schematic block diagram illustrating the three bits of a functional block circuit in Figure 9 Chain connection configuration; FIG. 11 is a schematic circuit diagram illustrating a functional block configured in FIG. 5 according to a fourth specific example of the present invention; FIG. 12 is a schematic block diagram illustrating five bits of a functional block circuit in FIG. Chain connection configuration; Figure 13 is a schematic circuit diagram illustrating the functional block configured in Figure 5 according to the fifth specific example of the present invention; Figures 14A to 14D are operation of asynchronous pipelines according to the present invention. The paper dimensions are applicable to Chinese national standards (CNS) A4 specifications (¾ 297 public love) ----- I i II l · II ['11--1 i Order " ----- III (Fill in this page again) Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 419391 B7 V. Individual timing diagrams of the invention description (ίο); Detailed description of the invention will now be described with reference to the attached circle for non-synchronous sensing according to the invention Differential logic (ASDL) circuit is fine Figure 5 is a schematic block diagram illustrating the asynchronous pipeline configuration according to the present invention. As shown in the figure, the asynchronous pipeline configuration includes a function block 202 that executes a logical operation to be performed and has a completion detection function therein, and Generate a completion signal for each logical operation; the control block 20I generates a signal cKi / CKib or Ei / Eib 俾 to facilitate the handshake of the control function block 202 when performing operations according to the request signals rEqIN, rEq〇uT; and the latch block 203 stores The data sent by the function block 202 outputs the stored data and simultaneously outputs the request signal REQOUT of the next stage. Here, the control block 201 is composed of a Miller C element. Specific examples of the functional block 202 according to the present invention are shown in Figs. 6, 7, 9, and 13 respectively. Fig. 6 is a schematic circuit diagram illustrating a non-synchronous charge cycle differential logic circuit according to a first specific example of the present invention. As shown in the figure, the asynchronous charge cycle differential logic circuit includes: an inverter X53 inverts the clock signal CKi and outputs an inverted clock signal CKb; an inverter X54 inverts the inverted phase Pulse signal CKib and output non-inverted clock signal CK; a pair of inverters X51, X52 are alternately coupled between output terminals CKb 'CK of inverters X53, X54, and latch output signals q, Qb, respectively; The NMOS transistor NM53 is connected between the output terminals Q and Qb and the output signal Q and Qb are equalized when the output signal CK from the inverter X54 is high; the logic circuit 202-1 is connected to the Chinese national standard according to this paper standard ( CNS) A4 specification (210 * 297 male i) (Please read the precautions on the back before filling this page)-IIII ί II ^ 1!] 1 !! 11 1 Printed clothing by the Intellectual Property Bureau of the Ministry of Economic Affairs, Consumer Consumption Cooperative 13

419S9I A? B7 五、發明說明() 輸入資料D ATAIN產生輸出端子Q、Qb間之電壓差異;及 NMOS電晶體NM54於來自控制方塊2〇1之致能信號招為内 時連結串接邏輯電路201-1接地。 反相器X51、X52分別係由互補成對電晶體(pM5】、 NM51)(PM52、NM52)串聯連結於時脈信號CKb、CK間組 成。 第7圖為示意電路說明根據本發明之第二具體例之非 同步電荷循環差動邏輯電路。如該圖所示,非同步電荷循 環差動邏輯電路包括:反相器X64係由電晶體PM64、NM64 串聯連結於電源電壓Vdd及地電壓Vss間組成,且分別於 其閘體接收致能信號Ei及時脈信號CKi,及於其汲極輸出 被反相的時脈信號CKb :反相器X65係由電晶體PM65、 NM65串聯連結於電源電壓Vdd及地電壓Vss間,且分別於 其閘體接收時脈信號CKib及致能信號Eib以及輸出非反相 時脈信號CK ;反相器χ61、χ62交叉耦合於時脈信號CKb 、CK間’且分別閂鎖輸出信號q、Qb ; NMOS電晶體63其 通道連結於輸出端子Q、Qb間,且當來自反相器Χ65之輸 出為高時等化輸出端子Q、Qb之電壓準位;信號 輸出單元202-3連結於電源電壓Vdd與輸出端子Q、Qb間, 且接收反相後之時脈信號CKb及輸出信號Q、Qb ’以及輸 出致能信號Eb ;反相器X63反相致能信號Eb,且輸出致能 信號Eo ;串接邏輯電路2〇2_2根據輸入資料DATAm產生輸 出端子Q、Qb間之電壓差;及nm〇S電晶體NM68於致能 k號£11)為高時連結串接邏輯電路2〇2>_2接地。 本紙張尺度關家標準(CNS^i.(2K]x297公楚) (請先閱讀背面之;i意事項再填寫本頁) ^衣·----I--訂---------,絲 經濟部智慧財產局員工消費合作社印 14 經濟部智慧財產局員工消費合作社印製 419891 B; 五、發明說明(12) 此處,於信號輸出單元202-3,被反相的時脈信號CKb 外加於PMOS電晶體PM63之閘體,PM63之源極被外加電 源電壓Vdd,輸出信號Q被外加至NMOS電晶體NM66之閘 體,NM66之源極被外加輸出信號Qb,輸出信號Qb被外加 至NMOS電晶體NM67之閘體,NM67之源極被外加輸出信 號Q,結果致能信號Eob由電晶體PM63、NM66、NM67之 汲極共通節點輸出。 反相器X61、X62分別係經由個別互補電晶體對(PM61 、NM61)(PM62、NM62)串聯連結於時脈端子CKb ' CK間 組成。 第9圊為示意電路圖說明根據本發明之第三具體例之 非同步電荷循環差動邏輯電路。如該圖說明,非同步電荷 循環差動邏輯電路包括:反相器X83反相被反相的致能輸 入信號Eib,且輸出致能信號Eo ;反相器X84反相未被反 相之致能輸入信號Ei,且輸出被反相致能信號Eob ; —對 反相器X81、X82交叉耦合於反相器X83及X84之輸出端子 Eo、Eob間,且個別閂鎖輸出信號Q、Qb;NMOS電晶體NM83 之通道係連結於輸出端子Q、Qb間,且當外加於其閘體之 被反相的致能輸入信號Eib為高時,等化輸出端子Q、Qb 之電壓準位;串接邏輯電路202-4根據輸入資料DATAIN產 生輸出端子Q、Qb之電壓差;及NM0S電晶體NM84其於 致能輸入信號Ei為高時連結串接邏輯202-4接地。 反相器X81、X82係由個別成對互補連結的電晶體 (PM8卜NM81)(PM82、NM82)串聯連結於致能信號Eo'Eob 本紙張又度適用中國國家標準(CNS)A4規格(2〗0 X 297公f > 15 II 一 ί n n I I If n c^i I t l i— I n 1^1 I 一OJ n ϋ —1 I 1 {請先閱讀背面之注意事項再填寫本頁) 419891 五、發明說明(η) 間組成。 第11圊為示意電路圖說明根據本發明之第四具體例之 非同步電荷循環差動邏輯電路。如第11圖之說明,非同步 電荷循環差勤邏輯電路包括·反相器X93反相未被反相之 致此輪入破Ei,反相器X94反相來自反相器X93之輸出 k號,且輪出致能信號Eo ; —對反相器χ91、X92交叉搞 合於電源電壓Vdd與反相器X93之輸出端子£i,間,以及分 別閂鎖輸出信號Q、Qb ; PMOS電晶體PM93之通道係連結 於輸出端子Q、Qb間,且當外加於其閘體之致能信號四為 低時,等化輸出端子Q、Qb之電壓準位;串接邏輯電路2〇2_5 根據輸入資料DATA IN產生輸出端子Q、Qb間之電廢差; 及NMOS電晶體NM93其於致能信號Ei為高時連結串接邏 輯電路202-5接地。 反相器X91、X92係經由串聯連結電晶體(pM91、 NM91)(PM92、NM92)於電源電壓Vdd與致能輸入信號Ei, 間之互補電晶體對組成。 第13圖為示意電路圖說明根據本發明之第五具體例之 非同步電荷循環差動邏輯電路。如該圊所示,非同步電荷 循環差動邏輯電路包括:一反相器係由電晶體PM1 〇4、 NM104串聯連結於電源電壓Vdd與地電壓Vss間組成,且 個別於其閘體接收致能輸入信號Ei及來自前一階段作業完 成信號Dni,以及輸出致能信Ei,; 一反相器χΐ〇3反相致能 信號Ei’,且輸出致能信號Ε〇 ;個別反相器Xioi、Xl〇2交 叉耦合於電源電壓Vdd與致能信號Ei間,且個別閂鎖輸出 本紙張坨度適用中固國家標準(CNS)A4規格(210 X 297公釐) : - 裝--- (請^-閱讀背面之主咅^事項再填寫本頁) 訂: --線 經濟部智慧財產局員工消費合作社印製 16 A7 一 413S91 — 五、發明說明(Η) 信號Q、Qb ; — PMOS電晶體ΡΜ103其通道係連結於輸出 端子Q、Qb間’且當致能輸入信號Ei為低時,等化輸出端 子Q、Qb ; —信號輸出單元202-7連結於接地與輸出端子q 、Qb間’接收致能信號E’及輸出信號Q、Qb,且輸出作 業完成信號Dno : —串接邏輯電路202-6根據可變資枓輸 入DATAIN輸出資料信號至輸出端子q、Qb;及NMOS電 晶體NM105其於致能輪入信號Ei為高時連結串接邏輯電路 202-6接地。 此處,於信號輸出單元202-7,致能輸入信號Ei被外 加NMOS電晶體NM103之閘體,NM103之源極接地,輸出 波Q被外加至P Μ 0 S電晶體Ρ Μ10 5之閘體,其源極被外 加輸出信號Qb,及輸出信號Qb係外加至PMOS電晶體 PM106之閘體,PM106之源極被外加輸出信號Q,如此由 電晶體NM103、PM105、PM106之共通〉及極節點輸出作業 完成信號Dno。 反相器XI01、XI02係經由將個別互補電晶體對 (PMl(H、NM101)(PM102、NM102)串聯連結於電源電壓Vcίd 與致能信號Ei’間組成。 如第8圊之說明,於閂鎖方塊203,確認信號ACKIN 外加至PMOS電晶體PM72之閘體,PM72之源極係連結至 電源電壓Vdd,也外加至NMOS電晶體NM72之閘體,NM72 之源極接地。輸入資料信號Din以及來自功能方塊202之輸 出信號OUT或—Οί7Γ外加至NMOS電晶體NM73之閘體, NM73之源極係連結至NMOS電晶體NM72之汲極,以及來 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) I—F I I· I- ί t, I n tlr i - n n E I n t— n n n tf f I— k 1 1 (靖it-閱璜背面之;i意事項再填寫本頁) 經濟部智慧財產局員工消費合作杜印製 17 經濟部智慧財產局員工湞費合作社印製 419391 A7 B7 五、發明說明(u) 自功能方魂202之致能信號E〇係外加至PMOS電晶體PM7 1 之閘體,PM71之源極係連結至pm〇S電晶體PM72之沒極 ’同時也外加至NMOS電晶體NM71之閘體,NM71之源極 係連结至NMOS電晶體NM73之沒極。電晶體pm71、NM71 之共通汲極節點係連結至閂鎖203-1之輸入端子,其中一 對反相器X72 ' X71係並聯反相連結用於輸出輸出信號out ,及設置延遲電路203-2用於廷遲來自功能方塊2〇2之輸出 信號Eo ’因此對次一階段產生請求信號REQOUT。 現在將參照第14A至14D圖說明根據本發明之非同步 管線配置之作業及效果。 於一狀態其中次一階段之請求信號REQOUT係於低準 位’當來自前一階段之請求信號REQIN係於高準位時,來 自控制方塊201之輸出信號CKi或Ei變高。結果,功能方 塊202進入評估期’評估輸出值’及產生於高準位之致能 信號Eo。 此處致能信號Eo被輸出作為前一階段之確認信號 ACKOUT。 隨後’閂鎖方塊203對來自功能方塊202之輸出資料進 行操作’延遲致能信號Eo經歷一段預定時間,及對次一 階段產生請求信號REQOUT,藉此通訊閂鎖期作業完成。 此處,當次一階段之請求信號REQ0UT處於高準位且 經歷一段預定時間時’來自控制方塊201之輸出信號CKi 或Ei變低’及功能方塊202返回預充電期。 如此功能方塊202之輸出端子Q、Qb等於『i/2Vda』, 本紙張尺度適用申國國家標準(CNS)A4規格(210 X297公笼) 18 I .----------^ I 裝--------訂·--------線- (請先閱讀背面之注意事項再填罵本頁) 41SS91 A7 B7 五、發明說明(16) 及功能方塊202於經過預定時間後輸出低準位的致能信號 Eo。 (請先閱讀背面之注意事項再填寫本頁) 換言之根據本發明具有管線配置之非同步系統係經由 根據第14A至第14D圖示例說明之時序重複進行前述過程 進行資料作業。結果,閂鎖方塊203閂鎖由功能方塊202輸 出資料’及傳輸資料至次一階段。 它方面,來自功能方塊202之資料透過閂鎖方塊203傳 輸至次一階段’閂鎖方塊203包括電晶體PM71、NM71-NM73、閂鎖203-1及延遲電路203-2,如第8圖之說明。 換言之’當確認信號ACKIN及致能信號Eo係於低準 位時’高準位信號外加至閂鎖2〇3_1,如此輸出資料信號 DATAOUT係以低準位輸出《於確認信號ACKIn及致能信 號Eo係於高準位狀態下,當輸入資料信號Din變高時,低 準位k號外加之閂鎖203-1 ’如此維持輸出資料信號 DATAOUT之前一低準位態。 於確認信號ACKIN與致能信號處於不同準位之案例 ’閂鎖203-1維持前一輸出準位。 經濟部智慧財產局員工消費合作社印*'1衣 延遲電路203-2接收致能信號E〇 ,延遲致能信號£〇經 歷預定時間,及輸出次-階段的請求信號REQ〇UT。 此外,功能方塊202係由非同步電荷循環差動邏輯電 路組成。現在解說根據本發明之具體例之功能方塊2〇2之 作業。 1*先說明執行第6圖所示電路之根據本發明之第一具 體例之功能方塊202之作業。 本紙張尺度過用中國國家標準(CNS)A4規格(21〇 297公釐) 19 A7 41GS91 ____B7_____ 五、發明說明(Π) 於來自控制方塊201之致能信號Ei處於低準位之狀態 下,當時脈信號CKi係於高準位而時脈信號CKib係於低準 位時’功能方塊202進入預充電期。高準位時脈信號cKi 被反相器X53反相,如此輸出作為低準位時脈信號CKb。 低準位時脈信號CKib由反相器X54反相,如此輸出為高準 位時脈信號CK。 因此,時脈信號CK、CKb分別係於高及低準位,如 此交叉耦合反相器X51、X52就輸出端子Q、Qb而言係維 持於關態。屬於等化電晶體之NMOS電晶體NM53藉高準 位時脈信號CK導通,然後輸出端子Q、Qb之電壓準位被 預充電至『l/2Vda』。 此處來自控制方塊201之致能信號Ei處於高準位,以 及NMOS電晶體NM54被導通狀態下,串接邏輯電路202-1 错由對輸入資料DATAIN執行邏輯運算而產生輸出端子q 、Qb間之電壓差。推定輸出端子q係於高準位。 隨後’當時脈信號CKi由高轉低’且時脈信號CKib由 低轉高時’功能方塊202進入評估期。反相器χ53反相低 準位時脈信號CKi及輸出南準位時脈信號CKb。反相器X54 反相高準位時脈信號CKib ’且輸出低準位時脈信號ck。 因此反相器X5 1、X52被致能’如此以高速評估輸出 端子Q、Qb之資料值。由於推定輸出端子Q係於高準位, 故PMOS電晶體PM51被斷路而NMOS電晶體NM51被導通 ,如此反相器X51下拉輸出端子Qb至低準位。此外,pM〇s 電晶體PM52被導通,NMOS電晶體NM52被斷路’如此反 本紙張尺度適用中國國家標準(CN幻人j規格(210 x 297公釐> t ------------ I ^--------^--------- 1 (請先閱讀背面之注意事項再填骂本頁) 經濟部智慧財產局員工消費合作社印制^ 20 —41咖1_:___ 五、發明說明(18 ) 相器X52上拉輸出端子Q至高準位。 (請先閱讀背面之注意事項再填茑本頁) 輸出端子Q、Qb個別準位根據前述作業傳輸至次一階 段,結果時脈信號CK、CKb也傳輸次一階段作為時脈信 號CKi、CKib。 其後,當時脈信號CKi由低轉高及時脈信號CKib由高 轉低時,功能方塊202返回預充電期。結果,高準位時脈 信號CKi被反相器X53反相,且輪出作為低準位時脈信號 CKb。此外’低準位時脈信號CKib被反相器χ54反相,及 輸出作為高準位時脈信號CK。 因時脈信號CK、CKb分別係於高及低準位,故交叉 耦合反相器X51 ' X52就輸出端子q、Qb而言被斷路,及 NMOS電晶體NM53響應高準位時脈信號CK被導通,藉此 連結輸出端子Q、Qb。結果,輸出端子Q、Qb被預充電至 『l/2Vda』。 換言之’根據時脈信號CKi、CKib之準位變化,預充 電期及評估期重複進行,如此輸入資料循序傳輸至次一階 段。 經濟部智慧財產局員工消費合作社印製 本發明之第一具體例於功能方塊202之階段作業僅延 遲短時間即極為有效。 現在參照第7圖說明根據本發明之第二具趙例執行第7 圖之電路之功能方塊202之作業。 第7圖之電路作業上極為類似第6圖之電路,但交又耦 合反相器X61、X62之致能時序除外。 以第7圊之電路為例,於時脈信號CKi改成高準位 21 本紙張尺度適用中國國家標準(CNS)A4規格(210x 297公釐) 419891 A7 B7 五、發明說明(19) 態下,電流端子之交叉耦合反相器X61、X62並未致能直 到前一階段之作業完成而致能輸入信號Ei變成激活為止。 於根據本發明之第二具體例之非同步電荷循環差動邏 輯電路中,當功能方塊202之一階段作業延遲相對長時間 時,因前評估造成的缺點可被克服。現在說明其作業細節 〇 當時脈信號CKi係於高準位而時脈信號CKib係於低準 位時,致能輸入信號Εί係於高準位,輸入致能信號Eib係 於低準位,及反相器X61、X62被去能。 此時,NMOS電晶體NM64被高準位時脈信號CKi導通 ,及時脈端子CKb係於低準位。PMOS電晶體PM65由低準 位時脈信號CKib被導通,及時脈信號CK係於高準位。 時脈信號CKb、CK分別係於低及高準位,如此交叉 耦合反相器X61、X62維持於關態。NMOS電晶體NM63由 高準位時脈信號CK導通,及輸出端子Q、Qb被預充電至 『l/2Vda』。 時脈信號CKb係於低準位,如此於信號輸出單元202-3,PMOS電晶體PM63被導通及以高準位輸出致能信號Eb 。反相器X63被外加高準位致能信號Eob,反相器X63輸出 低準位致能輸入信號Eo。 隨後當時脈信號CKi由高轉低且時脈信號CKib由低轉 高時,若致能輸入信號Ei由高轉低而輸入致能信號Eib由 低轉高,則NMOS電晶體NM64及PMOS電晶體PM65各自 被斷路,以及PMOS電晶體PM64及NMOS電晶體NM65各 本紙張尺度適用争囤國家標準(CNS)A4規格(210x 297公t ) I If I— n I— n n ^1* .^1 n 1 JT · - rf i ί i i n 04 a n n I— n n n I ^ >1 (請先閱讀背面之注音;事項再填寫本頁) 經濟部智慧財產局員工消費合作社印制农 22 A7 419S91 B7 五、發明說明(20) 自因此被導通。 (請先閱讀背面之注咅〕事項再填寫本頁) 如此,反相時脈信號CKb係於高準位而非反相時脈信 號CK係於低準位》結果交叉耦合反相器X61、X62就輸出 端子Q、Qb而言被置於作業態。 此時,於NMOS電晶體NM68藉由輸入致能信號Eib外 加至其閘體變成導通狀態下,串接邏輯電路202-2根據輸 入資料DATAIN產生輪出端子Q、Qb間之電壓差。此處例 如推定輸出端子Q係於高準位。 如此PMOS電晶體PM61被斷路及NMOS電晶體NM61 藉由外加輸出端子Q之高準位信號而被導通。因此反相器 X61以高速下拉輸出端子Qb之準位。它方面,PMOS電晶 體PM62被導通,而NMOS電晶體NM62藉由外加輸出端子 Qb之低準位而被斷路。結果,反相器X62高速上拉輸出端 子Q之準位。 此處,於信號輸出單元202-3,PMOS電晶體PM63藉 由外加高準位時脈信號CKb被斷路》但於上例推定輸出端 子Q係於高準位,如此NMOS電晶體NM66被藉此導通而致 能信號Eb變低。 經濟部知S慧財產局員工消費合作杜印制衣 如此,反相器X63反相處於低準位之致能信號Eob, 如此致能信號Eo變高。 然後,當時脈信號CKi由低轉高而時脈信號CKib由高 轉低時,若輸入致能信號Ei由低過渡至高而輸入致能信號 Eib由高過渡至低,則交叉耦合反相器X61、X62被去能, NMOS電晶體NM63被導通,如此輸出端子Q、Qb再度被 23 本紙張尺度適用中國國家標準(CNS)A4規格(210 x 297公釐) A7 419391 B7 _ 五、發明說明(21 ) 預充電至『l/2Vda』。 月’J述作業係於控制方塊201產生之信號Ej/Eib、 CKi/CKib準位改變時重複進行。 現在說明根據本發明之第三具體例執行第9圖所示電 路之功能方塊202之作業及效果。 當被反相之輸入致能信號E i b係於高準位及非反相之 輸入致能信號Ei係於低準位時,於等化期,高準位輪入致 能信號Eib由反相器X83反相且輪出作為低準位致能信號 E〇,而低準位輸入致能信號Ei由反相器χ84反相且輸出作 為南準位致能信號Eob。 如此致能信號Eo、Eob分別係於低及高準位,因此交 叉耗合反相器X81、X82就輸出端子Q、Qb而言維持於關 態。屬於等化電晶體之NMOS電晶體NM83藉由外加高準 位輸入致能信號Eib被導通,因此預充電輸出端子Q、Qb 之電壓準位至『l/2Vda』。 隨後,當輸入致能信號Eib由高轉低,而輸入致能信 號Ei由低轉高時’致能信號Eo轉高,致能信號E〇b轉低, 如此開始感測/評估期。 此時’於NM0S電晶體NM83由外加低準位輸入致能 Ίέ號Eib斷路而NMOS電畢體NM84由外加高準位輸入致能 信號Ei導通狀態下’串接邏輯電路202-4對輸入資料進行 邏輯運算’藉此產生輸出端子Q、Qb間之電壓差^此處例 如推定輸出端子Qb被放電至地電位,如此變低。 因此反相器X81、X82以高速評估輸出端子q、Qb間 本紙張K度適用中國國家標準(CNS)A4規格(210 x 297公t ) i』----II I — I l· I * I I I I---«— — tilt-- (請^閱讀背面之注意事項再填罵本頁) 經濟部智慧財產局員工消費合作社印?衣 24 經濟部智慧財產局員工消費合作社印製 41999i A7 _________B7 _ 五、發明說明(22 ) 之電壓差。前例係推定輸出端子Q之準位為高。因此pM〇S 電晶體PM81被斷路,NM0S電晶體NM81被導通,如此反 相器X81下拉輸出端子Qb至低準位。此外pM〇s電晶體 PM82被導通,NMOS電晶體NM82被斷路,如此反相器χ82 上拉輸出端子Q至高準位》 輸出端子Q、Qb之準位根據前述作業傳輸至次一階段 。致能彳§號Eo、Eob也被傳輸作為再次一階段的輸入致能 信號 Ei、Eib。 隨後輸入致能信號Ei由高轉低,致能信號Eib由低轉 高’如此開始等化期。此時反相器χ83反相高準位輸入致 能信號Eib且輸出低準位輸入致能信號E〇,而反相器χ84 反相低準位致能信號Ei且輸出高準位致能信號E〇b。 如此,致能信號Eob、Eo分別係於高及低準位,因而 父叉耦合反相器X81、X82進入關態。:NTMOS電晶體NM83 由高準位輸入致能信號Eib導通’且連結輸出端子q' Qb 。結果,輸出端子Q、Qb被預充電至r 1/2Vda』。 換言之’根據輸入致能信號Ei、Eib準位的改變,等 化期及感測/評估期重複交換,如此輸入資料DATAIN循序 傳輸至次一階段。 如第10圖所示,功能方塊之三位元連鎖連結配置可經 由串聯連結功能方塊電路211-213實施,功能方塊電路 211-213各自執行第9圖所示電路而從事前述作業。 現在解釋根據本發明之第四具體例執行第1丨圖所示電 路之功能方塊202之作業及效果。 ---I.----------,· ^--------訂---------線 (諳先閱讀背面之;i意事項再填駕本頁) 本紙張尺度適用中圈囷家標準(CNS)A4規格(2ιυ x 297公釐) 25 經濟部智慧財產局員工湞費合作社印制取 419991 at _ B7___ 五、發明說明(23 ) 輸入致能信號Ei係於低階及於評估期。反相器χ93、 X94循序反相低準位輸入致能信號及輸出低準位致能信號 Eo 〇 此時’來自反相器X93之輸出信號Ei,係於高準位,如 此反相器X91、X92就輸出端子Q、Qb而言交叉搞合且組 成上拉/下拉電路進入關態。PMOS電晶體PM93由準位輸 入致能信號Ei導通’因而預充電輸出端子q、Qb之準位至 『l/2Vda』。 CMIS反相器X9〗' X92個別之PMOS電晶體PM91、 PM92直接連結至電源電壓Vdd ’如此預充電電位於此電 路略比第9圖之電路更南。 隨後’當輸入致能信號Ei由低轉高時,開始感測/評 估期。 此時,於PMOS電晶體PM93被斷路及NM〇s電晶艘 NM93由外加高準位致能信號Ei導通之狀態下,串接邏輯 電路202-5對資料DATAIN執行邏輯運算,因而產生輸出端 子Q、Qb間之電壓差。此處例如推定輸出端子砂被放電 至地電位而變低。 因此,反相器X91、X92以高速評估輸出端子q、Qb 間之電壓差。於上例推定輸出端子Q之準位為高。因此, PMOS電晶體PM91被斷路,NMOS電晶體>^91被導通, 如此反相器X91下拉輸出端子Qb至低準位。此外,pM〇s 電晶體PM92被導通’ NMOS電晶體NM92被斷路,如此反 相器X92上拉輸出端子Q至高準位。 本紙張又度適用中國國家標準(CNS)A4規格(210 297公釐) (請先閱讀背面之汰意事項再填寫本頁) 裝·-------訂--------·線 26 419891 A7 __— ___ B7 五、發明說明(24 ) 根據前述操作’輸出端子Q'Qb之準位傳輪至再次一 階段。致能信號E〇也傳輸作為次一階段的輸入致能信號四 其後’輸入致此號Ei由高轉低,及開始評估期。此 時’反相器X93、X94循序反相低準位輸入致能信號m, 且輪出低準位致能信號Eo。 因此’致能k號Eo係於低準位及交叉耗合反相器1 、X92被去能。PMOS電晶體PM93由外加低準位輸入致能 "is號Ei被導通,及連結輸出端子Q、Qb。結果,輪出端子 Q、Qb被預充電至約Fi/2Vda』。 換言之’當輸入致能信號Ei之準位改變時,等化期及 感測/評估期重複交換,藉此循序傳輸輸入資料D ATAIN至 再次一階段》 如第12圖所示’功能方塊之五位元連結連鎖配置可經 由串聯連結功能方塊電路22卜225之資料端子實施,221-225各自執行第11圊所示電路用於從事前述操作6此種情 況下,輸入致能信號Ei具有預定延遲時間,且被傳輸至次 一階段後的階段。舉例言之’電路221之輸入致能信號透 過例如由一對反相器串聯連結組成的延遲電路而被輸入作 為電路223之致能信號Ei »結果獲得致能信號Eo及資料信 號Q。 最後,敘述根據本發明之第五具體例執行如第13圖所 示電路之功能方塊202之作業之效果。 當輸入致能信號Ei係於低準位以及來自前一階段之作 本紙張尺度適用中國國家標準(CNS)A4規格(210^ 297公f ) 一 27 - (請先閱讀背面之注意事項再填寫本頁) r 經濟部智慧財產局員工消費合作社印刹衣 經濟部智慧財產局員1消費合作社印製 419891 A7 _______B7 五、發明說明(25 ) 業完成信號Dni係於低準位,如此開始等化期時,pm〇S 電晶體PM104被導通’致能信號Ei’變高,如此反相器χιοι ' XI02處於關態。 此時’致能信號Ei’係於高準位,如此於信號輸出單 元202-7 ’ NMOS電晶體NM103被導通,及輸出低準位的 作業完成信號Dno。反相器XI03接收高準位致能信號Ei, ,輸出低準位致能信號Eo。 PMOS電晶體PM103由外加低準位輸入致能信號£丨被 導通’如此輸出端子Q、Qb被預充電至『i/2Vda』。 其後’當輸入致能信號Ei由低過渡至高時,pm〇S電 晶體PM103、ΡΜ1〇4被導通,輸出端子q、Qb彼此解除連 結’及致能信號Ei ’係於高阻抗態。 此時,於NMOS電晶體NM105由外加高準位致能信號 被導通之狀態下’串接邏輯電路202-6藉由對輸入資料 DATAIN執行運算而產生輸出端子Q、Qb間之電壓差。此 處例如推定輸出端子Qb變高。 > 來自前一階段之作業完成信號Dni轉高時,NMOS電 晶體NM104被導通,且致能信號Ei,變低。結果,交叉耦 合反相器XI01、X012被作動。反相器X〗03反相低準位致 能信號Ei’ ’如此致能信號Eo變高。 因此根據輸出端子Q之高準位,PMOS電晶體PM101 被斷路,NMOS電晶體NM101被導通,如此反相器χι〇1以 高速上拉輸出端子Qb之準位。根據輸出端子Qb之低準位 ’ PMOS電晶體PM102可導通’ NMOS電晶體NM102被斷 本纸張尺度適用中國國家標準(CNS)A4規格(210x 297公餐) 28 — -------r I --------^ --------丨線 (請先閱讀背面之江意事項再填冩本頁) 419891 A7 —…__B?____ 五、發明說明(26 ) 路’如此反相器X102以高速上拉輸出端子q之準位。 (請先閱讀背面之;1意事項再填寫本頁) 此外’於信號輸出單元202-7,因致能信號Ei,係於低 準位,NMOS電晶趙NM103因而被斷路。但輸出端子q係 於尚準位’輸出端子Qb係於低準位,如此pm〇S電晶體 PM106被導通,藉此輸出高準位的作業完成信號Dn〇。 隨後,當輸入致能信號Ei由高過渡至低時,若來自前 一階段之作業完成信號Dni由高轉低,因而PMOS電晶體 PM104被導通,則致能信號Ei,變高。如此,交又耦合反 相器又101、又102被去能,而卩1^〇8電晶體?1^103被導通, 藉此再度預充電輸出端子Q、Qb至『〗/2Vda』。 每當由控制方塊201傳輪至功能方塊2〇2之輸入致能信 號Ei之準位過渡時,重複執行前述作業。 根據前述本發明之具體例之串接邏輯電路2〇2_ 1、 202-2 ' 202-4、202-5、202-6可由互補輸出導通電晶體邏 輯網路替代’該網路係根據可變資料輸入於輸出端子Q、 Qb產生資料。419S9I A? B7 V. Description of the invention () Input data D ATAIN generates the voltage difference between output terminals Q and Qb; and NMOS transistor NM54 is connected to the logic circuit when the enable signal from control block 201 is internal. 201-1 is grounded. The inverters X51 and X52 are respectively composed of complementary paired transistors (pM5) and NM51 (PM52 and NM52) connected in series between the clock signals CKb and CK. Fig. 7 is a schematic circuit illustrating a non-synchronous charge cycle differential logic circuit according to a second specific example of the present invention. As shown in the figure, the non-synchronous charge cycle differential logic circuit includes: the inverter X64 is composed of transistors PM64 and NM64 connected in series between the power supply voltage Vdd and the ground voltage Vss, and receives the enable signal at its gate respectively. Ei clock signal CKi and clock signal CKb whose drain output is inverted: the inverter X65 is connected in series by the transistor PM65 and NM65 between the power supply voltage Vdd and the ground voltage Vss, and is separately connected to its gate. Receive clock signal CKib and enable signal Eib and output non-inverting clock signal CK; inverters χ61 and χ62 are cross-coupled to clock signals CKb and CK 'and latch output signals q and Qb respectively; NMOS transistor 63 The channel is connected between the output terminals Q and Qb, and the voltage level of the output terminals Q and Qb is equalized when the output from the inverter X65 is high; the signal output unit 202-3 is connected to the power supply voltage Vdd and the output terminal Between Q and Qb, and receive the inverted clock signal CKb and output signals Q, Qb 'and output enable signal Eb; inverter X63 inverts enable signal Eb and outputs enable signal Eo; serial logic The circuit 2〇2_2 generates an output terminal according to the input data DATAm The voltage difference between the sub-Q and Qb; and the nmOS transistor NM68 is connected to the serial logic circuit 2O2 > _2 when the enabling k number is £ 11) is high. The standard of this paper standard (CNS ^ i. (2K) x297). (Please read the back of the page; i-notes before filling out this page) ^ 衣 · ---- I--Order ------ ---, printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 14 printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 419891 B; 5. Description of the invention (12) Here, the signal output unit 202-3 is inverted The clock signal CKb is applied to the gate of the PMOS transistor PM63. The source of the PM63 is applied with the power supply voltage Vdd. The output signal Q is applied to the gate of the NMOS transistor NM66. The source of the NM66 is applied with the output signal Qb. Qb is applied to the gate of the NMOS transistor NM67, and the source of the NM67 is applied with the output signal Q. As a result, the enable signal Eob is output from the common node of the transistors PM63, NM66, and NM67. The inverters X61 and X62 are respectively The individual complementary transistor pairs (PM61, NM61) (PM62, NM62) are connected in series between the clock terminals CKb 'CK. Section 9 is a schematic circuit diagram illustrating a non-synchronous charge cycle differential according to a third specific example of the present invention. Logic circuit. As the figure illustrates, the asynchronous charge cycle differential logic circuit package : Inverter X83 inverts the inverted enable input signal Eib and outputs the enable signal Eo; Inverter X84 inverts the inverted enable input signal Ei and outputs the inverted enable signal Eob ;-The inverters X81 and X82 are cross-coupled between the output terminals Eo and Eob of the inverters X83 and X84, and the individual output signals Q and Qb are latched; the channel of the NMOS transistor NM83 is connected to the output terminals Q and Qb And when the inverted enabling input signal Eib applied to its gate body is high, the voltage levels of the output terminals Q, Qb are equalized; the serial connection logic circuit 202-4 generates the output terminal Q according to the input data DATAIN The voltage difference between Qb and Qb; and NM0S transistor NM84, which is connected to the series logic 202-4 when the enable input signal Ei is high. The inverters X81 and X82 are connected by individual complementary transistors (PM8 and NM81). ) (PM82, NM82) connected in series to the enable signal Eo'Eob This paper is again applicable to the Chinese National Standard (CNS) A4 specification (2) 0 X 297 male f > 15 II a nn II If nc ^ i I tli — I n 1 ^ 1 I —OJ n ϋ —1 I 1 {Please read the notes on the back before filling this page) 419891 V. Description of the invention (Η) composition. Section 11 is a schematic circuit diagram illustrating a non-synchronous charge cycle differential logic circuit according to a fourth specific example of the present invention. As illustrated in FIG. 11, the non-synchronous charge cycle differential logic circuit includes an inversion Inverter X93 is not inverted, this round breaks into Ei, inverter X94 inverts the output k number from inverter X93, and turns out the enable signal Eo; — crossover to inverters χ91, X92 Connected between the power supply voltage Vdd and the output terminal of the inverter X93, i, and latch the output signals Q, Qb respectively; the channel of the PMOS transistor PM93 is connected between the output terminals Q, Qb, and when applied to its gate When the body enable signal is low, equalize the voltage levels of the output terminals Q and Qb; connect the logic circuit 202_5 to generate the electrical waste difference between the output terminals Q and Qb according to the input data DATA IN; and the NMOS transistor The NM93 is connected to the ground logic circuit 202-5 when the enable signal Ei is high. The inverters X91 and X92 are composed of a complementary transistor pair connected in series with a transistor (pM91, NM91) (PM92, NM92) between the power supply voltage Vdd and the enable input signal Ei. Fig. 13 is a schematic circuit diagram illustrating a non-synchronous charge cycle differential logic circuit according to a fifth specific example of the present invention. As shown in this figure, the asynchronous charge cycle differential logic circuit includes: an inverter composed of a transistor PM104 and NM104 connected in series between the power supply voltage Vdd and the ground voltage Vss, and each of them is received by its gate body. Can input the signal Ei and the completion signal Dni from the previous stage, and output the enable signal Ei; an inverter χΐ〇3 inverts the enable signal Ei 'and outputs the enable signal Eo; the individual inverter Xioi X10 is cross-coupled between the power supply voltage Vdd and the enable signal Ei, and the individual latches output the paper. The paper is suitable for the China National Standard (CNS) A4 specification (210 X 297 mm):-installed --- ( Please ^ -Read the main instructions on the back and fill in this page) Order:-Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 16 A7 413S91 — V. Description of the invention (Η) Signal Q, Qb; — PMOS The channel of the crystal PM103 is connected between the output terminals Q and Qb, and when the enable input signal Ei is low, the output terminals Q and Qb are equalized; the signal output unit 202-7 is connected between the ground and the output terminals q and Qb. 'Receive enable signal E' and output signals Q, Qb, and output as Industry completion signal Dno: —series logic circuit 202-6 outputs data signals to output terminals q and Qb according to variable data input DATAIN; and NMOS transistor NM105 connects series logic when the enable turn-on signal Ei is high Circuit 202-6 is grounded. Here, at the signal output unit 202-7, the enable input signal Ei is added to the gate of the NMOS transistor NM103, the source of the NM103 is grounded, and the output wave Q is applied to the gate of the P MOS transistor PM 105. The source is externally applied with the output signal Qb, and the output signal Qb is externally applied to the gate of the PMOS transistor PM106. The source of the PM106 is externally applied with the output signal Q. Thus, the transistor NM103, PM105, and PM106 are used in common. The job completion signal Dno is output. The inverters XI01 and XI02 are formed by connecting individual complementary transistor pairs (PM1 (H, NM101) (PM102, NM102) in series) between the power supply voltage Vcίd and the enable signal Ei '. As explained in Section 8 (a), the latch Lock block 203 and confirm that the signal ACKIN is applied to the gate of the PMOS transistor PM72, the source of PM72 is connected to the power supply voltage Vdd, and also to the gate of the NMOS transistor NM72, and the source of the NM72 is grounded. Input data signals Din and The output signal OUT or —0ί7Γ from the function block 202 is applied to the gate of the NMOS transistor NM73, the source of the NM73 is connected to the drain of the NMOS transistor NM72, and the paper size applies the Chinese National Standard (CNS) A4 (210 X 297 mm) I—FII · I- ί t, I n tlr i-nn EI nt— nnn tf f I— k 1 1 (Jing-it—read on the back of the book; I will fill in this page if you want to know more) Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs, printed by the employee 17 Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs, printed by the cooperative 419391 A7 B7 V. Description of the invention (u) The enabling signal E0 from the function Fanghun 202 is added to the PMOS transistor The gate of PM7 1 and the source of PM71 are connected to pm〇S power. The body of PM72 is also added to the gate of NMOS transistor NM71. The source of NM71 is connected to the pole of NMOS transistor NM73. The common drain node of transistor pm71 and NM71 is connected to latch 203 -1 input terminals, of which a pair of inverters X72 'X71 are connected in parallel to invert the output signal out, and a delay circuit 203-2 is provided for the output signal Eo' from the function block 202 The request signal REQOUT is generated for the next stage. The operation and effect of the asynchronous pipeline configuration according to the present invention will now be described with reference to Figures 14A to 14D. The request signal REQOUT of the second stage in a state is at a low level when it comes from When the request signal REQIN in the previous stage is at the high level, the output signal Cki or Ei from the control block 201 becomes high. As a result, the function block 202 enters the evaluation period 'evaluation output value' and the enable signal generated at the high level Eo. Here, the enabling signal Eo is output as the acknowledgement signal ACKOUT of the previous stage. Then the 'latching block 203 operates the output data from the function block 202' The delayed enabling signal Eo undergoes a predetermined period And the request signal REQOUT is generated for the next stage, so that the communication latch period operation is completed. Here, when the request signal REQ0UT of the next stage is at a high level and a predetermined period of time elapses, the output signal from the control block 201 CKi or Ei goes low 'and function block 202 returns to the precharge period. In this way, the output terminals Q and Qb of function block 202 are equal to "i / 2Vda". This paper size applies to the national standard (CNS) A4 (210 X297 male cage) 18 I .---------- ^ I install -------- order · -------- line-(Please read the precautions on the back before filling in this page) 41SS91 A7 B7 V. Description of the invention (16) and function block 202 After a predetermined time has elapsed, a low-level enable signal Eo is output. (Please read the cautions on the back before filling this page) In other words, the asynchronous system with pipeline configuration according to the present invention repeats the foregoing process and data operations through the sequence illustrated in Figures 14A to 14D. As a result, the latch block 203 latches out the data 'from the function block 202 and transmits the data to the next stage. On the other hand, the data from the function block 202 is transmitted to the next stage through the latch block 203. The latch block 203 includes the transistor PM71, NM71-NM73, the latch 203-1 and the delay circuit 203-2, as shown in FIG. 8 Instructions. In other words, 'when the confirmation signal ACKIN and the enable signal Eo are at a low level', the high level signal is added to the latch 203_1, so the output data signal DATAOUT is output at a low level. "The confirmation signal ACKIn and the enable signal Eo is in a high level state. When the input data signal Din goes high, the low level k number plus the latch 203-1 'so as to maintain a low level state before the output data signal DATAOUT. In the case where the acknowledgement signal ACKIN and the enable signal are at different levels, the latch 203-1 maintains the previous output level. Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs of the Consumer Cooperative Cooperative Association * '1 The delay circuit 203-2 receives the enable signal E0, delays the enable signal £ 0 after a predetermined time, and outputs a sub-phase request signal REQ〇UT. In addition, the function block 202 is composed of a non-synchronized charge-cycle differential logic circuit. The operation of the function block 202 according to a specific example of the present invention will now be explained. 1 * First, the operation of the function block 202 according to the first specific embodiment of the present invention for executing the circuit shown in FIG. 6 will be described. This paper has been scaled to the Chinese National Standard (CNS) A4 (21,297 mm). 19 A7 41GS91 ____B7_____ 5. Description of the invention (Π) When the enable signal Ei from the control block 201 is at a low level, at that time When the pulse signal CKi is at a high level and the clock signal CKib is at a low level, the function block 202 enters a precharge period. The high-level clock signal cKi is inverted by the inverter X53, and thus output as the low-level clock signal CKb. The low-level clock signal CKib is inverted by the inverter X54, so that the output is a high-level clock signal CK. Therefore, the clock signals CK and CKb are at the high and low levels, respectively. Therefore, the cross-coupled inverters X51 and X52 are maintained in the off state with respect to the output terminals Q and Qb. The NMOS transistor NM53, which is an equivalent transistor, is turned on by the high-level clock signal CK, and then the voltage levels of the output terminals Q and Qb are precharged to "l / 2Vda". Here, the enable signal Ei from the control block 201 is at a high level, and the NMOS transistor NM54 is turned on. The logic circuit 202-1 is connected in series. The output terminal q and Qb are generated by performing a logical operation on the input data DATAIN. The voltage difference. The estimated output terminal q is at a high level. Subsequently, the function block 202 of "clock signal CKi changes from high to low" and clock signal CKib changes from low to high "enters the evaluation period. The inverter χ53 inverts the low-level clock signal CKi and outputs the south-level clock signal CKb. The inverter X54 inverts the high-level clock signal CKib 'and outputs a low-level clock signal ck. Therefore, the inverters X5 1, X52 are enabled 'so as to evaluate the data values of the output terminals Q, Qb at a high speed. Since the output terminal Q is estimated to be at a high level, the PMOS transistor PM51 is disconnected and the NMOS transistor NM51 is turned on, so that the inverter X51 pulls down the output terminal Qb to a low level. In addition, the pM〇s transistor PM52 is turned on, and the NMOS transistor NM52 is disconnected. 'So the paper size applies to the Chinese national standard (CN Fiction J specifications (210 x 297 mm > t -------- ---- I ^ -------- ^ --------- 1 (Please read the notes on the back before filling in this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs ^ 20 —41 Coffee 1_: ___ 5. Description of the invention (18) The phaser X52 pulls up the output terminal Q to a high level. (Please read the precautions on the back before filling this page) The individual levels of the output terminals Q and Qb are based on the foregoing. The job is transmitted to the next stage, and as a result, the clock signals CK and CKb are also transmitted as the clock signals CKi and CKib. Thereafter, when the clock signal CKi changes from low to high and the clock signal CKib changes from high to low, the function block 202 returns to the precharge period. As a result, the high-level clock signal CKi is inverted by the inverter X53, and is rotated out as the low-level clock signal CKb. In addition, the 'low-level clock signal CKib is inverted by the inverter χ54 The phase and output are used as the high-level clock signal CK. Because the clock signals CK and CKb are at the high and low levels, respectively, the cross-coupled inverter X51 'X52 The output terminals q and Qb are disconnected, and the NMOS transistor NM53 is turned on in response to the high-level clock signal CK, thereby connecting the output terminals Q and Qb. As a result, the output terminals Q and Qb are precharged to "l / "2Vda". In other words, 'the precharge period and the evaluation period are repeated according to the level changes of the clock signals CKi and CKib, so that the input data is sequentially transmitted to the next stage. The Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs printed the first The operation of a specific example in the phase of the function block 202 is extremely effective only with a short delay. Now referring to FIG. 7, the operation of the function block 202 of the circuit of FIG. 7 according to the second example of the present invention will be described. The circuit operation is very similar to the circuit in Fig. 6, except for the enabling sequence of the cross-coupling inverters X61 and X62. Taking the circuit in Fig. 7 as an example, the clock signal CKi is changed to a high level 21 paper standards Applicable to China National Standard (CNS) A4 specification (210x 297 mm) 419891 A7 B7 V. Description of the invention (19) In the state, the cross-coupled inverters X61 and X62 of the current terminal are not enabled until the operation of the previous stage is completed and The enable input signal Ei becomes active. In the asynchronous charge cycle differential logic circuit according to the second specific example of the present invention, when the operation of one stage of the function block 202 is delayed for a relatively long time, the disadvantages caused by the previous evaluation may be Overcome. The operation details will now be explained. 0 When the clock signal Cki is at a high level and the clock signal CKib is at a low level, the enable input signal Ε is at a high level and the input enable signal Eib is at a low level. Bits, and inverters X61, X62 are disabled. At this time, the NMOS transistor NM64 is turned on by the high-level clock signal CCl, and the clock terminal CKb is at a low level. The PMOS transistor PM65 is turned on by the low-level clock signal CKib, and the clock signal CK is at a high level. The clock signals CKb and CK are at the low and high levels respectively, so the cross-coupled inverters X61 and X62 are maintained in the off state. The NMOS transistor NM63 is turned on by the high-level clock signal CK, and the output terminals Q and Qb are precharged to "l / 2Vda". The clock signal CKb is at a low level, so in the signal output unit 202-3, the PMOS transistor PM63 is turned on and the enable signal Eb is output at a high level. The inverter X63 is applied with a high-level enable signal Eob, and the inverter X63 outputs a low-level enable input signal Eo. When the clock signal CKi changes from high to low and the clock signal CKib changes from low to high, if the enable input signal Ei changes from high to low and the input enable signal Eib changes from low to high, the NMOS transistor NM64 and the PMOS transistor PM65 is disconnected, and the paper sizes of PMOS transistor PM64 and NMOS transistor NM65 are applicable to the national standard (CNS) A4 specification (210x 297 g t) I If I— n I— nn ^ 1 *. ^ 1 n 1 JT ·-rf i ί iin 04 ann I— nnn I ^ > 1 (Please read the note on the back; please fill in this page for matters) The Intellectual Property Bureau, Ministry of Economic Affairs, Employee Consumer Cooperatives, Printed Agriculture 22 A7 419S91 B7 V. Invention Explanation (20) Since then it was turned on. (Please read the note on the back first) and then fill out this page) So, the inverted clock signal CKb is at the high level instead of the inverted clock signal CK at the low level. ”As a result, the cross-coupled inverter X61, X62 is put into operation as far as the output terminals Q and Qb are concerned. At this time, when the NMOS transistor NM68 is added to its gate body by the input enable signal Eib and turned on, the series logic circuit 202-2 generates a voltage difference between the wheel output terminals Q and Qb according to the input data DATAIN. Here, for example, it is estimated that the output terminal Q is at a high level. In this way, the PMOS transistor PM61 is disconnected and the NMOS transistor NM61 is turned on by applying a high level signal from the output terminal Q. Therefore, the inverter X61 pulls down the output terminal Qb at a high speed. On the other hand, the PMOS transistor PM62 is turned on, and the NMOS transistor NM62 is disconnected by applying a low level of the output terminal Qb. As a result, the inverter X62 pulls up the level of the output terminal Q at a high speed. Here, in the signal output unit 202-3, the PMOS transistor PM63 is disconnected by applying a high-level clock signal CKb. However, in the above example, it is estimated that the output terminal Q is at a high level, so that the NMOS transistor NM66 is used. Turn on and enable signal Eb goes low. The Ministry of Economic Affairs knows that the Shui Property Bureau employee consumer cooperation Du printed garments So, the inverter X63 reverses the enable signal Eob at a low level, so that the enable signal Eo becomes high. Then, when the clock signal CKi changes from low to high and the clock signal CKib changes from high to low, if the input enable signal Ei transitions from low to high and the input enable signal Eib transitions from high to low, the cross-coupling inverter X61 X62 and X62 are de-energized, and NMOS transistor NM63 is turned on, so the output terminals Q and Qb are again 23. This paper size is applicable to China National Standard (CNS) A4 (210 x 297 mm) A7 419391 B7 _ V. Description of the invention ( 21) Pre-charge to "l / 2Vda". The month's operation is repeated when the levels of the signals Ej / Eib and CKi / CKib generated by the control block 201 are changed. The operation and effect of executing the functional block 202 of the circuit shown in Fig. 9 according to the third specific example of the present invention will now be described. When the inverted input enable signal E ib is at a high level and the non-inverted input enable signal Ei is at a low level, during the equalization period, the high level turn-on enable signal Eib is inverted The inverter X83 is inverted and rotated out as the low-level enable signal E0, and the low-level input enable signal Ei is inverted by the inverter χ84 and output as the south-level enable signal Eob. In this way, the enabling signals Eo and Eob are respectively at the low and high levels, so the cross-consumer inverters X81 and X82 are maintained in the off state with respect to the output terminals Q and Qb. The NMOS transistor NM83, which is an equivalent transistor, is turned on by applying a high-level input enable signal Eib, so the voltage level of the pre-charge output terminals Q and Qb reaches "l / 2Vda". Subsequently, when the input enable signal Eib changes from high to low and the input enable signal Ei changes from low to high, the enable signal Eo goes high and the enable signal Eob goes low, so the sensing / evaluation period starts. At this time, 'in the state where the NM0S transistor NM83 is enabled by an external low level input, the Eib is disconnected, and the NMOS electric body NM84 is enabled by the external high level input enable signal Ei'. Connect the logic circuit 202-4 to the input data. Perform a logic operation to thereby generate a voltage difference between the output terminals Q and Qb. Here, for example, it is estimated that the output terminal Qb is discharged to the ground potential, and thus becomes low. Therefore, the inverters X81 and X82 use the high-speed evaluation output terminals q and Qb of this paper to comply with the Chinese National Standard (CNS) A4 specification (210 x 297 mm t) i ″ ---- II I — I l · I * III I --- «— tilt-- (Please read the notes on the back and fill in this page again) Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs? Clothing 24 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 41999i A7 _________B7 _ V. Voltage difference of invention description (22). The previous example assumes that the level of the output terminal Q is high. Therefore, the pM0S transistor PM81 is disconnected and the NMOS transistor NM81 is turned on, so that the inverter X81 pulls down the output terminal Qb to a low level. In addition, the pM0s transistor PM82 is turned on, and the NMOS transistor NM82 is disconnected. In this way, the inverter χ82 pulls up the output terminal Q to a high level. The levels of the output terminals Q and Qb are transmitted to the next stage according to the foregoing operation. The enabling signals 彳 § Eo and Eob are also transmitted as input enabling signals Ei and Eib in the next stage. Subsequently, the input enable signal Ei is changed from high to low, and the enable signal Eib is changed from low to high ', thus starting the equalization period. At this time, the inverter χ83 inverts the high-level input enable signal Eib and outputs a low-level input enable signal E0, and the inverter χ84 inverts the low-level enable signal Ei and outputs a high-level enable signal. E〇b. In this way, the enable signals Eob and Eo are respectively at the high and low levels, so the parent fork-coupled inverters X81 and X82 enter the off state. : The NTMOS transistor NM83 is turned on by the high-level input enable signal Eib ′ and connected to the output terminal q ′ Qb. As a result, the output terminals Q, Qb are precharged to r 1 / 2Vda ”. In other words, according to changes in the input enable signals Ei and Eib levels, the equalization period and the sensing / evaluation period are repeatedly exchanged, so that the input data DATAIN is sequentially transmitted to the next stage. As shown in FIG. 10, the three-bit chain connection configuration of the functional blocks can be implemented by serially connecting the functional block circuits 211-213, and the functional block circuits 211-213 each execute the circuits shown in FIG. 9 to perform the foregoing operations. The operation and effect of executing the functional block 202 of the circuit shown in Fig. 1 丨 according to the fourth specific example of the present invention will now be explained. --- I .----------, · ^ -------- Order --------- line (This page) The paper size applies to the Central Family Standard (CNS) A4 specification (2ιυ x 297 mm) 25 Printed by the Employees ’Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 419991 at _ B7___ V. Description of the invention (23) Enter The energy signal Ei is in the lower order and during the evaluation period. The inverters χ93 and X94 sequentially invert the low-level input enable signal and output the low-level enable signal Eo. At this time, the output signal Ei from the inverter X93 is at a high level, so that the inverter X91 X92 cross-connects with respect to output terminals Q and Qb and forms a pull-up / pull-down circuit into the off state. The PMOS transistor PM93 is turned on by the level input enable signal Ei ', so the level of the pre-charge output terminals q, Qb reaches "l / 2Vda". CMIS inverter X9: 'The individual PMOS transistors PM91 and PM92 of X92 are directly connected to the power supply voltage Vdd', so the pre-charge current is located slightly south of this circuit than the circuit in Fig. 9. Then, when the input enable signal Ei changes from low to high, the sensing / evaluation period starts. At this time, in a state where the PMOS transistor PM93 is disconnected and the NMOS transistor NM93 is turned on by an external high-level enable signal Ei, the serial logic circuit 202-5 performs a logical operation on the data DATAIN, thereby generating an output terminal. The voltage difference between Q and Qb. Here, for example, it is estimated that the output terminal sand is discharged to the ground potential and becomes low. Therefore, the inverters X91 and X92 evaluate the voltage difference between the output terminals q and Qb at high speed. In the above example, the level of the output terminal Q is estimated to be high. Therefore, the PMOS transistor PM91 is disconnected, and the NMOS transistor > ^ 91 is turned on, so that the inverter X91 pulls down the output terminal Qb to a low level. In addition, the pM0s transistor PM92 is turned on, and the NMOS transistor NM92 is disconnected, so that the inverter X92 pulls up the output terminal Q to a high level. This paper is again applicable to China National Standard (CNS) A4 specifications (210 297 mm) (Please read the notice on the back before filling out this page) -· Line 26 419891 A7 __— _ B7 V. Description of the invention (24) According to the aforementioned operation, the level of the output terminal Q'Qb is transferred to another stage. The enable signal E0 is also transmitted as an input enable signal of the next stage. Then, the input 'causes this number Ei to change from high to low, and the evaluation period starts. At this time, the inverters X93 and X94 sequentially invert the low-level input enable signal m, and turn out the low-level enable signal Eo. Therefore, the 'enable k' Eo is at a low level and the cross-consumption inverter 1, X92 is disabled. The PMOS transistor PM93 is enabled by an external low level input " is number Ei is turned on and connected to the output terminals Q, Qb. As a result, the round-out terminals Q, Qb are precharged to about Fi / 2Vda ". In other words, when the level of the input enable signal Ei is changed, the equalization period and the sensing / evaluation period are repeatedly exchanged, thereby sequentially transmitting the input data D ATAIN to another stage. "As shown in Figure 12" Function Block 5 The bit link chain configuration can be implemented through the data terminals of the serial connection function block circuit 22 and 225. 221-225 each executes the circuit shown in (11) for the aforementioned operation. 6 In this case, the input enable signal Ei has a predetermined delay. Time, and is transmitted to the stage after the next stage. For example, the input enable signal of the 'circuit 221 passes through a delay circuit composed of a pair of inverters connected in series, and is input as the enable signal Ei of the circuit 223. As a result, the enable signal Eo and the data signal Q are obtained. Finally, the effect of performing the operation of the function block 202 of the circuit shown in Fig. 13 according to the fifth specific example of the present invention will be described. When the input enable signal Ei is at a low level and the paper size from the previous stage is applicable to the Chinese National Standard (CNS) A4 specification (210 ^ 297 male f) 27-(Please read the precautions on the back before filling (This page) r Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, India ’s Consumer Cooperatives. Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs. 1 Printed by the Consumers ’Cooperative 419891 A7. At this time, the PM104 transistor PM104 is turned on and the 'enable signal Ei' becomes high, so that the inverter χιοι 'XI02 is in an off state. At this time, the 'enable signal Ei' is at a high level, so that the signal output unit 202-7 'NMOS transistor NM103 is turned on, and a low-level operation completion signal Dno is output. The inverter XI03 receives the high-level enable signal Ei, and outputs a low-level enable signal Eo. The PMOS transistor PM103 is enabled by an external low-level input enable signal, so that the output terminals Q and Qb are precharged to "i / 2Vda". Thereafter, when the input enable signal Ei transitions from low to high, the pMOS transistor PM103 and PM104 are turned on, the output terminals q and Qb are disconnected from each other 'and the enable signal Ei' is in a high impedance state. At this time, in a state where the NMOS transistor NM105 is turned on by an applied high-level enable signal, the serial logic circuit 202-6 generates a voltage difference between the output terminals Q and Qb by performing an operation on the input data DATAIN. Here, for example, it is estimated that the output terminal Qb becomes high. > When the operation completion signal Dni from the previous stage goes high, the NMOS transistor NM104 is turned on and the enable signal Ei becomes low. As a result, the cross-coupled inverters XI01 and X012 are activated. The inverter X〗 03 inverts the low-level enable signal Ei '' so that the enable signal Eo becomes high. Therefore, according to the high level of the output terminal Q, the PMOS transistor PM101 is disconnected, and the NMOS transistor NM101 is turned on. In this way, the inverter x1 pulls up the level of the output terminal Qb at high speed. According to the low level of the output terminal Qb, the PMOS transistor PM102 can be turned on. The NMOS transistor NM102 is broken. The paper size applies the Chinese National Standard (CNS) A4 specification (210x 297 meals). 28-------- r I -------- ^ -------- 丨 line (please read the Jiang Yi matters on the back before filling out this page) 419891 A7 —… __B? ____ V. Description of the invention (26) In this way, the inverter X102 pulls up the level of the output terminal q at a high speed. (Please read the back of this page first; and then fill in this page.) In addition, at the signal output unit 202-7, because the enable signal Ei is at a low level, the NMOS transistor Zhao NM103 is disconnected. However, the output terminal q is at the upper level 'and the output terminal Qb is at the lower level, so that the pMOS transistor PM106 is turned on, thereby outputting a high-level job completion signal Dn0. Subsequently, when the input enable signal Ei transitions from high to low, if the job completion signal Dni from the previous stage changes from high to low, and thus the PMOS transistor PM104 is turned on, the enable signal Ei becomes high. In this way, the AC and AC inverters 101 and 102 are deenergized, and the 卩 1 ^ 〇8 transistor? 1 ^ 103 is turned on, thereby pre-charging the output terminals Q and Qb again to "〖/ 2Vda". Whenever the transition from the control block 201 to the level of the input enable signal Ei of the function block 202 is transitioned, the foregoing operations are repeatedly performed. According to the foregoing specific example of the present invention, the serial logic circuits 202-2, 202-2 '202-4, 202-5, 202-6 may be replaced by complementary output conduction crystal logic networks. Data is input to output terminals Q and Qb to generate data.

經濟部智慧財產局員工湞費合作社印M 如前文討論’與同步系統不同,本發明之非同步系統 未採用通用時脈信號,如此並無時脈歪斜問題。此外,本 發明之非同步系統係藉由採用局部控制信號執行方塊間的 通訊,結果可減少時脈分散。 當使用習知差異串接電壓開關(DCVS)邏輯時,功率 顯著被消耗。但本發明之非同步系統具有功率有效作業特 性,如此可減少功率消耗。 根據本發明無須控制交叉耦合反相器對之PMOS電晶 29 本紙張尺度遇用中國國家標準(CNS) A4規格_(2〗0 X 297公餐) 419991 i、發明說明(27 ) 體之閾值電壓。因此無須高電壓來控制閾值電壓,結果獲 得高速處理而僅消耗小量功率。此外容易應用本發明之非 同步配置至數位設計而可降低成本。 因本發明可以若干形式具體表現而未背離及精髓或要 義特徵,也需瞭解前述具體例除非另行規定否則絕非受任 何前文說明之細節所限,而需視為廣義落入隨附之申請專 利耗圍界疋之精越及範圍内,因此全部屬於申請專利範圍 之要求及界線以内之變化及修改或其相當例意圖皆涵蓋於 隨附之申請專利範圍。 --.------------ '衣.-------訂---------線 (請先閲讀背面之注音事項再填寫本頁) 經濟部智慧財產局員工消費合作社印制π 30 本紙張尺度適用中國國家標準(CNS)A4規格(2】ϋ X 297公釐) 419S91 A7 B7 _;_ 五、發明說明(28 ) 元件標號對照 (請先閱讀背面之-1¾事項再填寫本頁) 101…控制方塊 102…功能方塊 103…完成偵測器 104···閂鎖方塊 104-1…閂鎖 2 01…控制方塊 201- 1…串級邏輯電路 202…功能方塊 202_1~2,4~6···串級邏輯電路 202- 3…信號輸出單元 203…閂鎖方塊 203- 1…閂鎖 經濟部智慧財產局員工消費合作社印製 211〜3…功能方塊電路 221〜5…功能方塊電路 CK…時脈信號 Ei…輸入致能信號 Ei’…輸出信號 £〇···低階致能信號 NM…NMOS電晶體 PM…PMOS電晶體 Q…輸出端子 Vdd···電源電壓 X-··反相器 本紙張尺度遇用中國國家標準(CNS)A4規格(210 x 297公釐)As discussed earlier, unlike the synchronous system, the asynchronous system of the present invention does not use a universal clock signal, so there is no clock skew problem. In addition, the asynchronous system of the present invention performs communication between blocks by using local control signals, and as a result, the clock dispersion can be reduced. When using conventional differential series voltage switch (DCVS) logic, power is significantly consumed. However, the non-synchronous system of the present invention has power-efficient operation characteristics, which can reduce power consumption. According to the present invention, it is not necessary to control the PMOS transistor of the cross-coupled inverter pair. 29 This paper size meets the Chinese National Standard (CNS) A4 specification_ (2〗 0 X 297 meals) 419991 i. The invention's description (27) body threshold Voltage. Therefore, no high voltage is required to control the threshold voltage, and as a result, high-speed processing is achieved without consuming a small amount of power. In addition, it is easy to apply the asynchronous configuration of the present invention to a digital design and the cost can be reduced. Because the present invention can be embodied in several forms without departing from the essence or essential features, it is also necessary to understand that the foregoing specific examples are not limited by any of the details described above unless otherwise specified, and need to be regarded as falling broadly into the accompanying patent application Consumption is within the scope and scope of the boundary, so all the requirements and limits within the scope of patent application and the changes and modifications within the boundary or its equivalent are intended to be included in the scope of the accompanying patent application. --.------------ 'yi .------- order --------- line (please read the note on the back before filling this page) Economy Printed by the Intellectual Property Cooperative of the Ministry of Intellectual Property, π 30 This paper size is applicable to the Chinese National Standard (CNS) A4 specification (2) ϋ X 297 mm) 419S91 A7 B7 _; _ 5. Description of the invention (28) Component reference comparison (please Read the -1¾ item on the back before filling in this page) 101 ... Control block 102 ... Function block 103 ... Complete detector 104 ... Latch block 104-1 ... Latch 2 01 ... Control block 201-1 ... Cascade Logic circuit 202 ... Function block 202_1 ~ 2, 4 ~ 6 ... Cascaded logic circuit 202-3 ... Signal output unit 203 ... Latch block 203-1 ... Latch Printed by employee consumer cooperative of Intellectual Property Bureau, Ministry of Economic Affairs 211 ~ 3 ... functional block circuit 221 ~ 5 ... functional block circuit CK ... clock signal Ei ... input enable signal Ei '... output signal £ 〇 ··· low-order enable signal NM ... NMOS transistor PM ... PMOS transistor Q ... Output terminal Vdd ··· Supply voltage X- ·· Inverter This paper size meets China National Standard (CNS) A4 specification (210 x 297 mm)

Claims (1)

ABCD 419S91 ~、申請專利範圍 1. 一種非同步感測差動邏輯電路,包含: 一控制方塊用於對來自前一階段之請求信號及對 次一階段之請求信號進行邏輯運算,及用於根據該運 算輪出第一或第二輸入致能信號及第一或第二時脈信 號; 一功能方塊用於根據來自控制方塊之第一或第二 輸入致能信號及第一或第二時脈信號對輸入資料進行 運算,及用於輸出第一或第二輸出致能信號及輸出資 料;以及 一閂鎖方塊,其係由來自次一階段之確認信號觸 發且經由對來自功能方塊之第一或第二輸出致能信號 及輸出資料執行運算而輸出對次一階段之請求信號及 一最終輸出資料。 2.如申請專利範圍第1項之電路,其中該第一及第二輸入 致1½號、第一及第二時脈信號,及第一及第二輪出 致能信號分別之相位相反。 3_如申請專利範圍第1項之電路,其中該功能方塊包含: 一第一反相器用於反相來自控制方塊之第—時脈 信號,及用於輸出一第三時脈信號; 一第二反相器用於反相來自控制方塊之第二時脈 信號’及用於輸出一第四時脈信號; 一第一PMOS電晶體,具有其源極連結至接收第 三時脈信號,其汲極連结至第一輪出節點,及其間體 連結至第二輪出節點; (請先閱讀背面之注意事項再填寫本萸) *1T -峡Ί 經濟部中央標率局貝工消費合作社印Τ 經濟部中央樣準局負工消費合作衽印製 A8 训m_I 六、申請專利範圍 ~ — -第二PMQS電晶體,具有其源極連結至接收第 三時脈信號’其汲極連結至第二輸出節點,及其閘體 連結至第一輸出節點; -第- NMOS電晶豸,具有其源極連結至接收第 四時脈信號,其沒極連結至第一輸出節點,及其間體 連結至第二輸出節點; 一第二NMOS電晶體,具有其源極連結至接收第 四時脈信號,其汲極連結至第二輸出節點,及其閘體 連結至第一輸出節點; 一第二NMOS電晶體,具有其源極及汲極分別連 結至第二及第四電晶體個別之閘極,且個別交替連結 至第一及第二輪出節點,且其閘體係連結至接收第四 時脈信號; 一串接邏輯電路用於根據輸入資料初始化第一及 第二輸出節點;以及 一第四NMOS電晶體用於根據來自控制方塊之第 輸入致能信號接地串接邏輯電路之内部切換元件。 4.如申請專利範圍第3項之電路,其中該功能方塊進一步 包含一輸出致能信號產生單元包括: 一第二PMOS電晶體,具有其源極連結至接收功 率電壓,其汲極連結至第一輸出致能信號輸出端子, 及其閘體連結至接收第三時脈信號;及 第五及第六NMOS電晶體,具有其汲極連結至第 二PMOS電晶體之汲極,也連結至接收第一輸出致能 本紙張尺度適财關家制^ ( CNS ) Α视ίΜ 2】GX297公釐) 33 (請先閲讀背面之注意事項再填寫本頁) 訂 經濟部中央標隼局負工消費合作社印« 4iGS9.i 錳 C8 —^一 ______ D8 中請專利範圍 信號輸出端子,其個別源極係連結至第一及第二輸出 即點,及其個別閘體係交替連結至第—及第二輪出節 點。 5.如申請專利範圍第3項之電路,其中該第一反相器包含 一第四PMOS電晶體具有其源極連結至接收電源 電壓,及其閘體連結至接收第一輸入致能信號;及 一第七NM0S電晶體具有其汲極連結至第一電晶 體之汲極,其源極連結至接收地電壓,及其閘體連結 至接收第一時脈信號;以及 第二反相器包含: 一第五PMOS電晶體具有其源極連結至接收功率 電壓,及其閘體連結至接收第二時脈信號;以及 一第八NMOS電晶體具有其汲極連結至接收第五 電晶體之汲極,其源極連結至接收地電壓,及其閘體 連結至接收第二輸入致能信號。 6,如申請專利範圍第3項之電路,其中該串接邏輯電路係 以導通電晶體邏輯_路置換,該網路可根據可變資料 輸出而對第一及第二輸出節點產生資料。 7.如申請專利範圍第〗項之電路,其中該功能方塊包含: 第一及第二輸出節點; —第一反相器用於反相來自控制方塊之第—輸入 致能信號,及用於輸出一第一輸出致能信號; —第二反相器用於反相來自控制方塊之第二輸入 本紙張( 2iex297H ) 34 (請先閱讀背面之注^^項再填寫本頁) 、1T Λ 經濟部中央橾準局貝工消費合作祍印製 419S91 韶 C8 D8 申請專利範圍 致能信號,及用於輸出一第二輸出致能信號; 一第一 PMOS電晶體具有其源極連結至接收第— 輸出致能信號,其汲極連結至第一輸出節點,及其問 體連結至第二輸出節點; —第一 PMOS電晶艘具有其源極連結至接收第一 輸出致能信號,其汲極連結至第二輸出節點,及其閘 體連結至第一輸出節點; 一第一 NMOS電晶體具有其源極連結至接收第二 輸出致能信號’其汲極連結至第一輪出節點,及其閘 體連結至第二輸出節點; 一苐一 NMO S電晶體具有其源極連結至接收第二 輸出致能信號’其汲極連結至第二輸出節點,及其閘 體連結至第一輸出節點; 一第二NMOS電晶體,具有其源極及沒極分別連 、名。至第二及第四電晶體個別之閉極,及分別交替連結 至第一及第二輸出節點,且具有其閘體連結至接收來 自控制方塊之第二輸入致能信號; 一串接邏輯電路用於根據輸入資料初始化第一及 第二輸出節點;以及 一第四NMOS電晶體用於根據來自控制方域之第 一輸入致能信號接地串接邏輯之内部切換元件。 8.如申凊專利範圍第7項之電路,其中該串接邏輯電路係 以導通電晶體邏輯網路置換,該網路可根據可變資料 輸出而對第一及第二輸出節點產生資料。 私紙張尺度適用中國國家橾準(CNS ) M规格(210><297公釐) 35 -----------.1 %------,玎------f . I 4* (請先閱讀背面之注意事項再填寫本頁) 經濟部中央標丰局員工消費合作社印衮 419S91 g _____ D8 六、申請專利範圍 9‘如申請專利範圍第1項之電路,其中該功能方塊包含: 第一及第二輸出節點; —第一反相器用於反相來自控制方塊之第一輸入 致能信號’及用於輸出反相後之致能信號; 一第二反相器用於再度反相該被反相的致能信號 且用於輸出一輸出致能信號; 一第一 PMOS電晶體具肴其源極連結至接收電源 電壓,其汲極連結至第一輪出節點,及其閘體連結至 第二輸出節點; 一第一 PMOS電晶趙具有其源極連結至接收電源 電壓,其汲極連結至第二輸出節點,及其閘體連結至 第一輸出節點: 一第一 NMOS電晶體具有其源極連結至接收來自 第一反相器之被反相的致能信號,其汲極連結至第一 輸出節點,及其閘體連結至第二輸出節點; 一第二NMOS電晶體具有其源極連結至接收來自 第一反相器之被反相的致能信號,其汲極連結至第二 輸出卽點,及其閉趙連結至第一輪出節點; 一第三NMOS電晶體,具有其源極及汲極分別連 結至第三及第四電晶體個別之閘極,及分別交替連結 至第一及第二輸出節點’且具有其閘體連結至接收來 自控制方塊之第一輸入致能信號; 一串接邏輯電路用於根據輪入資料初始化第—及 苐二輸出節點:以及 本紙張尺度逍用中國國家標準(CNS ) A4規格(2】OX29?公釐) 36 1^1 ^^^1 n^i n^— n··— ml n —^ϋ ^^^1 t (請先M讀背面之注意事項再填寫本頁) 418891 A8 B8 C8 D8 麵濟部中央橾率局—工消費合作杜印¾ 、申請專利範圍 一第二NMOS電晶體用於根據第一輸入致能信號 接地串接邏輯電路之内部切換元件。 W·如申請專利範圍第9項之電路,其中該功能方塊進一步 包含一運算完成信號產生單元包括: 一第四NMOS電晶體具有其源極連結至接收地電 壓,其及極連結至運算完成信號輸出端子,及其閘體 連結至接收第一致能信號之經反相的信號;以及 第四及第五PMOS電晶體具有其個別汲極連結至 第一電晶體之汲極’且連結至運算完成信號輸出端子 ,及個別源極連結至第一及第二輸出端子,及其個別 閘體交替連結至第一及第二輸出端子,及其個別閘體 交替連結至第一及第二輸出端子。 U•如申請專利範圍第9項之電路,其中該第一反相器包含 一第六PMOS電晶體具有其源極連結至接收電源 電壓,及其閘體連結至接收第一輸入致能信號;及 一第五NMOS電晶體具有其汲極連結至第六pm〇S 電晶體之汲極,且輸出被反相的致能信號,及源極連 結至接收地電壓,及其閘體連结至接收來自前—階段 之一運算完成信號。 12.如申請專利範圍第9項之電路,其中該串接邏輯係由導 通電晶體邏輯電路置換,該電路可根據一可變資料輸 入產生資料至第一及第二輸出節點。 U.如申請專利範圍第1項之電路,其中該閃鎖方塊包含: 本紙張尺度適用中賴家標準(CNs) .(训X加公楚) 37 .J 4------1T------Ϋ, (請先閲讀背面之注意事項再填寫本頁) 419S91 ABCD 六、申請專利範圍 … 一資料輸入單元其循序事聯連結第一及第二 PMOS電晶體及第一至第三NMOS電晶體介於電源電壓 與地電壓間,接收確認信號於第一 PMOS電晶體及第 三NMOS電晶體之閘極,接收輸出致能信號於第二 PMOS電晶體及第一 NMOS電晶體之閘體,及接收來自 功能方塊之輸出資料於第二NMOS電晶體之閘體; 一閂鎖單元具有其輸入端子連結至於資料輸入單 元之第二PMOS電晶體與第一 NMOS電晶體之一共通節 點,且用於閂鎖輸入其中之一信號;以及 一延遲單元用於延遲輸出致能信號,且產生對次 —階段之請求信號。 (請先聞讀背面之注意事項再填寫本頁) 經濟部中央橾隼局員工消费合作社印策 本紙張尺度逍用中國國家標準(CNS ) A4規格(21 〇 X四7公釐) 38ABCD 419S91 ~, patent application scope 1. A non-synchronous sensing differential logic circuit, comprising: a control block for performing logical operations on the request signal from the previous stage and the request signal from the next stage, and for The operation turns out the first or second input enable signal and the first or second clock signal; a function block is used to according to the first or second input enable signal and the first or second clock from the control block The signal operates on the input data and is used to output the first or second output enable signal and output data; and a latch block, which is triggered by a confirmation signal from the next stage and passes the first from the function block. Or, the second output enable signal and the output data perform an operation to output a request signal for a next stage and a final output data. 2. The circuit of item 1 in the scope of patent application, wherein the first and second inputs are 1½, the first and second clock signals, and the first and second round enable signals are in opposite phases. 3_ The circuit of item 1 in the scope of patent application, wherein the functional block includes: a first inverter for inverting the first clock signal from the control block, and for outputting a third clock signal; a first Two inverters are used to invert the second clock signal from the control block and to output a fourth clock signal; a first PMOS transistor having its source connected to receiving a third clock signal, and its sink The pole is connected to the first round of output nodes, and the interstitial body is connected to the second round of output nodes; (please read the notes on the back before filling in this card) * 1T-printed by the shellfish consumer cooperative of the Central Standards Bureau of the Ministry of Economic Affairs Τ Printed A8 training m_I by the Central Procurement Bureau of the Ministry of Economic Affairs VI. Patent application scope ~--The second PMQS transistor has its source connected to receive the third clock signal 'its drain connected to the The two output nodes and their gates are connected to the first output node; the -th NMOS transistor has its source connected to receive the fourth clock signal, its poles connected to the first output node, and its interstitial connection To the second output node; Two NMOS transistors with their sources connected to receive the fourth clock signal, their drains connected to the second output node, and their gates connected to the first output node; a second NMOS transistor with its source And drain are respectively connected to the individual gates of the second and fourth transistors, and are individually connected to the first and second round-out nodes, and their gate systems are connected to receive the fourth clock signal; a series of logic circuits It is used to initialize the first and second output nodes according to the input data; and a fourth NMOS transistor is used to ground the internal switching elements of the logic circuit in series according to the first input enable signal from the control block. 4. The circuit according to item 3 of the patent application scope, wherein the functional block further includes an output enable signal generating unit including: a second PMOS transistor having a source connected to the receiving power voltage and a drain connected to the An output enable signal output terminal and its gate are connected to receive the third clock signal; and the fifth and sixth NMOS transistors having their drains connected to the drains of the second PMOS transistor are also connected to the receiver The first output enables the paper size of the financial system ^ (CNS) Α ί 2 2 GX297 mm) 33 (Please read the precautions on the back before filling this page) Order the work of the Central Bureau of Standards of the Ministry of Economic Affairs Cooperative cooperative seal «4iGS9.i Manganese C8 — ^ 一 ______ D8 please patent range signal output terminal, its individual source is connected to the first and second output point, and its individual gate system is alternately connected to the first-and the first Second round out of nodes. 5. The circuit according to item 3 of the patent application, wherein the first inverter includes a fourth PMOS transistor having a source connected to receiving a power supply voltage and a gate connected to receiving a first input enable signal; And a seventh NMOS transistor having a drain connected to a drain of the first transistor, a source connected to a receiving ground voltage, and a gate connected to receiving a first clock signal; and a second inverter including : A fifth PMOS transistor has its source connected to the receiving power voltage, and its gate is connected to receive the second clock signal; and an eighth NMOS transistor has its drain connected to the sink of the fifth transistor The source is connected to the receiving ground voltage and the gate is connected to receiving the second input enable signal. 6. For the circuit in the third item of the patent application, wherein the serial logic circuit is replaced by a conducting crystal logic circuit, the network can generate data for the first and second output nodes according to the variable data output. 7. The circuit according to the item in the scope of patent application, wherein the functional block includes: first and second output nodes;-the first inverter is used to invert the first input enable signal from the control block, and is used for output A first output enable signal;-A second inverter is used to invert the second input paper (2iex297H) from the control block 34 (Please read the note ^^ on the back before filling this page), 1T Λ Ministry of Economy The Central Bureau of Standards and Technology, Shellfish Consumer Cooperation, printed 419S91 Shao C8 D8 patent application scope enable signal, and used to output a second output enable signal; a first PMOS transistor has its source connected to the receiving first-output Enable signal, its drain is connected to the first output node, and its body is connected to the second output node;-the first PMOS transistor has its source connected to receive the first output enable signal, its drain connected To the second output node, and its gate body is connected to the first output node; a first NMOS transistor has its source connected to receive the second output enable signal, its drain connected to the first output node, and Brake Connected to a second output node; a NMO S transistor has its source connected to receive a second output enable signal, its drain connected to a second output node, and its gate connected to a first output node; The second NMOS transistor has its source and terminal connected and named. To the individual closed poles of the second and fourth transistors, and alternately connected to the first and second output nodes, respectively, and having a gate body connected to receive a second input enable signal from the control block; a series connected logic circuit It is used for initializing the first and second output nodes according to the input data; and a fourth NMOS transistor is used for grounding and connecting logic internal switching elements according to the first input enable signal from the control domain. 8. The circuit of claim 7 in the patent scope, wherein the series logic circuit is replaced by a conducting crystal logic network, which can generate data for the first and second output nodes according to the variable data output. Private paper standards are applicable to China National Standards (CNS) M specifications (210 > < 297 mm) 35 -----------. 1% ------, 玎 ----- -f. I 4 * (Please read the precautions on the back before filling out this page) Printed by the Consumer Standards Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs 衮 419S91 g _____ D8 VI. Patent Application Scope 9 'Circuits such as the first patent application scope The function block includes: first and second output nodes; a first inverter for inverting a first input enable signal from the control block and an output enable signal after inversion; a second The inverter is used to invert the inverted enable signal again and used to output an output enable signal. A first PMOS transistor has a source connected to the receiving power voltage and a drain connected to the first round. The output node and its gate are connected to the second output node; a first PMOS transistor has its source connected to the receiving power voltage, its drain is connected to the second output node, and its gate is connected to the first output Node: a first NMOS transistor having its source connected to receiving a signal from a first inverter The inverted enable signal has its drain connected to the first output node and its gate connected to the second output node; a second NMOS transistor has its source connected to receive the inverted signal from the first inverter. Phase enable signal, its drain is connected to the second output pin, and its closed connection is connected to the first round output node; a third NMOS transistor with its source and drain connected to the third and the third, respectively The individual gates of the four transistors are alternately connected to the first and second output nodes, respectively, and have their gates connected to receive the first input enable signal from the control block; a series of logic circuits are used for Data Initialization—the second and the second output nodes: and the Chinese paper standard (CNS) A4 specification (2) OX29? Mm) 36 1 ^ 1 ^^^ 1 n ^ in ^-ml n — ^ ϋ ^^^ 1 t (please read the precautions on the back before filling in this page) 418891 A8 B8 C8 D8 Central Government Office of the Ministry of Health—Industrial and Consumer Cooperation Du Yin ¾ 、 Patent application scope-second NMOS The transistor is used to ground the logic circuit in series according to the first input enable signal. Switching element portion. W · The circuit of item 9 in the scope of patent application, wherein the functional block further includes an operation completion signal generating unit including: a fourth NMOS transistor having its source connected to the receiving ground voltage, and its sum connected to the operation completion signal The output terminal and its gate are connected to the inverted signal receiving the first enable signal; and the fourth and fifth PMOS transistors have their individual drains connected to the drain of the first transistor and are connected to the operation Complete signal output terminals, and individual sources connected to the first and second output terminals, and their individual gates alternately connected to the first and second output terminals, and their individual gates alternately connected to the first and second output terminals . U • The circuit of item 9 in the scope of patent application, wherein the first inverter includes a sixth PMOS transistor having a source connected to receiving the power supply voltage, and a gate connected to receiving the first input enable signal; And a fifth NMOS transistor has its drain connected to the drain of the sixth pMOS transistor, and the output is an enable signal inverted, the source is connected to the receiving ground voltage, and its gate is connected to Receives an operation completion signal from one of the previous stages. 12. The circuit of item 9 in the scope of patent application, wherein the series logic is replaced by a conducting crystal logic circuit, and the circuit can generate data to the first and second output nodes according to a variable data input. U. The circuit of item 1 in the scope of patent application, wherein the flash lock block contains: This paper size is applicable to CN standards. (Training X plus public Chu) 37 .J 4 ------ 1T- ----- Ϋ, (Please read the notes on the back before filling out this page) 419S91 ABCD VI. Patent application scope ... A data input unit which connects the first and second PMOS transistors and the first to the first The three NMOS transistors are between the power supply voltage and the ground voltage. They receive a confirmation signal at the gates of the first PMOS transistor and the third NMOS transistor, and receive an output enable signal at the second PMOS transistor and the first NMOS transistor. A gate body, and a gate body that receives output data from a function block to a second NMOS transistor; a latch unit having an input terminal connected to a common node between the second PMOS transistor and the first NMOS transistor of the data input unit And is used for latching one of the signals; and a delay unit is used for delaying the output enable signal and generating a request signal for the second stage. (Please read the precautions on the back before filling out this page) Printing policy of the Employees' Cooperatives of the Central Government Bureau of the Ministry of Economic Affairs This paper standard is in accordance with Chinese National Standard (CNS) A4 specification (21 × 47mm)
TW088109736A 1998-09-01 1999-06-10 Asynchronous sensing differential logic (ASDL) circuit TW419891B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019980035876A KR100300050B1 (en) 1998-09-01 1998-09-01 Asynchronous sensing differential logic circuit

Publications (1)

Publication Number Publication Date
TW419891B true TW419891B (en) 2001-01-21

Family

ID=19549170

Family Applications (1)

Application Number Title Priority Date Filing Date
TW088109736A TW419891B (en) 1998-09-01 1999-06-10 Asynchronous sensing differential logic (ASDL) circuit

Country Status (4)

Country Link
JP (1) JP4273278B2 (en)
KR (1) KR100300050B1 (en)
DE (1) DE19929600A1 (en)
TW (1) TW419891B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6356117B1 (en) * 2000-09-29 2002-03-12 Sun Microsystems, Inc. Asynchronously controlling data transfers within a circuit
JP5107033B2 (en) * 2004-06-15 2012-12-26 ザ リージェンツ オブ ザ ユニバーシティ オブ ミシガン Energy recovery boost logic
DE102005008367B3 (en) 2005-02-23 2006-10-19 Infineon Technologies Ag XOR circuit
US8527797B2 (en) * 2007-12-26 2013-09-03 Qualcomm Incorporated System and method of leakage control in an asynchronous system
WO2012120760A1 (en) * 2011-03-09 2012-09-13 ルネサスエレクトロニクス株式会社 Semiconductor device
FR3047136B1 (en) * 2016-01-27 2018-02-16 Commissariat A L'energie Atomique Et Aux Energies Alternatives SAVING DATA IN AN ASYNCHRONOUS CIRCUIT

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3183159B2 (en) * 1996-03-29 2001-07-03 日本電気株式会社 Synchronous DRAM

Also Published As

Publication number Publication date
JP2000188538A (en) 2000-07-04
KR20000018326A (en) 2000-04-06
DE19929600A1 (en) 2000-03-02
KR100300050B1 (en) 2001-09-06
JP4273278B2 (en) 2009-06-03

Similar Documents

Publication Publication Date Title
US6211704B1 (en) Asynchronous sensing differential logic (ASDL) circuit
Kong et al. Conditional-capture flip-flop for statistical power reduction
US8742796B2 (en) Low energy flip-flops
US6633188B1 (en) Sense amplifier-based flip-flop with asynchronous set and reset
TW200903514A (en) Level-converted and clock-gated latch and sequential logic circuit having the same
US20040100307A1 (en) Circuit for asychronous reset in current mode logic circuits
Kumar et al. A new leakage-tolerant high speed comparator based domino gate for wide fan-in OR logic for low power VLSI circuits
TW419891B (en) Asynchronous sensing differential logic (ASDL) circuit
Shah et al. A glitch free variability resistant high speed and low power sense amplifier based flip flop for digital sequential circuits
US8305125B2 (en) Low latency synchronizer circuit
JP4575300B2 (en) Master latch circuit with dynamic flip-flop signal level substitution
JP4589496B2 (en) Conditional capture flip-flop for power saving
US6078196A (en) Data enabled logic circuits
CN109525222B (en) Double-edge D trigger of single-phase clock
US20050083093A1 (en) Flip-flop
Rengarajan et al. Challenges to adopting adiabatic circuits for systems‐on‐a‐chip
US8421514B1 (en) Hazard-free minimal-latency flip-flop (HFML-FF)
KR102643441B1 (en) Clock generation circuit of semiconductor device
Lin et al. High speed soft-error-tolerant latch and flip-flop design for multiple VDD circuit
CN104022758B (en) A kind of band resets the power consumption equilibrium trigger of set port
JPH10335992A (en) Semiconductor integrated circuit device
Abhishek et al. Low Power DET Flip-Flops Using C-Element
JPH06152336A (en) Double edge trigger flip flop
Kumar et al. implementation and designing of low power SR flip-flop using 45nm CMOS technology
Inguva et al. Predictive Power Control Circuit Model With T-DET-FF for Hybrid Pulsed Latch Implementation.

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees