Ternary digit logic circuit

Abstract

A ternary logic circuit according to the present invention includes a pull-up device (100) and a pull-down device (200) connected in series between power voltage sources (V.sub.DD and GND), and an input voltage (V.sub.IN) source and output voltage (V.sub.OUT) source. When both the pull-up device (100) and the pull-down device (200) are turned off by an input voltage (V.sub.IN), both the pull-up device (100) and the pull-down device (200) operate as simple resistors which are affected only by an output voltage (V.sub.OUT) and form a ternary digit (1 state) through voltage division. When only one of the pull-up device (100) or the pull-down device (200) is turned on to allow a current to flow therethrough, V.sub.DD (2 state) or GND (0 state) is output as the output voltage (V.sub.OUT). Accordingly, a bit density can be remarkably increased.

Claims

1. A ternary logic circuit comprising: a pull-up device (100) and a pull-down device (200) connected in series between power voltage sources (V.sub.DD and GND); and an input voltage (V.sub.IN) source and an output voltage (V.sub.OUT) source, wherein, when both the pull-up device (100) and the pull-down device (200) are turned off by an input voltage (V.sub.IN), both the pull-up device (100) and the pull-down device (200) operate as simple resistors which are affected only by an output voltage (V.sub.OUT) and form a ternary digit (1 state) through voltage division, and when only one of the pull-up device (100) or the pull-down device (200) is turned on to allow a current to flow therethrough, V.sub.DD (2 state) or GND (0 state) is transferred as the output voltage (V.sub.OUT), wherein the ternary logic circuit has a current (I.sub.CON) component, which is not affected by the input voltage (V.sub.IN) and is affected only by the output voltage (V.sub.OUT), and a current (I.sub.EXT) component, which is affected by the input voltage (V.sub.IN) and is not affected by the output voltage (V.sub.OUT), the current (I.sub.CON), which is affected by the output voltage (V.sub.OUT), is realized via a junction BTBT current (I.sub.BTBT) independent of a gate voltage, the current (I.sub.EXT), which is affected by the input voltage (V.sub.IN), is realized via a subthreshold current (I.sub.sub), and characteristics of the BTBT current (I.sub.BTBT) and the subthreshold current (I.sub.sub) are obtained by a simple increase in a channel doping in a binary inverter, so that the binary inverter operates as a ternary inverter.

2. The ternary logic circuit of claim 1, wherein the current (I.sub.CON), which is affected by the output voltage (V.sub.OUT), has a current value I.sub.C when the output voltage (V.sub.OUT) is half of an operating voltage (V.sub.DD), that is, (I.sub.CON)=I.sub.C when V.sub.OUT=V.sub.DD/2, and the current (I.sub.EXT), which is affected by the input voltage (V.sub.IN), has a current value I.sub.E when the input voltage (V.sub.IN) is half of the operating voltage (V.sub.DD), that is (I.sub.EXT)=I.sub.E when V.sub.IN=V.sub.DD/2, and exponentially increases to a maximum current (I.sub.MAX) at a point where the input voltage (V.sub.IN) and the operating voltage (V.sub.DD) are equal to each other, that is, when V.sub.IN=V.sub.DD.

3. The ternary logic circuit of claim 2, wherein a current output from the pull-up device (100) and the pull-down device (200) is calculated by the following equation: $\begin{array}{c}{I}_{\mathrm{OUT}}\left(V_{\mathrm{IN}},V_{\mathrm{OUT}}\right)=I_{\mathrm{CON}}\left(V_{\mathrm{OUT}}\right)+I_{\mathrm{EXP}}\left(V_{\mathrm{IN}}\right),\\ =I_C\exp \left[\left(V_{\mathrm{OUT}}-\frac{V_{\mathrm{DD}}}{2}\right)\right]+I_E\exp \left[\left(V_{\mathrm{IN}}-\frac{V_{\mathrm{DD}}}{2}\right)\right]\end{array}$ wherein, in the equation, and are exponential coefficients of each current mechanism, and + and signs before and are respectively signs applied to the pull-down device (200) and the pull-up device (100).

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 illustrates a configuration diagram of a ternary logic circuit including a pull-up device and a pull-down device, according to the present invention, and an output current-input voltage characteristic graph of the pull-up device and the pull-down device.

(2) FIG. 2 illustrates an operating principle of an STI according to output current-output voltage characteristics of the pull-up device and the pull-down device.

(3) FIG. 3 illustrates a mathematical voltage transfer curve of STI.

(4) FIG. 4 illustrates a planar 32-nm nMOS device structure and BTBT occurrence rates at a low channel doping and a high channel doping.

(5) FIG. 5 illustrates output current-input voltage characteristics of ternary CMOS and a voltage transfer curve of STI according to a channel doping.

(6) FIG. 6 illustrates a comparison between simulated current-voltage data and modeled current-voltage based on [Equation 1].

(7) FIG. 7 illustrates voltage transfer characteristics and static noise margin analysis.

(8) FIG. 8 illustrates circuit configurations and truth tables of ternary inverter, ternary minimum (MIN), and ternary maximum (MAX) gates.

BEST MODE

(9) Hereinafter, embodiments of the present invention will be described in more detail with reference to the accompanying drawings. Terms and words used in the present specification and the claims should not be construed as limited to ordinary or dictionary terms, and should be construed in accordance with the meaning and concept consistent with the technical idea of the present invention based on the principle the inventors can properly define the concept of the terms so as to describe their invention in the best way.

(10) Therefore, the embodiments described in the present specification and the configurations illustrated in the drawings are only the most preferred embodiments of the present invention and do not represent all the technical ideas of the present invention. Therefore, it should be understood that various equivalents and modifications may be substituted for those at the time of the present application.

(11) FIG. 1 illustrates a configuration diagram of a ternary logic circuit including a pull-up device and a pull-down device, according to the present invention.

(12) As illustrated in FIG. 1(a), the ternary logic circuit according to the present invention includes a pull-up device 100 and a pull-down device 200.

(13) As illustrated in FIGS. 1(a) and 1(b), current-voltage characteristics of the ternary logic device according to the present invention have a current (I.sub.CON) component, which is not affected by an input voltage and is affected only by an output voltage, and a current (I.sub.EXT) component, which is affected by an input voltage and is not affected by an output voltage.

(14) The current I.sub.CON, which is affected by the output voltage, has a current value I.sub.C when the output voltage V.sub.OUT is half the operating voltage V.sub.DD (V.sub.OUT=V.sub.DD/2), and the current I.sub.EXT, which is affected by the input voltage, has a current value I.sub.E when the input voltage V.sub.IN is half the operating voltage V.sub.DD (V.sub.IN=V.sub.DD/2) and exponentially increases to a maximum current I.sub.MAX at a point where the input voltage V.sub.IN and the operating voltage V.sub.DD are equal to each other (V.sub.IN=V.sub.DD).

(15) In this case, I.sub.E, I.sub.C, and I.sub.MAX are characterized by the relationship I.sub.E<I.sub.C<I.sub.MAX.

(16) Complementary current-voltage characteristics for the operation of the ternary logic device according to the present invention may be represented by the following equation.

(17) $\left[\mathrm{Equation}1\right]$$\begin{array}{c}{I}_{\mathrm{OUT}}\left(V_{\mathrm{IN}},V_{\mathrm{OUT}}\right)=I_{\mathrm{CON}}\left(V_{\mathrm{OUT}}\right)+I_{\mathrm{EXP}}\left(V_{\mathrm{IN}}\right)\\ =I_C\exp \left[\left(V_{\mathrm{OUT}}-\frac{V_{\mathrm{DD}}}{2}\right)\right]+I_E\exp \left[\left(V_{\mathrm{IN}}-\frac{V_{\mathrm{DD}}}{2}\right)\right]\end{array}$

(18) In [Equation 1], and are exponential coefficients of each current mechanism, and + and signs before and are respectively applied to the pull-down device 200 and the pull-up device 100.

(19) Here, I.sub.MAX=I.sub.Eexp[(V.sub.DD/2)] is equal in both the pull-down device 200 and the pull-up device 100.

(20) The operation principle of the ternary logic circuit according to the present invention will be described with reference to [Equation 1] and FIG. 2.

(21) FIGS. 2(a) to 2(c) illustrate a voltage transition process of three states: a low 0 state, an intermediate 1 state, and a high 2 state. The low 0, the intermediate 1, and the high 2 states are determined by intersection points of the output current (I.sub.OUT)-output voltage(V.sub.OUT) curves of the pull-up device 100 and the pull-down device 200.

(22) To transition to the low 0 state or the high 2 state, the pull-up device 100 or the pull-down device 200 shows a dominant current of I.sub.EXP within a range of V.sub.IN>V.sub.IL (FIG. 2(a)) or V.sub.IN<V.sub.IH (FIG. 2(b)) as I.sub.OUT to I.sub.EXP and make a current path in GND (V.sub.OL) or the operating voltage (V.sub.DD) (V.sub.OH).

(23) While the output current V.sub.OUT transitions to V.sub.DD/2 around V.sub.IMH and V.sub.IML, the output current I.sub.OUT of the pull-up device 100 and the pull-down device 200 is similar to the sum of the constant current I.sub.CON and the exponential current I.sub.EXP, and this may induce a slow transition process.

(24) Finally, an additional intermediate 1 state V.sub.OM is obtained at one intersection point at V.sub.OUT=V.sub.DD/2 within a range of V.sub.IMH<V.sub.IN<V.sub.IML, and the output current I.sub.OUT of the pull-up device 100 and the pull-down device 200 is dominated only by the constant current I.sub.CON (FIG. 2(b)).

(25) When assuming a symmetrical device having I.sub.MAX>I.sub.C, an intermediate input voltage (V.sub.IM=V.sub.IMLV.sub.IMH) and a transition voltage (V.sub.TR=V.sub.IMHV.sub.IH=V.sub.ILV.sub.IML) may be determined according to [Equation 2] and [Equation 3] below.

(26) $\begin{array}{cc}{V}_{\mathrm{IM}}=V_{\mathrm{DD}}-\frac{2}{}\ln \left(\frac{I_{\mathrm{MAX}}}{I_C}\right)-\frac{2}{}\ln \left(\frac{}{2}\right)& \left[\mathrm{Equation}2\right]\\ V_{\mathrm{TR}}=\frac{}{}\frac{V_{\mathrm{DD}}}{2}+\frac{1}{}\ln \left(\frac{}{2}\right)& \left[\mathrm{Equation}3\right]\end{array}$

(27) Here, V.sub.IL, V.sub.IH, V.sub.IML, and V.sub.IMH are determined via a combination of current equations of the pull-up device 100 and the pull-down device 200.

(28) For example, when V.sub.IN>V.sub.DD/2, V.sub.IML is obtained from a relationship of {I.sub.EXP+I.sub.CON}.sub.pull-down={I.sub.CON}.sub.pull-up by dV.sub.OUT/dV.sub.IN=1.

(29) For operation of a STI, it is required to satisfy the conditions V.sub.IL<V.sub.DD, V.sub.IML>V.sub.DD/2, V.sub.IMH<V.sub.DD/2, and V.sub.IH>0 (GND). Consequently, the criteria for and are as follows.

(30) $\begin{array}{cc}{}^{}=\frac{}{\ln \left(10\right)}<\log \left(\frac{I_{\mathrm{MAX}}}{I_C}\right)/\frac{V_{\mathrm{DD}}}{2}{}^{}=\frac{}{\ln \left(10\right)}& \left[\mathrm{Equation}4\right]\end{array}$

(31) In FIG. 2, as a mathematical example, I.sub.MAX=10.sup.5A, I.sub.C=10.sup.8A, log(I.sub.MAX/I.sub.C)/(V.sub.DD/2)=6, =2, =10, and V.sub.DD=1 are used.

(32) FIG. 3 illustrates a voltage transfer curve of the standard ternary inverter (STI) having three output states (V.sub.OH=V.sub.DD, V.sub.OM=V.sub.DD/2, V.sub.OL=GND) as a function of and , and consequently, shows a change of V.sub.IM by [Equation 2] and the transition voltage by [Equation 3].

(33) As shown in a graph inserted into FIG. 3, the low transition voltage V.sub.TR and the high V.sub.IM may be obtained by larger and smaller within a range of and given by [Equation 4]. This is preferable for the voltage transfer curve of the ideal standard ternary inverter (STI).

(34) However, both the pull-up device 100 and the pull-down device 200 have a specific saturated value due to the log(/2)/ term in [Equation 2] and [Equation 3].

(35) In the term, such a nonlinear log(x)/x function is induced at V.sub.IMH and V.sub.IML around V.sub.DD/2 at which the output current I.sub.OUT of the pull-up device 100 and the pull-down device 200 is approximated as the sum of the constant current I.sub.CON and the exponential current I.sub.EXP.

(36) When considering only the transition voltage V.sub.TR, large seems to be preferable. However, when considering noise margin, it is expected that there will be an optimal condition for reasonable V.sub.IM.

(37) The constant current I.sub.CON independent of the input voltage, which is a current mechanism essential for the intermediate 1 state of the STI, may be realized by a junction BTBT current (I.sub.BTBT) independent of a gate voltage.

(38) In addition, the exponential current I.sub.EXP dependent on the input voltage for the 0 and 2 states may be made into a subthreshold current (I.sub.sub).

(39) FIG. 4 illustrates a cross-section of a 32-nm high-k/metal-gate planar nMOS and BTBT occurrence rates at a low channel doping (Nch=210.sup.18 cm.sup.3) and a high channel doping (Nch=210.sup.19 cm.sup.3). Basic structure information such as an equivalent oxide thickness of 1 nm, a high drain doping (HDD) of 110.sup.20 cm.sup.3, and a low drain doping (LDD) of 2.510.sup.19 cm.sup.3 was based on 32-nm low static power technology reported in ITRS. A device simulation was performed via Synopsys Sentaurus based on a BTBT model and a bandgap narrowing model.

(40) By increasing a channel doping, a maximum BTBT occurrence region moves from the LDD region below the gate to the HDD and body junction. This make a dominant off current mechanism become gate bias-independent I.sub.BTBT.

(41) FIG. 5 illustrates output current-input voltage characteristics of ternary CMOS and a voltage transfer curve of STI according to a channel doping. It can be seen that a simple increase in a channel doping changes the current-voltage characteristic from CMOS to tenary CMOS and changes the voltage transfer curve from binary inverter to ternary inverter.

(42) FIG. 6 shows that simulated current-voltage data is well matched with a modeled current-voltage based on [Equation 1].

(43) FIG. 7 illustrates voltage transfer characteristics and static noise margin analysis. SNM has 200 mV in a typical design (Nch=210.sup.19 cm.sup.3) and 230 mV in an ideal CMOS. In an SNM contour inserted into FIG. 7, it can be seen that NM increases as becomes larger and becomes smaller.

(44) A STI circuit having the above-described characteristics, according to the present invention, and a truth table thereof, a MIN gate circuit, to which the STI circuit is applied, and a truth table thereof, and a MAX gate circuit and a truth table thereof are illustrated in FIG. 8.

(45) In a case where the pull-up device 100 and the pull-down device 200 are not affected by the input voltage (in a case where both are turned off), both the two devices operate as simple resistors and form a ternary digit (1 state) through voltage division. On the other hand, when only one of the pull-up device 100 and the pull-down device 200 is turned on to allow a current to flow therethrough, V.sub.DD (2 state) or GND (0 state) is transferred.

(46) As described above, the current region which is not affected by the input voltage can be implemented by I.sub.BTBT characteristics which are not affected by the gate of the CMOS, and the current region increasing exponentially to the input voltage uses I.sub.sub of the CMOS.

(47) While the present invention has been particularly shown and described with reference to specific embodiments, it should be understood that the invention is not limited thereto and various modifications and changes may be made without departing from the scope of the present invention and the equivalents of the appended claims by those of ordinary skill in the art to which the invention pertains.

DESCRIPTION OF REFERENCE NUMERALS

(48) 100: pull-up device 200: pull-down device