 Open Access
 Authors : Anup Kumar Biswas
 Paper ID : IJERTV10IS040205
 Volume & Issue : Volume 10, Issue 04 (April 2021)
 Published (First Online): 26042021
 ISSN (Online) : 22780181
 Publisher Name : IJERT
 License: This work is licensed under a Creative Commons Attribution 4.0 International License
State Transition Diagram for A Pipeline Unit based on Single Electron Tunneling
Anup Kumar Biswas
Assistant Professor
Department of Computer Science and Engineering Kalyani Govt. Engineering College
Kalyani741235, Nadia, West Bengal, India,
Abstract: For low cost, low power consumption, high operating speed, and high integration densitybased electronic goods are economically indispensable in business, engineering, science and technology in the present era. Single Electron tunneling is one such approach by which all the logic gates can be implemented. Single Electron tunneling devices (SEDs) and Linear Threshold Gates (LTGs) have the capabilities of controlling the transport of only an electron through a tunnel junction. A single electron which has the charge sufficient to store information in a SED in the atmosphere of 0K. Power being consumed in the single electron tunneling circuits is very low comparing the (CMOS) circuits. The speed of the processing of LTG based device will be very close to electronic speed. The singleelectron transistor (SET) and LTG both attract the scientists, technologists and researchers to design and implement their large scale circuits for small cost of the ultralow power and its small size. All the tunneling events in the case of a LTGbased circuit happen when only a single electron tunnels from one conductor to other through the tunnel junction under the proper applied bias voltage and multiple input voltages. For implementing a state transition diagram for a pipeline unit, LTG would be a best candidate to fulfill the necessities requiring for its implementation. As far as Ultralow noise is concerned, LTG based circuit would be a best selection for implementing the desired tunneling circuits. Different LTGs, a DFlipflop, a 2:1 multiplexer are implemented, and above all, a state transition circuit for a pipeline is implemented also.
Key words: State transition, Electrontunneling, Coulomb blockade, pipeline, linear threshold gate

INTRODUCTION
We are interested in constructing a state transition diagram for a pipeline unit which can provide us with collision or collision free transition for latency m ( m is an integer number). Our intention for doing so is to utilize the electron tunneling phenomena through the SET as there is very ultra low power being consumed for a single electron while tunneling. For the purpose of the implementing of a state transition diagram circuit, one can take advantages of SET

TUNNEL JUNCTION AND SINGLE ELECTRON
TRANSISTOR
A tunnel junction is made up of two conducting materials and a thin insulating barrier between them. The tunnel junction must have a capacitance C and a resistance . The conducting electrodes of the tunnel junction will be a superconducting or semiconducting material. When we are considering them as superconducting, electron(s) having one elementary charge (1.60217662 Ã— 1019 Coulombs) carry the current through the junction.
We have the knowledge that current cant flow through the insulating barrier as it creates a barrier against the movement of an electron in the case of classical electrodynamics, whereas for the case of quantum mechanics, there must be a positive possibility that when an electron residing in one side of the barrier of the insulator in order to reach the another side of it, the electron to which the bias or input voltages are supplied can go to the other electrode. If bias voltage greater than the threshold voltage is applied properly, there will be a current flow. Avoiding other effects, the current will be following in proportion to the bias voltage applied as per the firstorderapproximationtunneling process. In electrical terms, a tunnel junction, of course, have a constant resistance R value relying on its barrier thickness a shown in Fig 1(a). When the two conducting materials are connected with an insulating layer between them together, there will also have a capacitance in the junction. In this context, the insulator represents itself as dielectric and two conducting plates with dielectric forms a capacitor C in the tunnel junction. For the discrete nature of electric charge in the tunneling phenomenon, current following through a tunnel junction is discontinuous i.e., a series of events in which merely one electron will be able to pass or tunnel through the tunnel junction at a particular time. Owing to the single electron tunneling through the junction, the tunnel capacitance is charged with an elementary charge (1.60217662 Ã— 10 19 Coulombs) developed a voltage according to the relation
= , where C=tunnel junction capacitance. When the
and other one can take that of LTG. In the present situation
we will accept the second case i.e. using LTG a state
cap
acitan
ce of the tunnel junction is ultrasmall in order of
transition diagram will be presented. The logic gates required to construct State transition diagram for a pipeline unit are being OR, AND, NAND. In addition to them (i) combinational circuits like 2:1 multiplexers and (ii) sequential circuits like D Flipflops will be necessary. All of them will be described step by step and their comparative studies will be shown in due places.
O(1018) , the voltage building up in the tunnel junction will
be adequate to prevent another electron(s) to tunnel through. We will be able to suppress electrical current when the bias voltage being applied is lower than the voltage created in the tunnel junction, as a result, the resistance of the device will no longer remain constant. The increment of the differential resistance of the tunnel junction around zero bias is said to be the Coulomb blockade [1, 10, 13, 15].
Fig.1 (a) Tunnel Junction
The principle of singleelectron technology [1, 2, 3, 4, 8] is developed on the basis of the single electron tunneling events as well as Coulomb blockade. Single electron tunneling circuits can be taken as a promising candidate for future VLSI circuits for its ultralow power consumption, ultrasmall size, reducing node capability, high integration density and rich functionality.
The output terminal 0 is directly connected to the common channel of the two SETs and to the ground via a capacitor
to put down charging effects.
Fig.2 (a) an Inverter, (b) Symbol
Fig.1 (b) Single Electron Transistor
A (SET) [2, 3, 5] made up of two tunnel junctions
The operation of the inverter can be described like this: – when the bias voltage is properly applied, the output 0 of the inverter will be high (logical 1) when the input voltage
is low and 0 value will be low (logical 0) when the input voltage is high. How does it happen? We set the voltages 1 = 0 2 =16mV along with the
and having their capacitances
1
1 2
and 2 with their
tuning gate voltages, at present, both for SET1 and SET2. Now if is logical 0, the SET1 is in conduction
resistances 1 and 2 respectively, two other capacitors
mode and the
is in Coulomb blockade. This results the
and 1 are shown in Fig. 1(b). They shares only a common inland bearing a low capacitance. The electric potential of the island can be tuned in (i.e., increased or decreased) by a third electrode, called gate, and the gate is being coupled to the island through a capacitor . An extra capacitance 1 which is intentionally connected to the island for the purpose of manipulating the gate voltage (input). The drain, source and gate voltages may be marked by , and respectively For proper operations of a SET, both of the
SET2
output being connected to bias voltage and causes the output voltage to be high. Coulomb blockade works against the steady flow of current because as soon as the high voltage (logic 1) is applied to the input, it causes to shift the induced charge on each of the islands of the two SETs by a fraction of the charge of an electron and keeps the SET1 in Coulomb blockade and the SET2 in conducting mode. Eventually, the output changes from high to low.
For the inverter, the parameters values chosen for the
resistances 1, and 2 should be greater than = h/ 2
25.8 K and charging energy E =e2/2C [where = +
simulation are: 1
=0, 2
=0.1Ã— ,
= 9, = 1 ,
4
4
10
C 1
= 1 ,
= 1 ,
= 1 ,
= 1 ,
= 1 , =
2+ + 1] should necessarily be greater than thermal
3 2 2 2
1 10
1 2 2 2
1
2
17 and
= 17 , R1 =R2=100K. For simulation
fluctuations , i.e, = > , the product of the 4
2 4
2
Boltzmann constant, k(1.38 49
1023 joule /K) , and the
purposes, the value of C is taken to be 1aF.
06 Ã—
temperature, T in Kelvin.

AN INVERTER & SYMBOL
For a stable operation and output, an inverter is unavoidable for the logic gate family of TLG and this is why we have described the architecture and the arrangement of its different components. The inverter [1, 7] is made up of two SETs connected in series as depicted in Fig. 2(a). Two input voltages ( ) of the same values are directly coupled to the islands of the SET1 and SET2 through two
physical capacitances 1 and 2 respectively. The island of each of the SET1 and SET2 has a size smaller than 10 nm of gold and their capacitances should be less than 1017 F.
Fig. 2(c) Inverter
Fig.2 (d) Simulation result of Inverter
h(x)= 1( Ã— ) – ..(2) where being the n Boolean inputs and being their
=
=
corresponding n integer weights.
The LTG compares the weighted sum of the inputs
=
=
1( Ã— ) with the threshold value . If the value of the weighted sum is being greater than the threshold or critical voltage value then the logic output of the LTG will be high (logical 1), otherwise it will be logical 0.
The tunnel junction capacitance and the output capacitance 0 are considered to be the two basic circuit elements in a LTG. The input signal vector elements
( , , , , ) are weighted by their corresponding
1 2 3
vector capacitances (, , , , ) and added to the
1 2 3
In this work, the Boolean logic inputs corresponding to the
voltages are taken as: logic 0 =0 Volts and logic
junction voltage, . Whereas, the input signal vector
elements ( , , , , ) are weighted by their
1=0.1Ã—
1 2 3
correspondin v to capac
ces (
) are
g ec r
itan
1 , 2 , 3 , ,
19
19
We assume, for simulation purposes, C=1aF then Logic
1= 0.1Ã— 1.602Ã—10 =0.1 Ã— 1.602 Ã— 102 =16.02 Ã—
1Ã—1018
103 =16.02 16 mV.

Multiple input threshold logic gate
being subtracted from the voltage, .
The critical voltage to enable tunneling action, and which acts as the intrinsic threshold of the tunnel junction circuit. The bias voltage connected to tunnel junction through the capacitance, , is used in order to adjust the gate threshold to the desired value . When a tunneling happens though the tunnel junction an electron goes through the junction from p to q as shown in Fig. 3.
+
+
k
k
The following notations are being used for the rest our discussion.
CP = Cb
g
k=1
CP..(3)
Cn = C0 + h
Cn (4)
l=1 l
= CP + CPCn + Cn (5)
When we are considering all voltage sources in Fig. 3 to be connected to ground, the circuit can be considered as three
capacitors namely, CP, Cn and , connected in series,. Here,
Fig. 3 Multiple input TLG
A multiple input threshold logic gate [1, 5, 6, 7, 14] consists of a tunnel junction having capacitance and resistance , two multiple inputs connected at points p and q. Each input voltage , for the top left side is connected to point
q through the capacitor and each input voltage , for
is thought of the sum of all 2term products of these three capacitances.
It is the time to find the expression regarding the critical voltage of the tunnel junction. We assume the capacitance of the tunnel junction to be and the remainder of the circuit having the equivalent capacitance to be , as
observed from the perspective of tunnel junction, we can
the bottom left side is connected to point p through the capacitor . Bias voltage is connected to point b
measure the critical voltage[1,6,7] for the tunnel junction as
through a capacitor as well. Point p is grounded
=
(6)
through a capacitor 0 . This multiple input TLG can also be
2( + )
termed as a JunctionCapacitor (CJ) circuit. Using the JunctionCapacitor circuit we will be able to implement the LTG being presented by the sgn function of h(x) expressed
= 2[ + ()]
=
( ) ()
by equations (1) and (2).
2[
+ ] ( + )
g(x) = sgn{h(x)} = {
0, () < 0
1, () 0
(1)
( + )
=
The parameters used for the constructing elements and simulations of different gates like AND, OR, NAND and
2[ ( + ) + ( ) ()]
( + )
( + )
= .(7)
2
When the voltage across the junction is , a tunnel event will happen through this tunnel junction if and only if the condition given below is satisfied.
TLG
2inout AND
2inout OR
2inout NAND
2inout NOR
3inout OR
Cb
10.6C
11.7C
13.2C
11.7C
11.8C
C0
8C
8C
9C
9C
7.8C
CP
1
–
–
0.5C
0.5C
0.4C
Cp
2
–
–
0.5C
0.5C
–
Cn
1
0.5C
0.5C
–
–
0.5C
Cn
2
0.5C
0.5C
–
–
–
0.1C
0.1C
0.1C
0.1C
0.1C
Cg1
= Cg2
0.5C
0.5C
0.5C
0.5C
0.5C
C2
= C3
0.1C
0.1C
0.1C
0.1C
0.1C
C2
= C3
0.5C
0.5C
0.5C
0.5C
0.5C
C1
= C4
4.25C
4.25C
4.25C
4.25C
4.25C
CL
9C
9C
9C
9C
9C
0 = 0 , 1 = 16 , = 105, =0.1aF, other
capacitance values are in terms of , where = 1
TLG
2inout AND
2inout OR
2inout NAND
2inout NOR
3inout OR
Cb
10.6C
11.7C
13.2C
11.7C
11.8C
C0
8C
8C
9C
9C
7.8C
CP
1
–
–
0.5C
0.5C
0.4C
Cp
2
–
–
0.5C
0.5C
–
Cn
1
0.5C
0.5C
–
–
0.5C
Cn
2
0.5C
0.5C
–
–
–
0.1C
0.1C
0.1C
0.1C
0.1C
Cg1
= Cg2
0.5C
0.5C
0.5C
0.5C
0.5C
C2
= C3
0.1C
0.1C
0.1C
0.1C
0.1C
C2
= C3
0.5C
0.5C
0.5C
0.5C
0.5C
C1
= C4
4.25C
4.25C
4.25C
4.25C
4.25C
CL
9C
9C
9C
9C
9C
0 = 0 , 1 = 16 , = 105, =0.1aF, other
capacitance values are in terms of , where = 1
  .(8)
NOR gates and other combinational or sequential circuits are given in Table1 [6,7].
Table1
If the junction voltage is less than the critical voltage i.e.
  <
, no tunneling events through the tunnel junction
happen. As a result, the tunneling circuit stays in a
.
Theoretically, the thresholds are being integer numbers. And the threshold logic equations for two input logic AND, OR, NAND and NOR gates can be expressed [6] as follows.
(, ) = { + 2} (9)
(, ) = { + 1}(10)
(, ) = { + 2} (11)
(, ) = { + 1} (12)
If the threshold "" equals to " "(i being an integer value), this implies that the gates perform their functions correctly for any value in the range [ 1 < ]. For the purpose of maximizing robustness for variations in parameter values, the threshold value = can be replaced by the average = 0.5. As a result, we will be able to express

OR GATE
To implement a state transition pipeline circuit some logic gates are essential, one of them is an OR gate.
For implementing the OR gate we shall use the parameters = =0.5aF,3 = 11.7, =
the threshold logic equations for the twoinput AND, OR, 1 2
NAND and NOR gates as below:
(, ) = { + 1.5} (13)
(, ) = { + 0.5} (14)
(, ) = { + 1.5} (15)
(, ) = { + 0.5}. (16)
The threshold gatebased implementations of the Boolean Logic gates have the same basic circuit topology will be drawn in the subsequent sections. The threshold gates consist of a bias capacitance , a tunnel junction capacitance , and an output capacitance 0 and they are connected in series. The AND and OR gates contain two
input capacitors CP CP . For both the and
= 4.25, 1 = 2 = 0.5, =
9, 0 = 8, =105, = 16 in Fig.4(a) and accordingly after running the simulator the output we get is given in Fig. 4(b).
1 2
gates, the two input capacitances are CP = CP = 0.5 for
1 2
1
1
2
2
positively weighted inputs. On the contrary, for the NAND and NOR gates, two capacitors bearing the values Cn = Cn = 0.5 for negatively weighted inputs.
Each threshold gates is augmented with an inverter/buffer for stable output. The logic function done by the buffered threshold gate is being the complement of which is done by the threshold gate itself. For instance, a buffered gate implements the function. For the rest of our discussion, when we are referring to a logic function such as AND, we enforce that the logic function totally performed by the entire gate (i.e., threshold gate plus an output buffer).
Fig. 4(a) OR gate
Fig. 4(b) simulation result of OR gate

AND GATE
Another logic gate named AND gate is invariantly required to implement the state transition pipeline circuit unit. For the purpose of implementing the AND gate we need to use the parameters = =0.5aF, =
Fig.6 (a) NAND Gate (b) Input A
1 2
= 4.25, 1 = 2 = 0.5, = 9, 3 = 10.6, 0 = 8, =105 in Fig. 5(a) and accordingly after running the simulator the output we get is given in Fig. 5(b).
Fig. 5(a) AND Gate
Fig. 5(b) inputoutput of AND gate

NAND GATE
One universal logic gate categorically named as NAND gate must be required to implement desired state transition pipeline circuit in this paper. To implement the NAND gate,
we will use the parameters = =0.5aF, =
Fig.6 (c) input B (d)Simulation result of NAND

DESIGN OF DFLIPFLOP USING LOGIC GATE The flipflop is a sequential circuit used for storing information with the help of two stable states either 0 state or 1 state. To construct a shift register, one type of Flip flop–a D Flip flop, will be needed. A classical D Flipflop is made up of four NAND gates and one inverter shown in Fig. 7(a). There are two inputs D input and Clock signal input, and two outputs Q and Q . The input output relationship of the D Flipflop is shown in the Table2.
Fig. 7(a) Diagram of Dflip flop Table2
The pattern of the D Flipflop drawn in Fig. 7(a), will be necessary in the case of implementing a D Flipflop with the help of LTG based NAND and a buffer/ inverter shown in Fig. 2(a) or 2(c).
1 2
= 4.25, 1 = 2 = 0.5, = 3 = 13.2, = 9, 0 = 8, =105 in Fig. 6(a) and accordingly after running the simulator the output we get is given in Fig. 6(b).
The same parameters indicated for the case of a NAND and an inverter will also be used in the LTGbased D Flipflop. The output is provided with the positive edged trigger of the clock pulses. The implemented D Flipflop required for our
desired state transition pipeline circuit unit is depicted in Fig. 8(a).
The Fig. 8(a) given below is a D Flipflop implemented by LTG logic gates and its input and output signals shown in the Fig. 8(b) through 8(e) are the DFlipflop input signal, Clock signal, output of the Q and output of Q respectively.
Fig. 8 (a) D Flipflop using one buffer and four NAND gate based on LTG
Fig. 8(b) D input (c) clock signal
(d) output of Q and (e) output of Q

2Ã—1 MULTIPLEXER
A multiplexer is a selection or control circuit that selects one input from many inputs with the assistance of control line(s). The simplest multiplexer is a 2:1 Multiplexer which is drawn on the basis of LTGbased AND and OR gates. The input output simulation relationships of a 2:1 Multiplexer are being depicted in Fig. 9(b) through 9(e). This multiplexer needs to be made for our tate transition circuit.
Fig. 9(a) A 2:1 Multiplexer
(b) (c)
(d) (e)
Fig. 9(b) input A (c) input B (d) control (e) output

A DYNAMIC PIPELINE AND ITS RESERVATION TABLE
The feedback and feedforward connections shown in the Fig. 10(a) make the scheduling of successive events into the pipeline a nontrivial task. By this connection we can take output not only from the last stage but can take from other stages also. From the different dataflow patterns, we can use the same pipeline for evaluating different functions like X and Y.
Fig. 10(a) A dynamic pipeline
The pipeline is really very interesting as nonlinear patterns are being followed there. A static pipeline is specified by a single reservation table whereas a dynamic pipeline can be specified by more than one reservation table [17]. Two reservation tables for the dynamic pipeline are drawn in Fig. 10(b) and 10(c) corresponding to a function X and a function Y.
Fig. 10(b) Reservation table for function X
Fig. 10(c) Reservation table for function X
Two state diagrams we have obtained from the two Reservation tables are given in Fig. 10(d) and 10(e). A state diagram is obtained for the reservation table in Fig. 10(c) by using a 7bit shift resister and another state diagram can be obtained for the reservation table given in Fig. 10(b) by using a 4bit shift resister. The process of making state transition diagram is given in sections 13 and 14.
Fig. 10(d) State diagram for function Y
Fig. 10(e) State diagram for function X

TERMINOLOGIES OF LATENCY ANALYSIS

Collision: Number of time units or clock cycles necessary between two initiations for a pipeline is considered as the between the initiations. The value of the Latency is a nonnegative integer number. A Latency of p indicates that two initiations are separated by p clock cycles associated with the same reservation table. Any attempts by two or more initiations to use the same stage (like ) of the pipeline at the same time will make a
.

Forbidden latencies: A collision implies that
happening between two initiations in the same pipeline. Alert is necessary so that all collisions can be avoided in scheduling a sequence of pipeline initiations. Some of the latencies causes collisions but some doesnt. Latencies which are responsible for causing collisions are called . Noted that to detect a forbidden latency, we need to investigate the distance between any two checkmarks in a row of the reservation table. For example, in the first row (stage 1) of the reservation table in Fig.10(c), the distances between column 1 and 6, between column 6 and 8 are 5 and 2 respectively, so the latencies 5 and 2 are forbidden.


COLLISIONFREE SCHEDULING STEPS
A transparent method for identifying collision free latencies finds out the possible issue sequences and this will be presented in this section. These sequences are termed as
. This transparent method is discussed in the following four steps:

Step1: We compile a list of issue latencies responsible
for causing collisions in each row of the reservation table. To do this, an example with a reservation table is given in Fig.11.
We must investigate the column distances between any two cross marks (Ã—) in the same row.
In stage1(first row) of the reservation table having column distance between two cross marks (Ã—) equals to (9 1) = 8.
So, there is a collision with an = 8. The latencies causing the collisions are presented below.
Stage1: 8
Stage2: 1 5 6
Stage3: 0
Stage4; 1
Stage5: 1
Fig.11 A reservation table having 5 stages & 9 Phases

Step 2: From the reservation table, forbidden list are collected by taking the set of all nonzero row forbidden latencies.
So the forbidden list is = (1, 5, 6, 8)
This list informs us that new issues can be constructed without collisions for the latencies of 2, 3. 5, 7, 9+. It is noted that + indicates any latency greater than that number considered. 9+ means the integer number 9 and the integer numbers greater than 9.

Step3: For the case of initial state, a collision vector is made up of the forbidden list. The vector is read from the right to left by using 1 for collision and 0 for collisionfree. Noted that the last nonforbidden latency of 9 is avoided in the collision vector.
In the collision vector, the rightmost element is 1 as
1 indicates a collision, 2, 3 and 4 are not in the forbidden state , so we put the 0, 0 and 0 for the positions of 2nd, 3rd and 4th in the collision vector. In the same manner, in positions 5th, 6th and 8th we put 1, 1 and 1. For the 7th position we set 0.

Step4: State transition diagram to be drawn
The collision vector we have got is V = (1 0 1 1 0 0 0
1). The transitions from the vector are 2, 3, 4, 7 and 9+. To get the next state collision vector, the present state collision vector is shifted rightward for each transition. Then the shifted collision vector is ORed with the initial collision vector, as a result we get a new collision state vector.


Permissible and forbidden latencies
From the collision vector discussed above we can make a state Diagram that specifies the permissible state transitions among successive initiations. The collision vector = ( , 1 , , 2 , 1) corresponds to the initial vector of the pipeline at time 1 and this is why it is said to be an
. We assume that p is permissible latency in the range {1,2,3, . , 1} i.e., 1 ( 1).
The next state we are considering for the pipeline at time +
is obtained with the help of an mbit shift register given in the Fig. 13(b). The initial collision vector is now stored into the resistor. The resistor is then shifted towards right. Every onebit shift towards right corresponds to an increase of the latency by 1. When a 0 exits from the right end after p shifts, it indicates that p is a . In the same manner, for a 1 being shifted out from the right end tells us a , and thus the corresponding latency will be . For an example, we consider an initial collision vector (1 0 1 1 0 1 0) and after right shifting the permissible and forbidden latencies are depicted in Fig. 12 below.
Fig. 12 Permissible or Collision after right shifting

State Diagrams
The Fig. 13(a) and 13(b) are the same logic circuits. First one is made up of CMOS gates and the second one is of LTGs. Initially there is no state stored in the shift register. So the initial value of ( , 1 , , 2 , 1) is taken to be (0, 0, , 0, 0).When the initial collision vector ( , 1 , , 2 , 1) is passed through the OR gates with (0, 0, , 0, 0), we get the same output of OR gates as the collision vector is. And this vector is loaded in the shift register if we set =0 and apply a positive edged trigger clock pulse. The register is then shifted towards right 1 bit. We know a 1bit shift represents an increase in the latency by 1. Similarly a twobit shift corresponds to an increase in the latency by 2 and so forth. In this shifting process when a 0(zero) imerges from the right end i.e. at the output terminal 1 after shifts, it indicates that is a
. In the same way, while a 1(one)
imerging from the right end i.e. at the output terminal 1 after shifts, it means a and the corresponding latency p is treated as .
In the Fig. 13(b), if =1 and positive triggered clock pulses are applied, the logical 0s enter from the left end of the register and is loaded into the register. what will be the next state after p number of 0s shifting towards right? This is obtained by bitwise ORing the initial collision vector ( , 1 , , 2 , 1 with the shifted register contents. For
instance, the initial collision vector we assumed to be = (1 0 1 1 0 1 0) and which will really be the shift registers contents. After 1 right shift the contents of the shift register will be = (0 1 0 1 1 0 1). These two vectors are bitwise ORed
and we have the next state (1 1 1 1 1 1 1 ). After 2 shift, 4 shift, 5 shift, 7 shift there will be collision states.
Fig. 13 (a) A new collisionstatecreating unit
Fig. 13 (b) A new collisionstatecreating unit based on LTG
With the help of diagram, drawn in the Fig. 13(b) as well as Fig. 13(a) we will be able to explain the creation of initial and new state diagram. Taking n=8 in the general initial collision vector V = ( , 1 , , 2 , 1), we have from reservation table in Fig. 11
V = (8 , 7, 6, 5, 4 , 3, 2 , 1)
= (1 0 1 1 0 0 0 1).
Initially all the register values are 0s, i.e. D = ( 8 , 7, 6, 5, 4 , 3, 2 , 1)
= (0 0 0 0 0 0 0 0)
These two vectors are bitwise ORed by the topmost level OR gate circuits in Fig.13(b) and we get the same values as vector V = (1 0 1 1 0 0 0 1). The three input terminals in the
left hand side for Fig. 13(b) are 0input, and
Clock signal. If we set 0 in the terminal , then all the second level circuits which are 2:1 Multiplexers select the outputs of all the OR gates, not select the 0input. The outputs of all the 2:1 Multiplexers are being present at input terminals of all the D Flipflops. At this time we apply first positive clock and all the data present at the input side of D Flipflops will enter into the Flipflops and their output will be D =(8 , 7, 6, 5, 4 , 3, 2 , 1)= (1 0 1 1 0 0 0 1).
Next step if we set 1 in the , then the 0 is selected by the leftmost 2:1 Multiplexer and other Multiplexers, from 2nd left to rightmost one, will select the values from (8 , 7, 6, 5, 4 , 3, 2)= (1 0 1 1 0 0 0).
Now if a clock pulse is applied, the output of the register will be (0 1 0 1 1 0 0 0), i.e. one right shift happens. This shifted vector is automatically ORed with the initial collision vector. Depending upon the value of the transition, we apply the clock pulses and get the required state transition from the output of the OR gates.
We can take two right shift as there is one outtransition =2. So after setting =1, two clock pulses are applied to get the new transition state 2 from the OR gates.
2 = (1 0 1 1 1 1 0 1)
We can repeat the process for the transitions 3, 4, 7, and 9+ . For the case of outtransition 3, we apply another clock pulse and take the output from OR gates and get 3
3 = (1 0 1 1 0 1 1 1)
In the same way for the next transition 4, 7 and 9+, we apply
the clock 4, 7, 9 and more pulses and obtain the next state transitions 4 , 7 , 9+ respectively from OR gates.
4 = (1 0 1 1 1 0 1 1)
7 = (1 0 1 1 0 0 0 1)
9+ = (1 0 1 1 0 0 0 1)
As we have obtained all the next transition states, we are
ready to draw an initial state transition diagram (STD) [Fig.13 (c)] based on these next transition states i.e, 3 , 4 ,
7 and 9+ .
Fig. 13(c) An initial state diagram
With the help of the Fig. 13(b) and the next state transitions
2 , 3 , 4 , 7 , 9+ , we can find all the new transition states and can draw the complete transition state diagram as shown in Fig. 13(d).
Fig. 13(d) complete transition state diagram

DISCUSSION
We are interested in determining the processing delay of any logic gates, combinational and sequential circuits based on the LTG gates. To find out the delay of a logic gate we must involve critical voltage given in equations (6) and (7), as well as the tunnel junction capacitance . However, we assume at T = 0K, the switching/processing delay of a logic gate can be calculated using the approach [7, 8].
delay=(ln ( )) / (  ) ..(17)
where is the junction voltage and is the critical (threshold) voltage
The slowest switching happens whenever the critical voltage
has the value lesser than the tunnel junction voltage , i.e., <  , but very close to it. This must happen, for example when only 1 is logic 1, resulting =11.8mV for the case of a 2input NOR gate, the critical voltage of the
tunnel junction voltage is 11.58mV. It is considered that the probability of error change =1012, = 105. We obtain a gate delay equal to 0.07281ln ( )ns = 1.675 ns. In the same manner, we are capable of calculating the circuit delays written in Table2. When a charge tunnels through the tunnel junction, the amount of total energy in the circuit changes before and after the tunneling. So the
difference between the energies before and after the tunneling event is calculated by the relation
=
= ( )..(18)
and it is the amount of switching energy consumed when a tunnel event occurs in the tunneling circuit.
We have drawn curves regarding the switching delay as a function of the switching error probability in Fig. 14(a) and the switching delay as a function of the unit capacitance C which is shown in Fig. 14(b).
Fig. 14(a) Delay vs. Error Probability
Fig. 14 (b) Delay Vs. capacitance
We have found out the area/element numbers for every case of gates, its switching delay, and switching energy consumption for the corresponding individual linear threshold gate (using the same methodology as adopted for the Boolean gates). The switching power vs. area/ element diagram regarding our present LTG based circuits is shown in Fig. 15.
Next, we have calculated those parameters for state transition circuits and that are presented in Table2.
In the Fig.13(b), time delays for the phase1 is 0.340ln( )ns, for phase2 is 0.5ln( ) ns, for phase3 is 0.66ln( ) ns, for phase4 is .82ln( ) ns, and for phasen is (0.340+0.160Ã— ( 1))ln( )ns. With the assistance of the data taken from table2, the processing/ switching delay of a State transition diagram of a pipeline unit having phases n is (0.340+0.160Ã— ( 1))ln( )ns. We have discussed about collision vector above having number of vector elements equal to 8. Given that the value of equals to 1010, so the time after which the 8th bit of the register content will emerge from the pipeline unit is
(0.340+0.160Ã— (8 1))ln( ) ns=26.13ns.
We are to determine the clock pulse duration and it is
determined by the maximum delay of an OR gate or a 2:1 Multiplexer or a D Flipflop. From the Table2 we are able to get their delays and the maximum delay of them is ( ) is given by
= max{ 0.102ln( 0.160ln( )}
), 0.124ln(
),
Fig. 15. Switching power vs. area/ element
Therefore the clock cycle time should be at least 8.887ns.
=8.887ns .(19)
Table2
So the last 8th bit output will emerge after 26.13ns. If we consider the Fib. 13(b) as a pipeline system,
then after every clock pulse of time 8.887ns one output will emerge from terminal 1 in Fig. 13(b). If the value of n is a big number then output rate or frequency of the pipeline will be
1 26.13ns
= 38.27MHz (20)

SWITCHING DELAYS OF LTG AND SET
The time delay/processing delay for a CMOS logic gate like NAND, NOR, XOR is 12ns [20], whereas the time required for tunneling through a single electron transistor (SET) is approximately 4ns [4, 5, 16, 18]. The XOR gate using conventional logic circuits requires 16 transistors, whereas the same function can be implemented with the help of just one SET [3, 5, 6, 11] i.e. number of nodes can be reduced to 1 in lieu of 16. Given that error probability is 1015 then the delay for the inverter will be 2.3025ns and similarly the other delays for the other gates can be calculated and re shown in Table3. It is clear that the LTG based circuit is faster than the SET based circuit when =1015. The comparison of delays for SET and LTG gate based circuits is drawn by a bar diagram in Fig.16.
Fig.16 Delay comparison of SET and LTG

CONCLUSION
How an electron tunnels through a single electron transistor and an inverter is discussed at first. A generic Linear Threshold logic Gate implementation is elaborately discussed and from which we have been able to derive a family of logic gates like AND, NAND, OR. All the gates along with a 2:1 Multiplexer and a D Flipflop have been implemented and are verified by means of simulation using SIMON. The number of elements for logic gates, their delays, power consumed by them are given in a tabular form and their related curves or bar diagram are also given in the annexed figures. By dint of all these LTG gates being required as well as an inverter, a pipeline circuit called new collisionstatecreating unit based on LTG having nstages, has been presented. With the help of this pipeline circuit we have been able to draw the state transition diagrams. In single electron tunneling technology, it is observed that the logic gates are at least 3 times faster than CMOS based logic gates. The atmosphere temperature is kept very close to 0K in real operation.

REFERENCES

Anup Kumar Biswas, Design of A Pipeline for A FixedPoint Multiplication using Single Electron Tunneling Technology, International Journal of Engineering Research & Technology (IJERT), Vol. 10 Issue 04, April2021 pp. 8698

Souvik Sarkar1, Anup Kumar Biswas2, Ankush Ghosp, Subir Kumar Sarkar1 Single electron based binary multipliers with overflow detection, International Journal of Engineering, Science and Technology Vol. 1, No. 1, 2009, pp. 6173

A. K. Biswas and S. K. Sarkar: An arithmetic logic unit of a computer based on single electron transport system: Semiconductor Physics, Quantum Electronics & OptElectronics. 2003. Vol 6. No.1, pp 9196

A.K. Biswas and S. K. Sarkar: Error Detection and Debugging on Information in Communication System Using Single Electron
Circuit Based Binary Decision Diagram. Semiconductor Physics Quantum electronics and opt electronics, Vol. 6, pp.18, 2003

Alexander N. Korotkov, Singleelectron logic and memory devices INT. ELECTRONICS, 1999, Vol. 86, No. 5, 511 547

Casper Lageweg, Student Member, IEEE, Sorin CotÂ¸ofana, Senior Member, IEEE, and Stamatis Vassiliadis, Fellow, IEEE Single Electron Encoded Latches and FlipFlops IEEE TRANSACTIONS ON NANOTECHNOLOGY, VOL. 3, NO. 2,
JUNE 2004

C. Lageweg, S. CotÂ¸fana, and S. Vassiliadis, A linear threshold gate implementation in single electron technology, in IEEE Computer Society VLSI Workshop, Apr. 2001, pp. 93 A. Korotkov, Singleelectron logic and memory devices, Int. J. Electron., vol. 86, no. 5, pp. 511547, 1999.

K. Likharev, Singleelectron devices and their applications, Proc. IEEE, vol. 87, pp. 606632, Apr. 1999.

A. Korotkov, R. Chen, and K. Likharev, Possible performance of capacitively coupled singleelectron transistors in digital circuits, J. Appl. Phys., vol. 78, pp. 25202530, Aug. 1995.

J. R. Tucker, Complementary digital logic based on the Coulomb blockade, J. Appl. Phys., vol. 72, no. 9, pp. 43994413, Nov. 1992.

A. Korotkov and K. Likharev, Singleelectronparametronbased logic devices, J. Appl. Phys., vol. 84, no. 11, pp. 61146126, Dec. 1998.

C. Wasshuber, H. Kosina, and S. Selberherr, SIMONA simulator for singleelectron tunnel devices and circuits, IEEE Trans. Computer Aided Design, vol. 16, pp. 937944, Sept. 1997.

A. B. Zorin, S. V. Lotkhov, H. Zangerle, and J. Niemeyer, Coulomb blockade and cotunneling in single electron circuits with onchip resistors: Toward the implementation of the R pump, J. Appl. Phys., vol. 88, no. 5, pp. 26652670, Sept. 2000.

T. Oya, T. Asai, T. Asai, T. Fukui, and Y. Amemiya, A majority logic device using and irreversible singleelectron box, IEEE Trans. Nanotechnol.,vol. 2, pp. 1522, Mar. 2003.

Z. Durrani, A. Irvine, and H. Ahmed, Coulomb blockade memory using integrated singleelectron transistor/metaloxide semiconductor transistor gain cells, IEEE Trans. Electron Devices, vol. 47, pp. 23342339, Dec. 2000.

J. Millman and C. C. Halkias; Integrated Electronics Analog and Digital Circuits and Systems_ McGraw Hill Education; 2 edition

Kai Hwang and Naresh Jetwani, Advanced Computer Architecture parallelism, scalability, programability chapter 6, 2nd edition McGraw Hill

Millman's Electronic Devices & Ciruits 4th Edition (English, Paperback, Millman Jacob)
BIOGRAPHY
Anup Kumar Biswas is an Assistant Professor in the department of Commuter Science and Engineering in Kalyani Govt. Engineering College. He is awarded his PhD[Engg.] degree in the stream of Electronics and Telecommunication Engineering from
Jadavpur University in the year 2006. He has engaged in teaching and research activities since the last 16 years. His Specialization field is nanotechnology specially single electron tunneling technology. Dr. Biswas has published several papers in various national, international conferences and journals.