Professional Documents
Culture Documents
Abstract—In this paper, Flash Analog to digital converter is which in turn offers higher speed. We have various types of
implemented whose resolution is 3-bits. The designed Flash ADC ADC architectures for instance successive approximation type
consists of a resistive ladder network, comparators, the ADC, Flash type, sigma-delta, etc. Among these Flash ADC is
thermometer to a binary encoder and the entire design is carried preferred since it offers high speed because of its parallel
out using LTspice tools employing 180nm technology. The architecture, the conversion time is not limited by resolution
reference voltage applied to the resistive ladder network is 1.8V. hence these ADC’s are utilized in those systems where
A two-stage operational amplifier is used as a comparator in the bandwidth with a wide range and high speed is required[2].
flash ADC. Binary code is obtained from the thermometer code
by utilizing a priority encoder. The major problem that usually Al-Ahsan Talukder and Shamim Sarker have implemented
appears in flash ADC is as the number of resolution bits flash ADC with 3-bit employing threshold inverter
increases, the Area, as well as the power consumption of the quantization(TIQ). The main feature of this technique is the
circuit, also increases. In this paper, we principally concentrated absence of separate reference voltage power supplies, unlike
to lessen the power consumption of the ADC by optimizing other Flash ADC implementations. It is possible to set the
encoder circuitry. With the purpose of reducing power switching voltage of the inverter by choosing nmos as well as
consumption, Encoder is implemented using 2:1 mux based on pmos transistors with suitable width to length ratios. Th is
various logics such as switch logic, pass transistor logic as well as architecture comprises of TIQ comparator, the thermometer to
CMOS logic. In addition to this, the Wallace tree encoder was
the binary encoder in addition to gain booster. Because of the
also implemented. Performance parameters of Flash ADC such
as conversion time as well as average power are calculated and
change in the dimensions of comparator Area changes [3].
compared. It is verified average power obtained using Wallace Sonu Kumar and Anjali Sharma proposed a strategy
tree encoder is 910pW and it is less compared to encoder employing CMOS technology that is demonstrated for
implemented with other designs. implementing opamp. They preferred CMOS technology for
designing an operational amplifier due to the fact that CMOS
Keywords—Average power, Conversion time, Thermometer to devices consume low static power and these devices are highly
binary encoder, comparator.
withstanding noise. The two-stage operational amplifier
performance parameters are obtained whose gain is 44.98dB,
I. INT RODUCT ION the phase margin is 63 degrees, the gain-bandwidth product is
Digital signal processing has advanced intensely due to the 33.4MHz, power consumption 276μW[4].
rapid expansion of science and technology. In the majority of Mirza Nemeth Ali Baig and Rakesh Ranjan have
the digital domains, signal processing offers several advantages implemented high-speed flash ADC for wireless LAN
such as flexibility in design and programmability, reduced applications. The designed 3-bit flash ADC is implemented
silicon area, high accuracy, as well as a smaller amount of using seven operational transconductance based comparators
power consumption. The design process is cost-effective and with the reference voltage of 250mV and a high-speed encoder
faster. Hence it is possible to design a system with a lesser area is implemented using full adders. This design is a flash -based
along with high speed. It is required to have an analog to digital ADC converter with a finite output resolution of three bits and
converter that offers much higher speed in wireless power consumption of about 223μW and resides in a chip area
communication, image processing, etc[1]. of 0.089287 mm2. The high-speed flash ADC is being
It is preferred to have digital systems that are portable and designed and verified using the CADENCE Virtuoso tool with
have prolonged battery life. This can be only possible by CMOS 180 nm technology. Since ADC is implemented by
developing applications that consume less power. Since ADC’s utilizing a full adder based encoder, The area is limited by the
act as front-end components in the majority of mixed-signal resolution[5].
systems, we focused to design ADC that consumes less power
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
Sarojini Mandal and J.K.das had implemented a 3-bit flash Resolution of flash ADC does not determine accuracy , it just
ADC using cascading full adder by using pass transistor logic indicates the total bits that are obtained at the output side.
that makes the circuit much faster. They have improved the
efficiency of flash ADC by improving the working of the
comparator by scaling down the length to width ratio of
transistors assuming transistors operating in the saturation
region. The gain of comparator increased by proper sizing o f
transistors. The encoder circuit is designed using a full adder
with 10 transistors. Chip complexity and area are reduced using
pass transistor logic employed in a full adder circuit. As full
adder is designed using pass transistor, logic levels will
deteriorate[6].
Jayesh. J. Vyas implemented flash ADC comprising 3-bit
resolution employing 180-nanometer technology files using
NGspice tool for high-speed applications. The response time of
the ADC and the comparator was calculated as 4.9 ns and
3.7ns. Power consumed by ADC is given as 50μW. Encoder
circuit is implemented using basic gates. The gate delay for the
inverter, two-input, three-input, and four-input NAND gates
are given as 0.29ns, 0.51ns, 0.37ns, 1.06ns respectively with Fig 1: 3-bit Flash ADC block diagram
2ns rise time and fall time for inputs. Comparator sensitivity
and comparator gain are calculated as 11mV and 29dB A. Resistive Ladder Network
respectively. the main drawback is delay increases linearly with
resolution since NAND gates were used[7].
Piyush V. konodiya had implemented flash ADC for high -
speed ultra wideband communications(it’s a form of
technology whose signal bandwidth is greater than 500MHz). .
Flash ADC employing ROM based encoder is designed and
reported Conversion time, as well as the power consumption of
the circuits, are given as 13.64mW and 1.135ns respectively. It
is not possible to suppress bubble errors utilizing a ROM
encoder because it is slow[8].
II. FLASH ADC ARCHITECTURE
Fig 2: Resistive Ladder Network
To implement N bit flash ADC, we require 2N -1 Figure 2 shows resistive ladder network. In this
comparators are needed. Similar to opamp, the comparator paper, we have implemented 3-bit flash ADC comprises of
comprises of two inputs where analog input is given to the resistive ladder network. It is possible to provide a steady
inverting terminal and reference voltage is applied to non- reference voltage to the comparators utilizing the resistive
inverting terminal. Comparators are divided with the help of ladder network. 2N resistors are required to design N-bit Flash
the resistive ladder network. 2N resistors are utilized to form a ADC. The main function of this ladder network is that it
resistive ladder network of N bit. Since we have implemented divides reference voltage across all resistors such that the
3-bit, the number of res istors required will be eight. The difference between the reference voltage of the corresponding
reference voltage is generated across the resistive ladder
two comparators will be the least significant bit value. Since
network between reference voltage and ground is equally
we have given reference voltage of 1.8 volts, each comparator
distributed and is differed by the least significant bit. the
comparator compares the reference voltage signal with the reference voltage will be differed by 255mV.
input analog signal and indicates output as logic high whenever B. Operational amplifier as comparator
analog input exceeds reference voltage and indicates logic low
output when analog input is smaller than reference voltage. The
output of comparators forms a thermometer code. Further, we
need to translate thermometer code into binary code[9].
The output generated mainly depends on the resolution. But
the major disadvantage is as the resolution increases, th e
number of comparators required will get increases . For
example if we need to implement flash ADC of 9 bit, we need Fig 3: Comparator block diagram
511 comparators which occupy a huge die area and dis sipates a Figure 3 shows comparator block diagram. The
large amount of power. Hence we need to reduce the power applied input signal is compared with the reference signal and
and area of flash ADC as they are major constraints. will provide output as shown below[11].
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
where bubble error suppression takes place using NAND gates Figure 9 shows the implementation of a full adder employing CMOS
technology. we can obtain sum and carry by adding input bits say for instance
with two or more inputs. if we apply input ABC as 111 the corresponding sum, as well as carry, will be 1
It is the simplest possible method to convert a thermometer and so on.
to binary code. All available inputs get passed across an equal
number of full adders hence propagation delay of all applied
inputs are the same which is the main benefit of this encoder
type. The speed of the encoder is high since it exhibits
pipelining.
As resolution increases the number of full adders to convert
thermometer code to binary code also increases. Hence area
occupied by the encoder will be increased that leads to large
power dissipation and chip cost.
A. Full adder implementation
Here we have implemented a simple one-bit conventional Fig 10: Full adder output
full adder with 14 PMOS and 14 NMOS transistors. Two
inverters are used to convert inverted sum and carry. Using the B. Wallace Tree Encoder
boolean function we can realize gate-level implementation and By utilizing adder, the Wallace tree encoder is built as it is
also we can design a transistor-level schematic for both sum the most elementary building block. As we are trying t o
and carry. The carry out internal node is forwarded to obtain implement a 3-bit flash ADC, the number of full adders needed
sum[16]. is four. The obtained number of ones on the comparator output
A schematic of the full adder is shown below is counted by the Wallace tree. The dominant problem is as
resolution increases, the number of full adders will be
Here, Sum =(A ْ B) ْ Cin) (3) increased. As full adder consumes more area, power
=A(B)’Cin+(A)’Bcin+AB(Cin)’+ABCin (4) consumption will be high. So here we are replacing full adders
with 2 to 1 multiplexer[12].
=[CO’*(ABCin)+(A+B+Cin)]’ (5)
Carry output=(A*B)+(B*C)+(A*C)
=A*B+Cin*(A+B) (6)
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
D. 2:1 Multiplexer Using Various Logic Styles that leads to the increase of parameters like area, power as well
as delay[15].
i) 2:1 Mux Using Switch Logic(Transmission Gate)
IV RESULTS
Transient Analysis
Transient analysis: Transient analys is is performed by
applying a sinusoidal waveform with a voltage of 1.8 volts by
applying a frequency of 1KHz and average power is
calculated.
A. Average Power
Fig 15: 2:1 mux using pass transistor logic Designing low power portable design systems is the main
Figure 15 shows 2:1 mux using pass transistor logic. challenge. power consumption of the circuit is reduced by
Here we have utilized only two transistors to implement mux. introducing various logic design styles in CMOS technology.
in CMOS circuits, the average power is estimated as the sum of
logic levels getting deteriorate using pass transistors. so to
static power, dynamic power as well as switching power. the
overcome that buffer is inserted to obtain a strong one and average power calculation is shown in the equation below.
strong zero[13]. Average power of the 3-bit Flash ADC using Wallace tree
ii) 2:1 Mux Using Complementary metal-oxide-semiconductor encoder is 910pW[14].
(CMOS) logic
Paverage=Pstatic+Pdynamic+Pswitching (7)
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
Fig 25: 3-bit Flash ADC using pass transistor logic 2:1 multiplexer
Fig 22: 3-bit Flash ADC using switch logic 2:1 multiplexer
Fig 23: Average power calculation for 3-bit Flash ADC using switch logic
encoder
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.
Proceedings of the Third International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV 2021).
IEEE Xplore Part Number: CFP21ONG-ART; 978-0-7381-1183-4
V. COMPARISON [5] Mirza Nemath Ali Baig, Rakesh Ranjan, “ Design and
T ABLE II: 3-BIT FLASH ADC PARAMET ER COMPARISON implementation of 3-bit High-speed flash ADC for wireless
LAN Applications”, IJARCCE, Vol 6, 2017.
Using Using Using Using CMO S [6] Sarojini Mandal, Dr. J.K Das, “ Design of 3-bit low power
Wallace Pass switch 2:1 mux flash ADC”, IJARCET, Volume 3 issue 4, April 2014.
tre e transistor logic 2:1 [7] Jayesh J. Vyas ,” Simulation of 3 bit Flash ADC in 0.18um
e ncoder logic 2:1 mux technology using Ngspice Tool for High-speed applications”,
IJSRD, volume 1, Issue 2, 2013.
mux
[8] Piyush. V. Kanodiya, Amisha. P. Naik, “Analysis and
Average 910pW 371μW 345μW 1.372nW design of flash Analog to digital converter for ultra-wideband
power applications”, IEEE 2011.
Resolution 3 bits 3 bits 3 bits 3 bits [9] Pradeep Kumar, AmitKolhe, "Design & Implementation of
Low Power 3-bit Flash ADC in 0.18μm CMOS", International
Journal of Soft Computing and Engineering (IJSCE), ISSN:
Conversion 74.2ns 124.7ns 189 ns 55 ns
2231-2307, Volume-1, Issue-5, November 2015.
speed
[10] Ashima Gupta, anil Singh, “Highly digital voltage
scalable 4-bit flash ADC”, IET circuits, devices &systems,
VI. CONCLUSION 2019.
[11] M.P. Ajanya, George Tom, “ Thermometer code to binary
In this paper, we have implemented flash ADC with a code converter for flash ADC –A Review”, ICCPCCT, 2018.
3-bit resolution. Design and simulation of 3-bit flash ADC are [12] M.P. Ajanya, George Tom, “Low power Wallace tree
carried out using the LTspice tool employing 180nm encoder for flash ADC”, IOP conference series: Material
technology model files. Parameters such as conversion time, science and engineering 2018.
the average power of 3-bit flash ADC are calculated and [13] Anjum Aara, “Design and implementation of CMOS and
compared. Mainly we focused to reduce the power CNT based 2:1 multiplexer at 32nm technology”, IJRET,
consumption of 3-bit flash Adc by optimizing the encoder Volume 6, August 2019.
circuitry. The encoder is implemented using different design [14] Neil H.E. Weste, David Harris “CMOS VLSI Design”.
styles such as Wallace tree encoder, 2:1 mux based encoder. It [15] G. Tretter, M. Khafaji, D. Fritsche, C. Carta And F.
is verified that 3 bit Flash ADC designed utilizing a Wallace Ellinger, "A 24 Gs/S Single-Core Flash Adc With 3 Bit
tree encoder offers less power compared to a 2:1 mux based Resolution In 28 Nm Low-Power Digital Cmos," 2015 Ieee
encoder. Also, we have implemented a 2:1 mux based Radio Frequency Integrated Circuits Symposium (Rfic),
encoder employing different design logics such as switch Phoenix, Az, 2015,
logic, pass transistor logic, and complementary metal-oxide-
semiconductor logic. out of these logics, 3-bit Flash ADC
implemented utilizing encoder with CMOS 2:1 mux consumes
less power and offers less delay. Further, it is intended to
increase the resolution of flash ADC and also to reduce the
average power consumed by the circuit by optimizing the flash
ADC circuitry.
VII. REFERENCES
Authorized licensed use limited to: East Carolina University. Downloaded on June 18,2021 at 23:49:10 UTC from IEEE Xplore. Restrictions apply.