Level-Up/Level-Down Voltage Level Shifter For Nano-Scale Applications
Level-Up/Level-Down Voltage Level Shifter For Nano-Scale Applications
Level-Up/Level-Down Voltage Level Shifter For Nano-Scale Applications
SRINIVASULU GUNDALA1,*,
M. MAHABOOB BASHA2, S. VIJAYAKUMAR3
1Department of Electronics and Communication Engineering, Lakireddy Bali Reddy
College of Engineering (Autonomous), Krishna Dt., Andhra Pradesh, India
2Department of Electronics and Communication Engineering, Sreenidhi Institute of
Technology and Management Studies (Autonomous), Chittoor Dt., Andhra Pradesh, India
*Corresponding Author: srinivasulugundala46@gmail.com
Abstract
Multi supply voltage domain is an ultimate approach for reducing power
consumption at system level. To interconnect multi supply voltage designs, and
to prevent static current, the Level Shifters (LSs) need to be employed. The
design of LSs with least power consumption and delay performance are the
primary design constraints. In this research, a new architecture voltage level-
up/level-down shifter for Nano-scale applications has been introduced with short
circuit aware Complementary Metal Oxide Semiconductor (CMOS) logic, which
executes level shifting operation with lower power consumption and delay. The
designed circuit is useful to shift from as low as 0.2 V to 1.2 V and vice versa.
Apart from operational range, the performance matrices like power, delay, and
duty cycle of the circuit were optimized to meet the demands of nano scale
applications. The proposed LS simulated using Synopsis tools at 90 nm
technology. From the results, it has been observed that, the average “level-up”
and “level-down” shift active power consumption is 9 nW and delay 1.5ns at 1
MHz signal frequency. The strength of the design has examined with different
load and working conditions. The area occupied by the proposed design from the
post-layout simulation is 5.87 µm2.
Keywords: Area, Delay, Level shifter, Multi-VDD systems, Nano-devices, Power
consumption.
745
746 S. Gundala et al.
1. Introduction
The wide performance requirements of nanoscale devices demand the extreme
operating voltages. Different computational performance can be obtained by
different operating voltages, leading to the development of multi supply system
approach [1]. In CMOS design, lower power supply voltage is an efficient practice
to lower the power consumption. This is because it results in a quadratic relation of
dynamic power consumption, as per Eq. (1), and the leakage power is governed by
exponential law. However, lowering the power supply voltage will degrade the
delay. To lower the dynamic, leakage powers and to maintain the required
propagation delay, a multi supply system approach is the ultimate solution; it uses
multiple power supplies, which use more than one power supply voltage to meet
the required power and delay constraints [2].
𝑃𝑑 = 𝐶𝑉 2 𝑓 (1)
In multi supply system, the logic gates on the noncritical paths can be operated
with low VDD is called as VDDL and gates on critical path can be operated with
high VDD which is called as VDDH to maximise the speed performance. This
phenomenon is called as clustered voltage scaling [3]. When a logic ‘1’ signal of
VDDL block drives the VDDH logic gates block, the P-Channel Metal Oxide
Semiconductor Field Effect Transistors (PMOSFETs) of VDDH operating logic
gate blocks may not become perfectly ON, may become partially ON or weakly
ON, hence there will be large static current flow in VDDH logic gates blocks. This
will lead to high static power in VDDH logic gate blocks, as shown in Fig. 1. To
prevent this, high static current and static power the level shifter (LS) are needed
to be used. The LS is an interfacing circuit which shifts the VDDL signal equal to
VDDH signal. When a logic ‘1’ signal of VDDH gates block drives the VDDL
logic gates block, MOSFET of logic gate blocks become partially ON or weakly
ON, hence no level shifter is needed and even if required, cascading of inverters
are enough but additional circuits are needed.
However, the use of LSs induces the area overhead with some LS Power
consumption and LS delay. The primary challenge in multi VDD system is
designing efficient LS to reduce the overheads cause by the inclusion of LSs. In
this research we have introduced novel LS, which utilize CMOS topology to design
new LS. The introduced “Level-up/level-down Voltage Level Shifter (VLS) for
The comparison is done with level shifting logic gates and dedicated level
shifters. The DCVSL LS is a ratio circuit which has current contention between
MP2-MP3 and MN1-MN2 MOS transistors. As a result “pull-up and pull-down”
impedances have to be suitably designed [5, 6]. This circuit is suitable to perform
level shifting from threshold level to VDDH, useful to perform voltage level-up or
level-down can be done by changing VDDH to VDDL or cascading of inverters are
needed to be adopted [7-13].
The LSs [14-16] describe the LSs applications in various applications like IoT,
Biomedical and Bio sensors based on the essentiality of LSs in allied applications.
The issues pertaining to “Wilson-current mirror-based level shifter” were addressed
in [17], which use multi-VT transistors for attaining wide conversion range. As the
focus is on design performance optimization in low voltage operations, the existing
work is needed to be optimised for wide range of supply voltages. The LSs in [18,
19] developed using reduced swing inverter, probably which suffers from weak
pull-up and delays are not possible to scale by power supply voltages. In level
shifter in [20] is a multi-stage topology that reduces the contention, effectively
evades the serious upsizing of pull-down network, which results in increased circuit
complexity and delay in comparison with single stage LSs.
The LSs in [21, 22] are also based on DCVSL structure that consumes high static
power when the input is stable. However, the current disputation between “pull-up and
pull-down” network during output switches is more, which results in more delay and
dynamic power, the impact of this is much more when VIN is lower than threshold.
Sizing is a very straightforward remedy to stabilise the strength between “pull-up and
pull-down” network, which penalties the impractical aspect ratios [23, 24].
Gundala et al. [25] proposed an LS called as : Active Volta Level Shifter”
(AVLS) that performs “level-up shift, level-down shift, and blocking”. The actual
level shifting action takes place by two transmission gates. There is a possibility of
both transmission gates ON when VIN equals to VDDL. Hence there may be higher
static power consumption at any logic level of its input.
Luo et al. [26] proposed an LS called as “Modified Wilson Current Mirror
Hybrid Buffer” (MWCMHB) useful for wide range voltage shifting applications,
and incorporated a provision called bidirectional level shifting, balancing
transitions, poses current mirror circuit having higher static power consumption,
and second stage is a NOR gate with output inverter, employ dual power supplies.
Being dual power supplies the routing congestion may occur at physical design. In
[11, 27] the source currents are enabled/disabled based on the input transitions,
which extensively shrinks the standby power, increase in delay and energy. The
LSs proposed in [28-33] have a considerable level of contention between the pull-
up and the pull-down circuitry in the logic ‘High-to-Low transition’ of the VIN
leads to increase in delay and power consumption.
where Vth is the thermal voltage, n is the sub threshold swing coefficient constant,
η is the Drain-Induced Barrier Lowering coefficient, is the linearized body effect
coefficient, and I0 is given by Eq. (3).
𝑊 2 1.8
𝐼0 = 𝜇𝑜 𝑐𝑜 𝑉𝑡ℎ 𝑒 (3)
𝐿
Fig. 9. Average power of VLS @ CL and VDDH = 1.2 V and VDDL = 0.2 V.
Fig. 10. Delay of VLS @ CL and VDDH = 1.2 V and VDDL = 0.2 V.
Fig. 11. Layout of the voltage LS for dual supply applications [28].
Fig. 12. Layout of the two stage LS using Wilson current mirror [30].
From Figs. 15 (a) and (b), it is clear that the area in terms of λ and µm have least
values for the proposed method than the other three types due to the smaller device
count and the resulting area implementation for the proposed VLS circuit. Here, λ
based area is calculated with the help of Lambda based minimum width and spacing
rules as specified in many literatures [39-41]. The advantage of lambda-based
design rule is that any change in a design to test with lower technology models or
scaling down to lower technology makes it easier to use the existing architecture to
measure the area with the new (low technology) without any change. Here, only
the value of lambda is going to vary (lower). The values in µm are also given in
Table 2 for comparison.
(a) Area in λ2 .
(b) Area in µm2
5. Conclusion
The proposed Input controlled level-up/level-down voltage level shifter has a
capability of performing two types of shifts (Level up or Level down) without any
control signal. The type of shifting either up or down takes place automatically just
based on its input VIN. The design has been characterized by performing power,
delay, and load analysis. These analyses proved that the design consumes very low
power and delay, even under varying load conditions. The average up and down shift
active power is 9nW, quite appropriate for all nano-scale applications. The average
power consumed by the proposed LS is nearly 1/10th of design [28]. The designs in
[31, 33, 29] have delay of 16ns, 21.2ns and 3.5ns respectively, whereas, proposed
design have 2.5ns for level up operation. The overall Power Delay Product (PDP) of
the proposed design has 0.03fJ. The proposed design occupies nearly 50% less layout
area than [28, 30, 32] at 90 nm technology.
Abbreviations
CMOS Complementary Metal Oxide Semiconductor
DCVSL Differential Cascade Voltage Switch Logic
FET Field Effect Transistor
NMOS n-channel Metal Oxide Semiconductor FET
PDP Power Delay Product
PMOS p-channel Metal Oxide Semiconductor FET
VDDH Higher supply Voltage
VDDL Lower Supply Voltage
VIN Input Voltage
VLS Voltage Level Shifter
VOUT Output Voltage
References
1. Blaauw, D.; Sylvester, D.; Dutta, P.; Lee, Y.; Lee, I.; Bang, S.; Kim, Y.; Kim,
G.; Pannuto, P.; Kuo, Y.-S.; Yoon, D.; Jung, W.; Foo, Z.; Chen, Y.-P.; Oh, S.;
Jeong, S.;and Choi, M. (2014). IoT design space challenges: Circuits and
systems. 2014 Symposium on VLSI Technology, 1-2.
2. Gutnik; V.; and Chandrakasan, A.P. (1997). Embedded power supply for low
power DSP. IEEE Transactions on VLSI Systems, 5(4), 425-435.
3. Usami, K.; Igarashi, M.; Minami, F.; Ishikawa, T.; Kanzawa, M.; Ichida, M.;
and Nogami, M. (1998). Automated low power technique exploiting multiple
supply voltages applied to a media processor. IEEE Journal of Solid State
Circuits, 33(3), 463-472.
4. Diril, A.U.; Dhillon, Y.S.; Chatterji, A.; and Singh A.D. (2005). Level-shifter
free design of low power dual supply voltage CMOS circuits using dual
threshold voltages. IEEE Transactions on VLSI Systems, 13(9), 1103-1107.
5. Romli, N.B.; Mamun, N.; Arif, M.; Bhuiyan, S.; and Husain, H. (2006). Design
of a low power dissipation and low input voltage range level shifter in Cedec
0.18-μm CMOS process. World Applied Sciences Journal, 19(8), 1140-1148.
6. Chen, T.-H.; Chen, J.; and Clark, L.T. (2006). Subthreshold to above threshold
level shifter design. Journal of Low Power Electronics, 2(2), 251-258.
21. Gundala, S. (2019). A leakage power aware transmission gate level shifter.
International Journal of Engineering and Advanced Technology, 8(4), 1527-1530.
22. Matsuzuka, R.; Hirose, T.; Shizuku, Y.; Shinonaga, K.; Kuroki, N.; and Numa,
M. (2017). An 80mV to 1.8 V conversion range low energy level shifter for
extremely low voltage VLSIs. IEEE Transactions on Circuits and Systems I,
64(8), 2026-2035.
23. Gundala, S.; Venkata K. Ramanaiah, V.K.; and Padmapriya, K. (2015). A
novel high performance dynamic voltage level shifter. ARPN Journal of
Engineering and Applied Sciences, 10(10), 4424-4429.
24. Gundala, S.; Ramanaiah, V.K.; and Kesari, P. (2014). High speed energy
efficient level shifter for multi core processors. International Conference on
Circuits, Communication, Control and Computing, Banglore, India, 393-397.
25. Gundala, S.; Ramanaiah, V.K.; and Kesari, P. (2014). Nanosecond delay level
shifter with logic level correction. 2014 International Conference on Advances
in Electronics Computers and Communications, Banglore, India, 26-30.
26. Luo, S.-C.; Huang, C.-J.; and Chu, Y.-H. (2014). A wide-range level shifter
using a modified wilson current mirror hybrid buffer. IEEE Transactions on
Circuits and Systems I, 61(6), 1656-1665.
27. Lütkemeier, S.; and Rückert, U. (2010). A sbthreshold to above- threshold
level shifter comprising a Wilson current mirror. IEEE Transactions on
Circuits and Systems II, 57(9), 721-724.
28. Hosseini, S.R.; Saberi, M.; and Lotfi, R. (2017). A high-speed and power-
efficient voltage level shifter for dual-supply applications. IEEE Transactions
on Very Large Scale Integration (VLSI) Systems, 25(3), 1154-1158.
29. Lotfi, R.; et al. (2018). Energy-efficient wide-range voltage level shifters
reaching 4.2 fJ/transition. IEEE Solid-State Circuits Letters, 1(2), 34-37.
30. Maroof, N.; et al. (2017). A fast and energy-efficient two-stage level shifter
using the controlled Wilson current mirror. Journal for Control, Measurement,
Electronics, Computing and Communications, 58(4), 473-478.
31. Zhao, W.; Alvarez, A.B.; and Ha, Y. (2015). A 65-nm 25.1-ns 30.7-fJ robust
subthreshold level shifter with wide conversion range. IEEE Transactions on
Circuits and Systems II, 62(7), 671-675.
32. Gundala, S.; Basha, M.M.; and Vijayakumar, S. (2020). Double current limiter
high performance voltage level shifter for IoT applications. 2020 5th
International Conference on Communication and Electronics Systems
(ICCES), 281-284.
33. Nan, L; Zeng, X.;, Li, W.; Wang, Z.; and Dai, Z. (2018). A single-supply sub-
threshold level shifter with an internal supply feedback loop for multi-voltage
applications. IEICE Electronics Express, 15(6), 1-10.
34. Vijayakumar, S.; and Korah, R. (2014). Circuit level 32 nm, 1-bit MOSSI-ULP
adder: power, PDP and area efficient base cell for unsigned multiplier. IEICE
Electronics Express, 11(6), 1-7.
35. Shanmugam, V.; and Reeba K. (2012). Area and power efficient hybrid
PTCSL MUX Design. European Journal of Scientific Research, 83(1), 39-52.