Microsoft Word - ETASR_V12_N4_pp8891-8895 Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8891-8895 8891 www.etasr.com Khmailia et al.: Design of a Low Power CMOS Inverter with the VBB Stack Approach Design of a Low Power CMOS Inverter with the VBB Stack Approach Samah Khmailia Department of Physics Faculty of Science University of Tunis El Manar Tunis, Tunisia samehkhmailia@gmail.com Jilani Rouabeh Department of Technology Faculty of Science University of Tunis El Manar Gafsa, Tunisia rouabeh.jilani@yahoo.com Abdelkader Mami Department of Physics Faculty of Science University of Tunis El Manar Tunis, Tunisia abdelkader.mami@fst.utm.tn Received: 12 February 2022 | Revised: 21 February 2022 | Accepted: 16 April 2022 Abstract-Due to the exponential advancement in nanotechnology devices, low energy consumption has become a significant concern of researchers and VLSI designers. In this paper, the Variable body bias (VBB) and the stack approach are used simultaneously to reduce the leakage power of a CMOS inverter in standby mode. This new technique is called the VBB stack approach. The simulations have been conducted on the LT spice simulator. The power evaluation has been determined and compared between the conventional approach, the stack approach, and the VBB stack approach. The results have demonstrated the performance of the VBB stack approach. The power consumption in the VBB stack approach has decreased by 23% compared to the conventional approach and by 10% compared to the stack approach. Keywords-CMOS inverter; VLSI; power dissipation; leakage current; low power; VBB stack approach I. INTRODUCTION The continuous demand for lightweight portable devices such as laptops, tablets, and smart phones has increased the need to reduce the processor size [1, 2]. VLSI designers have made possible to put millions of transistors on a single chip while maintaining good device performance [3]. The oxide thickness of transistor has been shrinking. The channel length has also become short, but produces increased leakage current in standby mode. With the continuous decreasing of technology size, the leakage power in standby mode is becoming the main contributor of total power consumption. The high power dissipation has become a major issue in digital circuit design with each new technology emergence [4]. The main effect of high leakage power is the increase in temperature, which in some cases leads to device breakdown [5]. Power optimization has become an important research field for VLSI designers. Many techniques have been proposed, each one delivering a new way to decrease power leakage, but the shortcomings of each push researchers to try to discover a better approach. In this work, the stack approach and the VBB approach are used simultaneously to ensure the best power optimization of a CMOS inverter. The proposed technique is called the VBB stack approach. II. LITERATURE REVIEW At all levels of VLSI design steps, many approaches heve been proposed to reduce power dissipation. Authors in [6] presented low power half adder, full adder, half substractor, and full substractor using the CMOS technology. Authors in [7] developed a full adder design using the modified GDI technique to reduce power consumption. Authors in [8] analyzed a MOS transistor to decrease leakage current in standby mode based on forward and reverse body biasing techniques. Authors in [9] used the substrate biasing technique to reduce the standby leakage power of a nanometer scale CMOS circuit. The substrate biasing VSB of a CMOS transistor has a direct effect on the threshold voltage Vth, which can be controlled by varying the substrate potential. The equation that shows the impact of substrate biasing on threshold voltage is: ��� = ���� + �( |2∅ − �� | − |2∅ |) (1) where ∅ is the flatband voltage, � is the substrate effect coefficient, ���� is the threshold voltage with zero body biasing, and �� is the source to substrate voltage. The threshold voltage is directly proportional with the body potential VBB (VSB). Figure 1 illustrates the conventional CMOS connections and the VBB connections. Corresponding author: Samah Khmailia Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8891-8895 8892 www.etasr.com Khmailia et al.: Design of a Low Power CMOS Inverter with the VBB Stack Approach Fig. 1. Body connections: (a), (b) conventional bias, (c), (d) VBB technique. Commonly, the body (substrate) of an NMOS transistor is connected to the ground and the body of the PMOS transistor is connected to VDD [10]. With the VBB technique, the body terminal of PMOS is connected to the positive voltage, and the body terminal of NMOS is connected to the negative voltage to increase the threshold voltage. Authors in [11] presented the stack technique as a method to reduce leakage power. This method is based on the principle that each transistor can be replaced by two half size transistors associated in series [11, 12]. Fig. 2. The stacking technique. If only one transistor "N" is in standby mode, the potential in the source node is zero, hence there is not a self-reverse bias effect and the leakage current increases. When both transistors N1 and N2 are in standby mode, a low drain current occurs. A positive potential VN appears. The gate to source voltage (VGS1) of transistor N1 reduces because the sub-threshold current reduces which has as effect an increase to the threshold voltage. This effect is known as the stacking effect [13]. Authors in [11] used the sleepy stack approach to reduce the power consumption of a CMOS inverter. The sleepy stack approach consists of dividing each transistor into two half size transistors. A sleep transistor is placed in parallel to one of the transistors. During the standby mode, the sleep transistors are put off. The threshold voltage increases and the leakage current reduces. The increase of area is the major penalty of this approach. Authors in [14] used the MTCMOS technology to design of a low power XNOR gate. This method is based on adding two sleep high threshold voltage transistors to the principal circuit. While low threshold transistors are active to realize the principal function, the high threshold transistors are turned on. During the stand-by mode, the high threshold transistors are turned off, hence the subthreshold leakage current is cut off and the static power decreases. The MTCMOS technique reduces the leakage current but increases the propagation delay time. Fig. 3. The sleepy stack approach. Fig. 4. The MTCMOS technique. Authors in [15] proposed the use of the LECTOR technique to cut down the leakage current of a NAND gate [17]. In this approach, two leakage control transistors LC1 and LC2 are introduced between the pull up and pull down network. The source of each leakage transistor controls the gate terminal of the other. The introduction of leakage control transistors increases the resistance between Vdd and Gnd, thus reducing the leakage current. The lector technique reduces the leakage current but increases the area and the propagation delay time. III. LOW POWER INVERTER USING THE VBB STACK APPROACH The inverter is the basic design element of digital circuits. A good understanding of its behavior is necessary in order to build more complex circuits. It has one input and one output. Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8891-8895 8893 www.etasr.com Khmailia et al.: Design of a Low Power CMOS Inverter with the VBB Stack Approach The output is always the complement of the input. Which means a logical 0 at the input produces a logical 1 at the output and vice versa [10, 11]. Generally, a CMOS logic circuit consists of a symmetrical and complementary pairs of PMOS blocks placed on the circuit as a pull up Network (PUN) and a NMOS block connected as a PULL- down network (PDN). The conventional CMOS inverter consists of two complementary transistors connected to the same input. Fig. 5. The LECTOR technique. Fig. 6. The conventional CMOS inverter. In active mode, the transistors behavior is comparable to that of switches. When the input is at low logic level (0), the P- MOS transistor is conductive and acts like a closed switch, the Current flows from the Vdd supply to the output. When the inverter input is at high logic level (1), the PMOS transistor is cut off, disconnecting the output from the positive voltage Vdd and the N-MOS transistor is conductive and acts like a closed switch, pulling the output to low level (0). The power dissipation in the CMOS inverter consists of two components, the dynamic and the static power. Dynamic power permits to charge and discharge the load during the active mode [12]: 2 dyn L dd P C V f= ∗ ∗ (2) where �� is the load capacitance, ��� is the supply voltage. Ideally a PMOS transistor must be in standby mode when Vgs < Vth (static power), Vgs is the voltage from the gate to the source, and Vth is the threshold voltage. Due to reduced transistor size, the gate oxide and the channel length become short which produces leakage power or static power. As reported by VLSI designers, static power may in the future dominate the total power consumed by CMOS devices as technology size continue to shrink [10, 16] ������� = ��� ∗ �������� (3) where �� !"!# is the leakage current during the standby mode. A major part of the leakage current is caused by the sub- threshold current [12]. ��$% = ��$%� &'()&*+ ,&- .1 − )&0(&- 1 (4) where �2340 is the zero bias electron mobility: �2340 = 6 77. �9: ;<= > . �?2 (5) where �? is the thermal voltage, Vth is the threshold voltage, η is the sub- threshold swing coefficient, Vgs is the transistor gate to source voltage, Vds is the drain to source voltage, μeff is the electrons mobility, �9: is the gate oxide capacitance per unit area, and < and = are the width and length of the channel respectively. (a) (b) Fig. 7. Low power CMOS inverter (a) stack method, (b) VBB method. The sub- threshold current Isub is inversely proportional to the threshold voltage ��� . Therefore, an improvement of the threshold voltage may resolve the problem of growing leakage power. The stack and VBB techniques are frequently used to improve the threshold voltage in consequence decrease the leakage power in VLSI circuits and devices. Each method Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8891-8895 8894 www.etasr.com Khmailia et al.: Design of a Low Power CMOS Inverter with the VBB Stack Approach increases the substrate potential in a different way which decreases the substrate leakage current. To take advantage of the stack and VBB approaches, we have used them together in the same circuit. The new technique is called the VBB stack approach. Each transistor in the conventional inverter is divided into two equal transistors. A positive potential is applied on the body terminal of PMOS transistors. A negative potential is applied on the body terminal of NMOS transistors. A self-reverse bias potential appears in all transistors, thus the threshold voltage increases and the leakage current decreases. The schematic circuit design and the simulation of the inverter with the VBB stack approach have been done on the LT-Spice tool. Fig. 8. Design of the inverter using the VBB stack approach. IV. RESULTS AND DISCUSSION To evaluate the performance of the proposed method, the inverter is simulated on the LT-Spice with supply voltage variation from 0.2V to 1V. Simulations have been carried out for the inverter in the conventional approach, the stack approach, and the VBB stack approach. Power dissipation and delay have been calculated with variation in Vdd . TABLE I. POWER DISSIPATION AND DELAY VARIATIONS AS FUCTIONS OF THE SUPPLY VOLTAGE Supply voltage (V) 0.2 0.4 0.6 0.8 1 Conventional approach Power dissipation (nW) 111 615 1604 3044 4897 Delay (nS) 7.757 7.757 7.510 7.493 7.441 Stack approach Power dissipation (nW) 82 443 1214 2460 4197 Delay (nS) 7.645 7.561 7.544 7.510 7.491 VBB stack approach Power dissipation (nW) 71 368 997 2036 3535 Delay (nS) 7.634 7.562 7.561 7.527 7.508 From the obtained results, it is observed that power dissipation increases simultaneously when the supply voltage increases. Power dissipation in the VBB stack approach is minimal in comparison with the other techniques, but the delay increases slightly due to the increase in the threshold voltage of the device. We can observe from Table I that by using the stack technique we can save 27% power as compared to the conventional method, but in this case the delay is increased slightly. By using the proposed VBB stack technique, power dissipation is approximately 37% less than in the conventional method, but with a slight increase in delay of 2%. Figure 9 shows the power dissipation characteristics as functions of the supply voltage. Fig. 9. Power dissipation characteristics. The main achievement is the inverter with the proposed VBB stack approach is that it shows much better performance in terms of consumed power as compared to the conventional and stack methods. Hence, the proposed VBB stack approach technique is an optimum alternative method to realize low power devices with accepted performance in VLSI design. V. CONCLUSION The demand for low power VLSI circuits increases due to the continuous development of wireless technologies which require a limited source of power. Power dissipation is becoming a major issue in today’s digital circuits. So, the main objective of VLSI designers is to reduce power consumption as much as possible. In this paper, the VBB stack approach is presented as a solution to prevent the problem of leakage power. A comparative analysis of logic CMOS inverters that were designed with the conventional technique, the stack technique, and the VBB stack technique was conducted in this paper. To evaluate the performance of the VBB stack approach, all CMOS inverter designs have been simulated in LT-Spice, with supply voltage variation from 0.2V to 1V. From the experimental results, it is observed that power dissipation in the VBB stack technique is much less than in the other techniques with a minor increase in delay. REFERENCES [1] H. Ghabri, D. B. Issa, and H. Samet, "Performance Optimization of 1-bit Full Adder Cell based on CNTFET Transistor," Engineering, Technology & Applied Science Research, vol. 9, no. 6, pp. 4933–4936, Dec. 2019, https://doi.org/10.48084/etasr.3156. [2] M. Ma, Z. Li, and Z. Yao, "Current-mode CMOS Active Inductor with Applications to Low-Voltage Oscillators," Engineering, Technology & Applied Science Research, vol. 3, no. 6, pp. 540–543, Dec. 2013, https://doi.org/10.48084/etasr.317. [3] W. W. Kai, N. binti Ahmad, and M. H. bin Jabbar, "Variable Body Biasing (VBB) based VLSI Design Approach to Reduce Static Power," Engineering, Technology & Applied Science Research Vol. 12, No. 4, 2022, 8891-8895 8895 www.etasr.com Khmailia et al.: Design of a Low Power CMOS Inverter with the VBB Stack Approach International Journal of Electrical and Computer Engineering, vol. 7, no. 6, pp. 3010–3019, Dec. 2017, https://doi.org/10.11591/ijece.v7i6. pp3010-3019. [4] B. L. Dokic, "A Review on Energy Efficient CMOS Digital Logic," Engineering, Technology & Applied Science Research, vol. 3, no. 6, pp. 552–561, Dec. 2013, https://doi.org/10.48084/etasr.389. [5] M. J. Rani and S. Malarkkan, "Analysis of Pseudo-NMOS Logic with Reduced Static Power in Deep Sub- Micron Regime," International Journal of Advances in Electronics Engineering, vol. 2, no. 3, pp. 233– 236, Dec. 2012. [6] A. A. Beltran Jr, K. Nones, R. L. Salanguit, J. B. Santos, J. M. R. Santos, and K. J. Dizon, "Low Power NAND Gate–based Half and Full Adder / Subtractor Using CMOS Technique," Journal of Robotics and Control, vol. 2, no. 4, pp. 252–257, Jul. 2021, https://doi.org/10.18196/jrc.2487. [7] H. Pasha, "Design and Analysis of Low Power High Speed Full Adder Cell using Modified GDI Technique at 90nm Technology," vol. 4, no. 8, pp. 43–48, Aug. 2016, https://doi.org/10.17148/IJIREEICE.2016.4810. [8] P. Kalyani, M. M. Latha, and P. C. Sekhar, "Analysis of MOS transistor behavior with Forward and Reverse Body biasing in Subthreshold region," vol. 13, no. 19, pp. 14236–14240, 2018. [9] J. W. Chun and C.-Y. R. Chen, "Leakage power reduction using the body bias and pin reordering technique," IEICE Electronics Express, vol. 13, no. 3, 2016, Art. no. 13.20151052, https://doi.org/10.1587/elex. 13.20151052. [10] M. Zabeli, "The most significant MOSFET parameters impact in CMOS inverter switching characteristics," International Journal Of Circuits, Systems And Signal Processing, vol. 12, pp. 565–572, May 2018. [11] L. Kumre, B. P. Shrivastava, and N. Rai, "Comparative Analysis Of Cmos Inverter For Low Leakage Power," International Journal Of Scientific & Technology Research, vol. 8, no. 9, pp. 1598–1601, Sep. 2019. [12] M. V. Ramanaiah, S. Alluri, B. R. Naik, and N. S. S. Reddy, "Transistor sizing of CMOS VLSI Circuits in Deep Submicron Technology," International Journal of Innovative Technology and Exploring Engineering, vol. 8, no. 11S2, pp. 15–29, Oct. 2019, https://doi.org/ 10.35940/ijitee.K1004.09811S219. [13] M. Gangele and K. P. Patra, "Comparative Analysis of Lector and Stack Technique to Reduce the Leakage Current in CMOS Circuits," International Journal of Research in Engineering and Technology, vol. 4, no. 7, pp. 92–100, Jul. 2015, https://doi.org/10.15623/ijret.2015. 0407015. [14] S. Nehra and P. K. Ghosh, "Design of a Low Power XNOR gate Using MTCMOS Technique," Advance in Electronic and Electric Engineering, vol. 3, no. 6, pp. 701–710, 2013. [15] N. Hanchate and N. Ranganathan, "LECTOR: a technique for leakage reduction in CMOS circuits," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 12, no. 2, pp. 196–205, Oct. 2004, https://doi.org/10.1109/TVLSI.2003.821547. [16] S. I. Padma, M. Shamila, P. Rahima, and M. A. Devi, "Leakage Power Reduction Techniques for Nanoscale in CMOS VLSI Systems Using Microwind Eda Tool," International Research Journal of Engineering and Technology, vol. 7, no. 5, pp. 2835–2840, May 2020.