Printer Friendly

A Glitch-Free Novel DET-FF in 22 nm CMOS for Low-Power Application.

1. Introduction

Today, the flip-flops are widely used for data storage. The performance and fault tolerance ability of the devices are precisely affected by the flip-flops reliability, speed, and power consumption. As a result, this is needed to design the flip-flops for lowest power consumption, propagation delay, area, and highest reliability with fault tolerance ability. Present studies have shown that device scaling reduces the device capacitances and the supply voltage requirements, and circuit becomes more vulnerable to the glitches. When particles touch the drain side of a MOSFET, electron hole combinations are generated. The reverse biased electric field produces a drift transient current [1, 2]. Voltage transient as a result of the collected charge is called a transient fault. In memory elements, transient faults may be produced by the preceding combinational circuit glitches.

Low-power consumption may be obtained efficiently by voltage supply scaling. In CMOS designs, power consumption due to the glitches cannot be ignored as the portion of power consumption varies somewhat between 9% and 38% [3]. Today, in digital integrated circuits, the designing of energy efficient designs is one of the great challenges for the researchers [4]. Power consumption in clock distribution network is very significant, which may account 45% of the total system power [5]. Clock network consumes more power; consequently, this is necessary to reduce the number of overall clocks. To reduce the number of clocks, the true single-phase clock (TSPC) technique has been advised with the basic registers [6]. To reduce the clock power consumption, the clock frequency can be scaled down, by sampling the data on both of the falling and rising edges of the clock, without altering the system throughput. The dual edge triggered procedure reduces the 50% power dissipation of the clock network system. However, dual edge triggered designs are more complex as compared to single edge triggered (SET) designs, but this can be more energy efficient [7].

The rest part of this paper is arranged as follows. Section 2 reports the existing designs. We have interpreted the proposed work in Section 3. Section 4 describes the result and analysis. Finally, Section 5 concludes this work.

2. Existing Designs

Dual edge triggered flip-flops provide the equal data rate as single edge triggered flip-flops at the half of the clock frequency, which leads to the reduction in power dissipation of digital synchronous logic designs [8, 9]. Figure 1 shows the dual data rate flip-flop (DDR-FF) which has a lower clock load because of its simple configuration, lower activity factor, and hard edge quality factor [10]. Devarapalli et al. [10] presented a robust LM_C dual edge flip-flop by using C-elements, where the direct clock pulses were used to latch the data for the reduction of clock dynamic power consumption without any additional pulse generator circuit. This flip-flop design offers the more robust solution for dual data rate (DDR) flip-flops due to its simplicity with less number of transistors. Bonetti et al. [11] proposed a dual edge triggered flip-flop, as shown in Figure 2, to overcome the built-in clock overlap threat, by using true single-phase clock circuits instead of an inverted clock and clarify the issue of clock overlapping by using the TSPC circuits and an internal dual feedback structure.

Figure 3 shows the LG_C circuit design [12] which improved with common Latch MUX dual edge triggered flipflops; because of this fact, flip-flops internal circuit node value never changes with the changes in the input value. The LG_C circuit design presented improvement in the energy dissipation. The glitch-free LG_C dual edge triggered flip-flop was designed by three C-element circuits, with two internal latches A and B and one output latch Q, with inverting topology because of the transistor level implementations.

Generally, the C-element structures are used to reduce the switching activity. The 2P-1N three-transistor structure [13] and C-element circuit [14] and their truth tables are shown in Figure 4. The C-element structure is a 3-terminal structure with the two inputs and a single output. The C-element circuit has the special feature that when both inputs are equal, thereupon output switches to its input values; when the inputs are not equal, thereupon its output goes to the high impedance state, which means output will be in its past state. The 2P-1N three-transistor structure is different from the C-element circuit: (1) when the inputs are not equal, then the output may not go to the high impedance state. (2) The C-element circuit has four transistors, but this structure has three transistors.

3. Proposed Novel DET-FF

Present day, the main goal of the researchers is to achieve small area, low power, and high speed in the field of VLSI designing. Therefore, several approaches have been considered by the researchers in VLSI applications to achieve these motives. We have designed a low-power glitch-free novel dual edge triggered flip-flop circuit design, as shown in Figure 5. Existing DET-FF designs [10-12] are designed by using only the C-element circuits. In the proposed design, we have used 2P-1N three-transistor structure with the combination of C-element circuit.

Consequently, we have presented a novel robust low-power glitch-free dual edge triggered flip-flop design by mixed combination of 2P-1N and C-element structures which can work accurately at low power supply and provides the totally glitch-free output which can increase the system efficiency. The proposed circuit model consumes less power and provides glitch-free output. The working of the proposed design is as follows: case 1, the glitch either filter out or propagate to the output node Q; case 2, if glitch propagates to the output node Q, then it is filtered out through the feedback path. We just explain how this design avoids the glitches which may occur at the input node from the preceding combinational circuits.

Initially, let the nodes M and N be M = 0, N = 0, X = 1, and Y = 1, and the output be Q = 1 and clock clk = 0. At the node M, if any glitch occurs from the preceding combinational circuit, then the value of the node M will be changed from 0 to 1. Now for the first 2P-1N structure, the inputs are M = 1 and clk = 0, so for this input combination its output becomes X = 0. For the second 2P-1N structure, the inputs are M = 1 and clk = 1, so for this input combination its output becomes Y = 0. Thus, for the C-element circuit, input combination becomes X = 0 and Y = 0; therefore, C-element circuit's output switches to its input values, then N = 0, and the final output becomes Q = 1. Consequently, we can see that there is no change in the output state.

Now, let the nodes M and N be M = 1, N = 1, X = 0, and Y = 0, and the output be Q = 0 and clock clk= 0. At the node M, if any glitch occurs from the preceding combinational circuit, then the value of the node M will be changed from 1 to 0. For the first 2P-1N structure, the inputs are M = 0 and clk = 0; therefore, its output becomes X = 1. For the second 2P-1N structure, the input combination becomes M = 0 and clk = 1, and for this input combination, its output goes to the high impedance state; therefore, its output is in its previous state, that is, Y = 0. Now for C-element circuit, the input combination becomes X = 1 and Y = 0; therefore, its output goes to the high impedance state means its output is in its past state, that is, N = 1, and then output becomes Q = 0; consequently, again we can see that there is no change in the output state.

Now, let the nodes M and N be M = 0, N = 0, X = 1, and Y = 1, and the output be Q = 1 and clock clk= 1. At the node M, if any glitch occurs from the preceding combinational circuit, then the value of the node M will be changed from 0 to 1. Now, for the first 2P-1N structure, the inputs are M = 1 and clk = 1, so for this input combination, its output becomes X = 0. For the second 2P-1N structure, the inputs are M = 1 and clk = 0, so for this input combination, its output becomes Y = 0. Thus, for the C-element circuit, input combination becomes X = 0 and Y = 0; therefore, C-element circuit's output switches to its input values, then N = 0, and the final output becomes Q = 1. Consequently, again there is no change in the output state.

Now, let the nodes M and N be M = 1, N = 1, X = 0, and Y = 0, and the output be Q = 0 and clock clk= 1. At the node M, if any glitch occurs from the preceding combinational circuit, then the value of the node M will be changed from 1 to 0. For the first 2P-1N structure, the inputs are M = 0 and clk = 1, and for this input combination, its output goes to the high impedance state; therefore, its output is in its previous state, that is, X = 0. For the second 2P-1N structure, the input combination becomes M = 0 and clk = 0, and for this input combination, its output becomes Y = 1. Now, for C-element circuit, the input combination becomes X= 0 and Y= 1; therefore, its output goes to the high impedance state which means that output is in its past state, that is, N = 1, and then the final output becomes Q = 0; consequently, again there is no change in the output state. Thus, the faulty input value is left out without any cost in area, power, and time. Therefore, the proposed unique glitch-free novel dual edge triggered flip-flop design is completely glitch-resistant and has high speed and high efficiency.

4. Results and Analysis

The proposed low-power glitch-free novel dual edge triggered flip-flop design is performed through the SPICE simulator with Predictive Technology Model (PTM) 22 nm CMOS technology [15] with a power supply of 1V. The channel lengths of all of the transistors are fixed to 22 nm. To simulate the circuit, the temperature is set to 27[degrees]C and the clock frequency is fixed to 500 MHz.

In the higher technologies, the operating voltage decreases and the frequency increases. Therefore, we have taken 1 V operating voltage, and the frequency is fixed to 500 MHz. When the operating frequency is high enough, the output will not reduce. The glitch-free novel dual edge triggered flip-flop design is simulated through the SPICE simulator, and the results are calculated and verified. The performance evaluation results are reported in Table 1. We note that the glitch-free novel dual edge triggered flip-flop design has a small number of transistors, the lowest average power consumption, the lowest power delay product (PDP), and the lowest propagation delay as compared to existing dual edge triggered flip-flop designs. Consequently, the proposed glitch-free novel dual edge triggered flip-flop design has high speed and high efficiency in comparisons with existing dual edge triggered flip-flop designs.

Comparative analysis of average power consumption, propagation delay, and area (number of transistors) for different glitch-free dual edge triggered flip-flop designs are shown in Figure 6, which reports that the proposed novel circuit design has the lowest average power consumption, the smallest area, and lowest propagation delay in comparison with other existing dual edge triggered flip-flop designs.

The propagation delay analysis of glitch-free novel dual edge triggered flip-flop is presented in Figures 7 and 8 with different parameters. The propagation delay increases with the frequency increment (from 25 MHz to 1000 MHz), as indicated in Figure 7. To reduce the delay, we should increase power supply voltage but in higher technology (22 nm) the power supply voltage decreases and the frequency increases, and consequently, delay increases with the frequency increment.

Due to the power supply voltage reduction, power consumption decreases but propagation delay increases. The propagation delay increases with the increase in the temperature (from - 100[degrees]C to +100[degrees]C), as indicated in Figure 8. A higher temperature will decrease electrons and holes mobility and threshold voltage also; therefore, delay increases with the increase in the temperature.

The average power consumption analysis of glitch-free novel dual edge triggered flip-flop is presented in Figures 9 and 10. Due to the propagation delay increment, the average power consumption increases with the frequency increment (from 25 MHz to 1000 MHz), as indicated in Figure 9. Here also, due to the propagation delay increment, the average power consumption increases with the temperature increment (from -100[degrees]C to +100[degrees]C), as indicated in Figure 10.

5. Conclusion

We have presented a low-power glitch-free novel dual edge triggered flip-flop which is designed with the mixed combination of 2P-1N and C-element structures. The glitch-free novel dual edge triggered flip-flop design is the novel and unique design because it is constructed by using two fault resistant structures. If any glitch affects one of the structures, then it is corrected by the other structure. The glitch-free novel dual edge triggered flip-flop provides the totally glitch-free output and can increase the system efficiency. The presented novel design can reduce the 50% power consumption and contribute to the total system power savings. By using the fault resistant structure and internal dual feedback structure proposed design obtained robust and static operation. To reduce the input loading, the two structures are merged to share transistors connected with the data input. The proposed glitch-free novel dual edge triggered flip-flop has the lowest power consumption and lowest power delay product (PDP) as compared to the existing dual edge triggered flip-flop designs [10-12]. The proposed design has the smallest number of transistors and consequently occupies a small area and has the lowest propagation delay, thus providing high speed and high efficiency. To compare the novel design with existing designs, we have set 1 V power supply voltage and a system clock frequency of 500 MHz, and the proposed design was implemented with 22 nm CMOS technology and simulated through the SPICE simulator.

https://doi.org/10.1155/2018/2934268

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

[1] C.-M. Hsieh, P.C. Murley, and R. R. O'Brien, "Collection of charge from alpha-particle tracks in silicon devices," IEEE Transactions on Electron Devices, vol. 30, no. 6, pp. 686-693, 1983.

[2] R. C. Baumann, "Soft errors in advanced semiconductor devices-part I: the three radiation sources," IEEE Transactions on Device and Materials Reliability, vol. 1, no. 1, pp. 17-22, 2001.

[3] L. Benini, M. Favalli, and B. Ricco, "Analysis of hazard contribution to power consumption in CMOS IC's," in Proceedings of the International Workshop on Low Power Design, pp. 27-32, Napa, CA, USA, April 1994.

[4] R. G. Dreslinski, M. Wieckowski, D. Blaauw, D. Sylvester, and T. Mudge, "Near-threshold computing: reclaiming Moore's law through energy efficient integrated circuits," Proceedings of the IEEE, vol. 98, no. 2, pp. 253-266, 2010.

[5] H. Kawaguchi and T. Sakurai, "A reduced clock-swing flipflop (RCSFF) for 63% power reduction," IEEE Journal of Solid-State Circuits, vol. 33, no. 5, pp. 807-811, 1998.

[6] Y. Yuan and C. Svensson, "High-speed CMOS circuit technique," IEEE Journal of Solid-State Circuits, vol. 24, no. 1, pp. 62-70, 1989.

[7] M. Alioto, E. Consoli, and G. Palumbo, "DET FF topologies: a detailed investigation in the energy-delay-area domain," in Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS), pp. 563-566, Rio de Janeiro, Brazil, May 2011.

[8] N. Nedovic and V. G. Oklobdzija, "Dual-edge triggered storage elements and clocking strategy for low-power systems," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 13, no. 5, pp. 577-590, 2005.

[9] A. G. M. Strollo, E. Napoli, and C. Cimino, "Analysis of power dissipation in double edge-triggered flip-flops," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 8, no. 5, pp. 624-629, 2000.

[10] S. V. Devarapalli, P. Zarkesh-Ha, and S. C. Suddarth, "A robust and low power dual data rate (DDR) flip-flop using C-elements," in Proceedings of the 11th International Symposium on Quality Electronic Design, IEEE, 978-1-4244-6455-5/10, pp. 147-150, San Jose, CA, USA, March 2010.

[11] A. Bonetti, A. Teman, and A. Burg, "An overlap-contention free true-single- phase clock dual-edge-triggered flip-flop," in Proceedings IEEE International Symposium on Circuits and Systems (ISCAS), pp. 1850-1853, Lisbon, Portugal, May 2015.

[12] S. Lapshev and S. M. R. Hasan, "New low glitch and low power DET flip-flops using multiple C-elements," IEEE Transactions on Circuits and Systems, vol. 63, no. 10, pp. 1673-1681, 2016.

[13] A. K. Pudi N S and M. S. Baghini, "Robust soft error tolerant CMOS latch configurations," IEEE Transactions on Computers, vol. 65, no. 9, pp. 2820-2834, 2016.

[14] D. E. Muller, "Theory of asynchronous circuits," Internal Rep. no. 66, Digital Computer Lab, University Illinois at Urbana-Champaign, Champaign, IL, USA, 1955.

[15] Berkeley predictive technology model," 2016, http://www.ptm.asu. edu/~PTM/.

Sumitra Singar (iD), (1) N. K. Joshi, (1) and P. K. Ghosh (2)

(1) College of Engineering and Technology, Mody University of Science and Technology, Lakshmangarh, Sikar, Rajasthan 332311, India

(2) NSHM Knowledge Campus, Durgapur, West Bengal, India

Correspondence should be addressed to Sumitra Singar; sumitra1988singar@gmail.com

Received 21 November 2017; Accepted 4 February 2018; Published 29 March 2018

Academic Editor: Marco Rossi

Caption: Figure 1: DDR-FF (LM_C) design [10].

Caption: Figure 2: TSPC design [11].

Caption: Figure 3: LG_C design [12].

Caption: Figure 4: (a) 2P-1N structure and its truth table. (b) C-element circuit and its truth table.

Caption: Figure 5: Proposed design with 2P-1N structure and C-element circuit.

Caption: Figure 6: Power, delay, and area comparisons of different DET-FF designs.

Caption: Figure 7: Delay analysis of proposed DET-FF design with different frequencies.

Caption: Figure 8: Delay analysis of proposed DET-FF design with different temperatures.

Caption: Figure 9: Average power consumption analysis of the proposed DET-FF design with different frequencies.

Caption: Figure 10: Average power consumption analysis of the proposed DET-FF design with different temperatures.
Table 1: Result analysis of different dual edge triggered flip-flops.

DET-FFs                Reference   Reference   Reference   Proposed
                         [10]        [11]        [12]       design

[P.sub.avg.] cons.       8.094       7.825       8.988       3.049
([mu]w)
[t.sub.p(D-Q)] (ps)     72.872      65.536      84.297      26.995
[t.sub.p(CLK-Q)]        55.021      43.213      63.281      25.896
(ps)
PDP (fJ)                 0.589       0.513       0.758       0.082
Number of                 28          38          28          18
transistors
COPYRIGHT 2018 Hindawi Limited
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research Article
Author:Singar, Sumitra; Joshi, N.K.; Ghosh, P.K.
Publication:Journal of Nanotechnology
Date:Jan 1, 2018
Words:3053
Previous Article:Assessing the Static and Dynamic Sensitivity of a Commercial Off-the-Shelf Multicore Processor for Noncritical Avionic Applications.
Next Article:Nanobismuth: Fabrication, Optical, and Plasmonic Properties--Emerging Applications.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters