2010•Unpublished venueRequires access

Asymmetrical Positive Feedback Adiabatic Logic for Low Power and Higher Frequency

V. S. Kanchana Bhaaskaran

Open publisher page 7 citations

Abstract

This paper presents the quasi-adiabatic Asymmetrical Positive Feedback Adiabatic Logic (APFAL) for low power operation through energy recovery technique. The topology of a logic gate defines the logic effort and it determines the gate sensitivity. The APFAL strives to reduce the logic effort of one arm of the 2N2P latch which results in reduced values of adiabatic and non-adiabatic power components. The use of asymmetric complementary functional blocks in the sense-amplifier structure achieves this. Furthermore, the APFAL incurs reduced transients and minimized floating node problems. It is a diode-free and dual rail logic offering both the true and complementary outputs. It achieves significant reduction in switched capacitance resulting in faster response. Efficient energy recovery is achieved for frequency range of up to 500 MHz. The need for reduced interconnects and realization of less leakage are the added advantages. Validation is done through full-custom designed arithmetic circuits. Comparison with static CMOS and PFAL circuits are made to validate the design. In post-layout simulations, the 8-bit APFAL multiplier achieves an adiabatic gain of 14.91 at 100 MHz to 6.45 at 500 MHz against the static CMOS counterpart. Energy savings of 27% and 22.5% are achieved against the optimized PFAL 4-bit CLA and 8-bit multiplier equivalent circuits respectively, at 500 MHz.

About this research paper

What this paper is about

This paper presents the quasi-adiabatic Asymmetrical Positive Feedback Adiabatic Logic (APFAL) for low power operation through energy recovery technique. The topology of a logic gate defines the logic effort and it determines the gate sensitivity. The APFAL strives to reduce the logic effort of one arm of the 2N2P latch which results in reduced values of adiabatic and non-adiabatic power components. The use of asymmetric complementary functional blocks in the sense-amplifier structure achieves this. Furthermore, the APFAL incurs reduced transients and minimized floating node problems. It is a diode-free and dual rail logic offering both the true and complementary outputs. It achieves significant reduction in switched capacitance resulting in faster response. Efficient energy recovery is achieved for frequency range of up to 500 MHz. The need for reduced interconnects and realization of less leakage are the added advantages. Validation is done through full-custom designed arithmetic circuits. Comparison with static CMOS and PFAL circuits are made to validate the design. In post-layout simulations, the 8-bit APFAL multiplier achieves an adiabatic gain of 14.91 at 100 MHz to 6.45 at 500 MHz against the static CMOS counterpart. Energy savings of 27% and 22.5% are achieved against the optimized PFAL 4-bit CLA and 8-bit multiplier equivalent circuits respectively, at 500 MHz.

Why it matters

OpenAlex reports 7 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This paper presents the quasi-adiabatic Asymmetrical Positive Feedback Adiabatic Logic (APFAL) for low power operation through energy recovery technique. The topology of a logic gate defines the logic effort and it determines the gate sensitivity. The APFAL strives to reduce the logic effort of one arm of the 2N2P latch which results in reduced values of adiabatic and non-adiabatic power components. The use of asymmetric complementary functional blocks in the sense-amplifier structure achieves this. Furthermore, the APFAL incurs reduced transients and minimized floating node problems. It is a diode-free and dual rail logic offering both the true and complementary outputs. It achieves significant reduction in switched capacitance resulting in faster response. Efficient energy recovery is achieved for frequency range of up to 500 MHz. The need for reduced interconnects and realization of less leakage are the added advantages. Validation is done through full-custom designed arithmetic circuits. Comparison with static CMOS and PFAL circuits are made to validate the design. In post-layout simulations, the 8-bit APFAL multiplier achieves an adiabatic gain of 14.91 at 100 MHz to 6.45 at 500 MHz against the static CMOS counterpart. Energy savings of 27% and 22.5% are achieved against the optimized PFAL 4-bit CLA and 8-bit multiplier equivalent circuits respectively, at 500 MHz.

Key concepts: Adiabatic circuit, Adiabatic process, Pass transistor logic, Logic gate, Electronic engineering, Logic family, CMOS, Electronic circuit

Related papers

Back to paper searchBrowse research topicsOriginal source
Asymmetrical Positive Feedback Adiabatic Logic for Low Power and Higher Frequency — Research Paper | ScholarLens