Energy Efficient Dynamic Logic Based on TGDI Cell for Portable Electronics

The design of energy-efficient circuits is crucial for portable devices in order to enhance battery life. The delay, power, and area reduction are central themes to make circuits suitable for portable electronics. The Gate Diffusion Input (GDI) technique has emerged in the low-power circuit design paradigm. However, it suffers from a threshold drop at the output, and several modifications subsequently appear in open literature. This study examines the problem of restricted voltage swing at an intermittent node in the current Dynamic Gate Diffusion Input (DGDI), which could result in higher power consumption and delay. The approaches of Dynamic Full Swing Gate Diffusion Input (DFSGDI) and Dynamic Transmission Gate Diffusion Input circuit (DTGDI) are proposed in this paper to determine full swing at the intermittent node. The proposals are verified for functionality and performance using the Cadence Virtuoso tool and the 32nm Carbon Nanotube Field Effect Transistor (CNTFET) model. The XOR gate based on DFSGDI and DTGDI is 23.6% faster than the existing counterpart. The average power consumptions are almost identical in DGDI and DTGDI, while DFSGDI shows an increase of 76.9% as compared to DGDI. The current drawn from the power supply is observed to be <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0.891}\ \boldsymbol{\mu} \mathbf{A},\ \mathbf{1.563}\ \boldsymbol{\mu} \mathbf{A}$</tex>, and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0.857}\ \boldsymbol{\mu} \mathbf{A}$</tex>, respectively, for DGDI, DFSGDI, and DTGDI. Therefore, the product of delay and power (PDP) of the DTGDI outperforms DGDI by 25.2%. Hence, portable systems and low-power applications, which are energy efficient, gain significant benefits from the incorporation of XOR logic based on DTGDI. Monte Carlo simulations show the resilience of the proposals against process variations.