Design and analysis of power efficient flip flop and its application in nanometre technology
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The global consumer electronics market is expanding at a rapid rate. Due to a rise in device complexity, speed, and density, power dissipation has become a significant issue for the semiconductor industry. The requirement for lower power Very Large Scale Integration (VLSI) circuits is driven by both technological and commercial requirements. Flip-Flop (FF) is one of the most important and vital element in the design of digital circuits. Switching power accounts for roughly 90% of power consumption in CMOS designs and 75% of overall power consumption in a system-on-chip. Five different types of D flip-flops are investigated in this work. These types are single edge triggered flip-flop, dual edge triggered flip-flop, pulse triggered flip flop, sense amplifier based flip-flop and self-gated flip-flop. All the types were individually studied with its counterpartand#8223;s chapter wise. Single edge, dual edge, pulse triggered and sense amplifier flip-flops are simulated in Tanner EDA SPICE tool at 32 nm CMOS technology node whereas self-gated flip-flop was simulated in HSPICE at 16 nm length and in CMOS and beyond CMOS devices like the graphene nano ribbon, carbon nanotube, FinFET. Performances of flip-flops were analyzed based on area, speed and power. Three new flip-flop designs are proposed specifically targeting power efficient applications. Among these three novel flip-flop designs, two are sense amplifier based flip-flop and one a self-gated flip-flop. To ascertain the circumstances where the proposed design should be used, extensive simulations were performed. In proposed design 1, at nominal operating conditions, power consumption is reduced by 14% and at maximum voltage of 1.1 V by 10%, the overall reduction of 8.3% in Power Delay Product (PDP) is observed at nominal voltage. At 100 MHz of frequency, the power consumption is reduced by 15%. In terms of data activities, the power at 100% activity is reduced by 14.2%, at 75% activity by nearly 12%. The proposed design 2 achieved a reduction of minimum 22.67% in power