Year of Publication
Master of Science in Electrical Engineering (MSEE)
Electrical and Computer Engineering
Dr. Himanshu Thapliyal
The growing data-intensive applications that run on IoT edge devices require the circuit to be low-power consumption and energy-efficient for limited resources. As conventional Complementary Metal-Oxide-Semiconductor (CMOS) scales down to the nanometer technology node, it reaches its limits, such as leakage and power consumption. Adiabatic logic and approximate computing are emerging techniques for the low-power circuit. Adiabatic logic can recycle energy which is a promising solution for building energy-efficient circuits. However, the power clock scheme and dual-rail structure of adiabatic logic increase the overall area. Power consumption is further reduced by applying approximate computing while reducing the complexity and size of the circuit.
Therefore, to investigate the benefits of approximate computing combined with adiabatic logic, we propose two adiabatic logic based approximate adders. The proposed approximate adders use the advantage of dual-rail logic to shrink the overall size and reduce energy consumption. The two proposed designs are True Sum Approximate Adder (TSAA) and True Carry-out Approximate Adder (TCAA). TSAA approximates the Carryout based on the accurate Sum, and TCAA approximates the Sum based on the precise Carryout.
We performed simulations using 45nm technology in Cadence Spectre. Comparing with CMOS based accurate mirror adder (AMA) at 100 MHz, a power-saving of 83.26% and energy saving of 66.54% in PFAL based TSAA (PFAL: Positive Feedback Adiabatic Logic) is achieved. Further, we achieved a power saving of 87.22% and an energy saving of 74.43% in PFAL based TCAA compared to CMOS based accurate mirror adder (AMA). It is illustrated that PFAL based TCAA consumes 24.0% less power and energy per cycle compared to PFAL based TSAA.
Further, we have proposed the True Sum Approximate Adder (TSAA) and the True Carry-out Approximate Adder (TCAA) that are energy-efficient and secured against DPA attacks. At 12.5 MHz operating frequency and 45 nm technology node, the DPA-resistant adiabatic TSAA and TCAA achieved power savings of 95.4% and 95.48%, energy savings of 90.80%, and 90.96% in comparison with the standard CMOS AMA.
Digital Object Identifier (DOI)
This work was partially supported by National Science Foundation CAREER Award (No. 1845448) from 2020 to 2021.
Yang, Wu, "Combining Approximate Computing And Adiabatic Logic For Low-Power And Energy-Efficient Iot Edge Computing" (2021). Theses and Dissertations--Electrical and Computer Engineering. 168.