A Configurable CORDIC and PSADIC Fusion Architecture for Nonlinear Edge-Computing

The computation of some crucial and nonlinear functions on the edge has to consider the area and latency. The traditional coordinate rotation digital computer (CORDIC) systems are widely used for these key functions, such as the sine, cosine, or the square root. For the co-calculations of the division and the square root, CORDIC requires more iteration cycles and resources, whereas the power series approximate digital computer (PSADIC) approach can half the iteration cycles. In this context, a configurable CORDIC and PSADIC fusion architecture is proposed here for edge-computing. It integrates five modes of CORDIC and PSADIC to calculate some crucial nonlinear functions. Each mode is simulated and verified on software and hardware, achieving the mean relative error distance (MRED) lower than 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−4</sup>. Under the TSMC 40nm process, the area of proposed architecture is increased by 18.3% compared to the PSADIC one, but the five modes, including all the calculations from CORDIC and PSADIC, are implemented.

Authors

Institutions

Publication Details

Published
2024-10-25
DOI
https://doi.org/10.1109/icicm63644.2024.10814567
Primary Topic
Analog and Mixed-Signal Circuit Design
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A Configurable CORDIC and PSADIC Fusion Architecture for Nonlinear Edge-Computing

Jesús Ureña, Álvaro Hernández, Dian Tian, Jing Zhang et al.
Analog and Mixed-Signal Circuit Design
article

A Configurable CORDIC and PSADIC Fusion Architecture for Nonlinear Edge-Computing

Jesús Ureña, Álvaro Hernández, Dian Tian, Jing Zhang, Ningmei Yu, Minghui Xie
article en

Abstract

The computation of some crucial and nonlinear functions on the edge has to consider the area and latency. The traditional coordinate rotation digital computer (CORDIC) systems are widely used for these key functions, such as the sine, cosine, or the square root. For the co-calculations of the division and the square root, CORDIC requires more iteration cycles and resources, whereas the power series approximate digital computer (PSADIC) approach can half the iteration cycles. In this context, a configurable CORDIC and PSADIC fusion architecture is proposed here for edge-computing. It integrates five modes of CORDIC and PSADIC to calculate some crucial nonlinear functions. Each mode is simulated and verified on software and hardware, achieving the mean relative error distance (MRED) lower than 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−4</sup>. Under the TSMC 40nm process, the area of proposed architecture is increased by 18.3% compared to the PSADIC one, but the five modes, including all the calculations from CORDIC and PSADIC, are implemented.

Universidad de Alcalá (ES), Xi'an University of Technology (CN)
National Natural Science Foundation of China, Ministerio de Ciencia e Innovación
Openalex Percentile: Top 77%
Analog and Mixed-Signal Circuit Design
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.