Full Custom CORDIC Arithmetic Unit Design
Yijun Dai
Abstract
Yijun Dai
Abstract
Floating-point trigonometric computing is a fundamental operation in navigation systems,3D image processing,radar signal preprocessing and so on.This paper presents a floating point trigonometric computing circuit using a CORDIC algorithm and a full custom layout method,and the output data are compatible with the IEEE-754 signal precision floating-point standard.It describes the CORDIC algorithm principle and chooses a pipeline structure based on the principle of priority on performance.A static circuit structure and a full custom layout are given in the SIMC 0.13μm 1P8M CMOS process.The silicon area of data path is 605 284μm2 and the critical path delay is 3.013ns in the SS(Slow-Slow) corners.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Floating-point trigonometric computing is a fundamental operation in navigation systems,3D image processing,radar signal preprocessing and so on.This paper presents a floating point trigonometric computing circuit using a CORDIC algorithm and a full custom layout method,and the output data are compatible with the IEEE-754 signal precision floating-point standard.It describes the CORDIC algorithm principle and chooses a pipeline structure based on the principle of priority on performance.A static circuit structure and a full custom layout are given in the SIMC 0.13μm 1P8M CMOS process.The silicon area of data path is 605 284μm2 and the critical path delay is 3.013ns in the SS(Slow-Slow) corners.
Key concepts: CORDIC, Computer science, Pipeline (software), Floating-point unit, Computer hardware, Path (computing), Critical path method, Floating point