A Novel SIMD-Optimized Implementation for Fast and Memory-Efficient Trigonometric Computation

TL;DR

This paper introduces a SIMD-optimized trigonometric computation method that is significantly faster and more memory-efficient than existing implementations, especially benefiting resource-constrained devices like low-end FPGAs.

Contribution

The paper presents a novel SIMD-optimized approach for trigonometric functions that outperforms inbuilt and Taylor-based methods in speed and memory usage.

Findings

5x faster than inbuilt C++ functions

Requires no precomputations

Reduces hardware usage significantly

Abstract

This paper proposes a novel set of trigonometric implementations which are 5x faster than the inbuilt C++ functions. The proposed implementation is also highly memory efficient requiring no precomputations of any kind. Benchmark comparisons are done versus inbuilt functions and an optimized taylor implementation. Further, device usage estimates are also obtained, showing significant hardware usage reduction compared to inbuilt functions. This improvement could be particularly useful for low-end FPGAs or other resource-constrained devices.