知识卡片
迭代更新旋转因子会积累误差
FFT逐次相乘更新旋转因子可省去每轮重新计算,但浮点误差会随迭代累积,输入越大越需留意。
性能优化因此伴随数值精度代价;预计算因子表是一种可评估的替代。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- 第30章 多项式与快速傅里叶变换 · 查看原书依据
The downside of iteratively updating ω is that round-off errors can accumulate, especially for larger input sizes. Several techniques to limit the magnitude of FFT round-off errors have been proposed, but are beyond the scope of this book. If several FFTs are going to be run on inputs of the same size, then it might be
参考来源
- 位置:《算法导论(第4版)》第30章《多项式与快速傅里叶变换》(OEBPS/chapter030.xhtml,文本字符[40000,44125))。
- 结论依据:原文说明位逆序、FFT电路、精确变换或数值稳定性;相应图和公式已核对。
- 原始内容:The downside of iteratively updating ω is that round-off errors can accumulate, especially for larger input sizes. Several techniques to limit the magnitude of FFT round-off errors have been proposed, but are beyond the scope of this book. If several FFTs are going to be run on inputs of the same size, then it might be