Floating-point Error Analysis of two Efficient Algorithms used in the Fast Computation of two-dimensional DFTs

Page 1

Page 2

Page 3

Page 4