# 0.7 Dft and fft: an algebraic view  (Page 4/5)

 Page 4 / 5

## Matrix notation

We denote the $N×N$ identity matrix with ${I}_{N}$ , and diagonal matrices with

${diag}_{0\le k

The $N×N$ stride permutation matrix is defined for $N=KM$ by the permutation

${L}_{M}^{N}:\phantom{\rule{4pt}{0ex}}iK+j↦jM+i$

for $0\le i . This definition shows that ${L}_{M}^{N}$ transposes a $K×M$ matrix stored in row-major order. Alternatively, we can write

${L}_{M}^{N}\phantom{\rule{3pt}{0ex}}\text{:}\phantom{\rule{3pt}{0ex}}i\phantom{\rule{3pt}{0ex}}↦\phantom{\rule{3pt}{0ex}}\mathrm{iM}\phantom{\rule{3pt}{0ex}}\text{mod}\phantom{\rule{3pt}{0ex}}N-1,\phantom{\rule{5pt}{0ex}}\text{for}\phantom{\rule{3pt}{0ex}}0\le i

For example ( $·$ means 0),

${L}_{2}^{6}=\left[\begin{array}{cccccc}1& ·& ·& ·& ·& ·\\ ·& ·& 1& ·& ·& ·\\ ·& ·& ·& ·& 1& ·\\ ·& 1& ·& ·& ·& ·\\ ·& ·& ·& 1& ·& ·\\ ·& ·& ·& ·& ·& 1\end{array}\right].$

${L}_{N/2}^{N}$ is sometimes called the perfect shuffle.

Further, we use matrix operators; namely the direct sum

$A\oplus B=\left[\begin{array}{c}A\\ & B\end{array}\right]$

and the Kronecker or tensor product

$A\otimes B={\left[{a}_{k,\ell }B\right]}_{k,\ell },\phantom{\rule{1.em}{0ex}}\text{for}\phantom{\rule{4.pt}{0ex}}A=\left[{a}_{k,\ell }\right].$

In particular,

${I}_{n}\otimes A=A\oplus \cdots \oplus A=\left[\begin{array}{c}A\\ & \ddots \\ & & A\end{array}\right]$

is block-diagonal.

We may also construct a larger matrix as a matrix of matrices, e.g.,

$\left[\begin{array}{cc}A& B\\ B& A\end{array}\right].$

If an algorithm for a transform is given as a product of sparse matrices built from the constructs above, then an algorithm for the transpose orinverse of the transform can be readily derived using mathematical properties including

$\begin{array}{cc}{\left(AB\right)}^{T}={B}^{T}{A}^{T},\hfill & {\left(AB\right)}^{-1}={B}^{-1}{A}^{-1},\hfill \\ {\left(A\oplus B\right)}^{T}={A}^{T}\oplus {B}^{T},\hfill & {\left(A\oplus B\right)}^{-1}={A}^{-1}\oplus {B}^{-1},\hfill \\ {\left(A\otimes B\right)}^{T}={A}^{T}\otimes {B}^{T},\hfill & {\left(A\otimes B\right)}^{-1}={A}^{-1}\otimes {B}^{-1}.\hfill \end{array}$

Permutation matrices are orthogonal, i.e., ${P}^{T}={P}^{-1}$ . The transposition or inversion of diagonal matrices is obvious.

The DFT decomposes $\mathcal{A}=\mathbb{C}\left[s\right]/\left({s}^{N}-1\right)$ with basis $b=\left(1,s,\cdots ,{s}^{N-1}\right)$ as shown in [link] . We assume $N=2M$ . Then

${s}^{2M}-1=\left({s}^{M}-1\right)\left({s}^{M}+1\right)$

factors and we can apply the CRT in the following steps:

$\begin{array}{ccc}& & \mathbb{C}\left[s\right]/\left({s}^{N}-1\right)\hfill \\ & \to & \mathbb{C}\left[s\right]/\left({s}^{M}-1\right)\oplus \mathbb{C}\left[s\right]/\left({s}^{M}+1\right)\hfill \end{array}$
$\begin{array}{ccc}& \to & \underset{0\le i
$\begin{array}{ccc}& \to & \underset{0\le i

As bases in the smaller algebras $\mathbb{C}\left[s\right]/\left({s}^{M}-1\right)$ and $\mathbb{C}\left[s\right]/\left({s}^{M}+1\right)$ , we choose $c=d=\left(1,s,\cdots ,{s}^{M-1}\right)$ . The derivation of an algorithm for ${DFT}_{N}$ based on [link] - [link] is now completely mechanical by reading off the matrix for each of the threedecomposition steps. The product of these matrices is equal to the ${DFT}_{N}$ .

First, we derive the base change matrix $B$ corresponding to [link] . To do so, we have to express the base elements ${s}^{n}\in b$ in the basis $c\cup d$ ; the coordinate vectors are the columns of $B$ . For $0\le n , ${s}^{n}$ is actually contained in $c$ and $d$ , so the first $M$ columns of $B$ are

$B=\left[\begin{array}{cc}{I}_{M}& *\\ {I}_{M}& *\end{array}\right],$

where the entries $*$ are determined next. For the base elements ${s}^{M+n}$ , $0\le n , we have

$\begin{array}{ccc}\hfill {s}^{M+n}& \equiv & {s}^{n}\phantom{\rule{4.pt}{0ex}}\text{mod}\phantom{\rule{4.pt}{0ex}}\left({s}^{M}-1\right),\hfill \\ \hfill {s}^{M+n}& \equiv & -{s}^{n}\phantom{\rule{4.pt}{0ex}}\text{mod}\phantom{\rule{4.pt}{0ex}}\left({s}^{M}+1\right),\hfill \end{array}$

which yields the final result

$B=\left[\begin{array}{cc}{I}_{M}& \phantom{-}{I}_{M}\\ {I}_{M}& -{I}_{M}\end{array}\right]={DFT}_{2}\otimes {I}_{M}.$

Next, we consider step [link] . $\mathbb{C}\left[s\right]/\left({s}^{M}-1\right)$ is decomposed by ${DFT}_{M}$ and $\mathbb{C}\left[s\right]/\left({s}^{M}+1\right)$ by ${DFT\text{-3}}_{M}$ in [link] .

Finally, the permutation in step [link] is the perfect shuffle ${L}_{M}^{N}$ , which interleaves the even and odd spectral components (even and odd exponents of ${W}_{N}$ ).

The final algorithm obtained is

${DFT}_{2M}={L}_{M}^{N}\left({DFT}_{M}\oplus {DFT\text{-3}}_{M}\right)\left({DFT}_{2}\otimes {I}_{M}\right).$

To obtain a better known form, we use ${DFT\text{-3}}_{M}={DFT}_{M}{D}_{M}$ , with ${D}_{M}={diag}_{0\le i , which is evident from [link] . It yields

$\begin{array}{ccc}\hfill {DFT}_{2M}& =& {L}_{M}^{N}\left({DFT}_{M}\oplus {DFT}_{M}{D}_{M}\right)\left({DFT}_{2}\otimes {I}_{M}\right)\hfill \\ & =& {L}_{M}^{N}\left({I}_{2}\otimes {DFT}_{M}\right)\left({I}_{M}\oplus {D}_{M}\right)\left({DFT}_{2}\otimes {I}_{M}\right).\hfill \end{array}$

The last expression is the radix-2 decimation-in-frequency Cooley-Tukey FFT. The corresponding decimation-in-time version isobtained by transposition using [link] and the symmetry of the DFT:

${DFT}_{2M}=\left({DFT}_{2}\otimes {I}_{M}\right)\left({I}_{M}\oplus {D}_{M}\right)\left({I}_{2}\otimes {DFT}_{M}\right){L}_{2}^{N}.$

The entries of the diagonal matrix ${I}_{M}\oplus {D}_{M}$ are commonly called twiddle factors .

The above method for deriving DFT algorithms is used extensively in [link] .

To algebraically derive the general-radix FFT, we use the decomposition property of ${s}^{N}-1$ . Namely, if $N=KM$ then

a perfect square v²+2v+_
kkk nice
algebra 2 Inequalities:If equation 2 = 0 it is an open set?
or infinite solutions?
Kim
y=10×
if |A| not equal to 0 and order of A is n prove that adj (adj A = |A|
rolling four fair dice and getting an even number an all four dice
Kristine 2*2*2=8
Differences Between Laspeyres and Paasche Indices
No. 7x -4y is simplified from 4x + (3y + 3x) -7y
is it 3×y ?
J, combine like terms 7x-4y
im not good at math so would this help me
how did I we'll learn this
f(x)= 2|x+5| find f(-6)
f(n)= 2n + 1
Need to simplify the expresin. 3/7 (x+y)-1/7 (x-1)=
. After 3 months on a diet, Lisa had lost 12% of her original weight. She lost 21 pounds. What was Lisa's original weight?
preparation of nanomaterial
Yes, Nanotechnology has a very fast field of applications and their is always something new to do with it...
can nanotechnology change the direction of the face of the world
At high concentrations (>0.01 M), the relation between absorptivity coefficient and absorbance is no longer linear. This is due to the electrostatic interactions between the quantum dots in close proximity. If the concentration of the solution is high, another effect that is seen is the scattering of light from the large number of quantum dots. This assumption only works at low concentrations of the analyte. Presence of stray light.
the Beer law works very well for dilute solutions but fails for very high concentrations. why?
how did you get the value of 2000N.What calculations are needed to arrive at it
Got questions? Join the online conversation and get instant answers!