1: \begin{abstract}

2: In this paper, we provide a solution to the open problem of computing the

3: Fourier transform of a binary function defined over $n$-bit vectors

4: taking $m$-bit vector values.

5: In particular, we introduce the two-modular Fourier transform (TMFT) of a binary function

6: $f:G\rightarrow {\cal R}$, where $G = (\mathbb{F}_2^n,+)$ is the group

7: of $n$ bit vectors with bitwise modulo two addition $+$, and ${\cal R}$ is a finite commutative ring of

8: characteristic $2$.

9: Using the specific group structure of $G$ and a sequence of nested subgroups of $G$,

10: we define the fast TMFT and its inverse.

11: Since the image ${\cal R}$ of the binary functions is a ring,

12: we can define the convolution between two functions $f:G\rightarrow {\cal R}$.

13: We then provide the TMFT properties, including the convolution theorem, which

14: can be used to efficiently compute convolutions.

15: Finally, we derive the complexity of the fast TMFT and the inverse fast TMFT.

16: \end{abstract}