ft_kernel: Fourier transform of kernel

In gbasulto/kerdec: Kernel deconvolution methods

Description

Evaluate Fourier transform of kernels and product kernels.

Usage

ft_kernel(t, ker)

Arguments

t

A vector of size n of a nxd matrix where the Fourier transform of a kernel or a d-product kernel (if t is a matrix) will be evaluated.

ker

Character or number specifying the kernel. All the kernels are symmetric and they have a FT supported on [-1, 1]. The possible choices are: 1 or "sinc" Sinc kernel 2 or "triangular" Kernel whose FT is a triangle 3 or "triweight" Kernel whose FT is proportional to triweight kernel 4 or "tricube" Kernel whose FT is proportional to tricube kernel 5 or "flat-top" A kernel with FT equal to one around zero and decreasing linearly to zero at -1/2 and 1/2

Details

The Fourier transform (FT) of all the kernel functions considered for this package have a bounded support, as it is usual in kernel deconvolution methods. All the kernels are symmetric.

See vignette for detailed examples.

Value

A vector of size n.

Examples

t <- seq(-1.5, 1.5, 0.05)
vals <- ft_kernel(t, "flat-top")
plot(t, vals, type = "l")

gbasulto/kerdec documentation built on June 5, 2019, 10:58 a.m.