I have data that is a function of frequency. The frequency points are

defined in the logarithmic scale.

I need to perform an inverse Fourier transform on this data. However,

since the frequency points are not linearly-spaced, the results of this

operation are not accurate.

Is there an accurate interpolation algorithm that would give me data

for linearly-spaced frequency points, given data for log-scaled

frequency points?

Approaching the problem from a different angle: have you successfully

implemented your own version of a DFT function for log-spaced data?

Tips, pointers and any help are appreciated!

Thanks for your help,

Nikhil

There are methods for computing the Fourier transform of irregularly

sampled data that should work.

OUP

1. linear-space image converted to log-space

2. interpolation: unevenly spaced x-y data pairs --> evenly spaced data pairs

3. Interpolating data in the space between two polynomials in simulink

4. interpolate irregularly spaced imported data

5. Interpolating logarithmic spaced data

6. Matlab plot command interpolates irregurly spaced data?

7. How the data from user space is passing to kernel space?

8. SCSI WRITE_10, Passing data from User Space to Kernel Space

9. Communicating Data from User Space to Kernel Space

10. printf to pad with spaces BUT trips over when data being padded contains spaces

11. Physical Volumes - Space Allocated Vs Space used by data.

12. Linear spacing of points in log scale

13. Updates to data based on a query (adding spaces before the data)

14. Used/Free Space Data/Transaction Log File

15. Merge Data from Access 2002 has data in blank spaces.

2 post • Page:**1** of **1**