Re-create missing negative frequencies?

  • Context: Undergrad 
  • Thread starter Thread starter fleem
  • Start date Start date
  • Tags Tags
    Frequencies Negative
Click For Summary

Discussion Overview

The discussion revolves around the handling of complex frequency domain data that lacks negative frequencies. Participants explore whether it is valid to fill in these missing negative frequencies using the complex conjugate of the corresponding positive frequency values, particularly in the context of real-world measuring instruments.

Discussion Character

  • Technical explanation
  • Conceptual clarification

Main Points Raised

  • One participant inquires if they can fill in negative frequency data with the complex conjugate of positive frequency data, suggesting an example with specific values.
  • Another participant references a previous thread that seems to confirm the initial inquiry, implying that using complex conjugates is acceptable.
  • A third participant asserts that if the original time-domain signal is real, then using the complex conjugate for negative frequencies is valid, citing the "reality condition" from the Fourier transform context.

Areas of Agreement / Disagreement

There appears to be a general agreement among participants that filling in negative frequencies with complex conjugates is acceptable, provided the original time-domain signal is real. However, the discussion does not explore any potential counterarguments or alternative methods.

Contextual Notes

The discussion assumes that the original time-domain signal is real without explicitly verifying this condition, which may affect the validity of the proposed approach.

fleem
Messages
440
Reaction score
0
I am being given complex frequency domain data from a measuring instrument. The data does not include negative frequencies. I have some existing software to process this data, but that software currently expects the spectrum data to also include negative frequencies. I want to continue using that software for now without change. So my question is, can I fill in the missing negative frequency data with the complex conjugate values of the corresponding positive frequency data? For example, if the positive frequency for 2GHz is 3+4j, then can I assume the value for -2GHz is 3-4j. (Note that I realize this may be wasteful of CPU because redundant data will be processed, but that's not an issue at the moment--it might be later). Also, I don't care if absolute magnitudes are not preserved, as long as any scaling is constant. Any help, ramblings, guesses will be appreciated.
 
Physics news on Phys.org
Provided the original time-domain signal is real, not complex (which it probably is in a real-world measuring instrument), then yes you can.

(Find the words "reality condition" in the Wikipedia article Fourier transform.)
 
DrGreg said:
Provided the original time-domain signal is real, not complex (which it probably is in a real-world measuring instrument), then yes you can.

(Find the words "reality condition" in the Wikipedia article Fourier transform.)

Great. Thanks for the help!
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
1K
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 15 ·
Replies
15
Views
1K
Replies
17
Views
6K
Replies
1
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 11 ·
Replies
11
Views
5K