Skip to content

Generalized spiral points #12

New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Open
wants to merge 5 commits into
base: master
Choose a base branch
from
Open

Conversation

narahahn
Copy link
Member

@narahahn narahahn commented Dec 4, 2017

The so-called generalized spiral points are generated according to Rakhmanov 1991.

  • The azimuth and elevation angles are given in closed form
  • The computation requires the number of microphones. This causes the same inconsistency problem we had before

@narahahn narahahn requested a review from trettberg December 4, 2017 13:43
@trettberg
Copy link
Contributor

Currently we use only quadrature grids:
The grid determines ISHT matrix Y. The SHT matrix is then given by Y^H D, where diagonal matrix D contains the quadrature weights.

This generalized spiral is not a quadrature grid. The SHT matrix must be computed via Y^+.

This makes me wonder:

  1. Should the toolbox contain non-quadrature grids?
    Since Y^+ can be computed almost surely, these grids should be "particularly good" in some sense.

  2. Is the generalized spiral a "particular good" grid?

(IMO merging should wait if this is not clear yet.)

@narahahn
Copy link
Member Author

narahahn commented Dec 5, 2017

Good questions. Unfortunately, I don't have a satisfying answer. The advantages I see so far are just that it has a closed form, and it can be computed for any given number. Clearly, the usage of the generalized spiral grid as a microphone distribution needs to be further studied. Quoting Saff and Kuijlaars,

... one can introduce a parameter C in place of 3.6 and adjust its value appropriately for the application at hand

so I will test the grid for different values of C in the context of sound field analysis. I'll come back when I find out something new.

# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants