aspcol.soundfieldestimation.posterior_mean_omni

aspcol.soundfieldestimation.posterior_mean_omni(p, pos, exp_center, max_order, wave_num, prior_covariance, noise_power, prior_mean=None)

Computes the posterior mean of the spherical harmonic coefficients, under a Gaussian assumption

Parameters:

  • p (ndarray of shape (num_real_freqs, num_mics)) – complex sound pressure for each microphone and frequency

  • pos (ndarray of shape (num_mics, 3)) – positions of the microphones

  • exp_center (ndarray of shape (1, 3)) – expansion center of the spherical harmonics

  • max_order (int) – maximum order of the spherical harmonics

  • wave_num (ndarray of shape (num_real_freqs,)) – wavenumber, defined as w / c where w is the angular frequency and c is the speed of sound

  • noise_power (float) – variance of the noise in the measurement equation. The noise covariance matrix is assumed to be a identity matrix scaled by this value

  • prior_mean (ndarray of shape (num_real_freqs, num_coeffs), optional) – mean of the prior distribution for the sherical harmonic coefficients If not provided, the prior mean is assumed to be zero

  • prior_covariance (ndarray of shape (num_real_freqs, num_coeffs, num_coeffs),) – or ndarray of shape (num_real_freqs,) if the covariance is a scalar matrix or float if the covariance is a scalar matrix, identical for all frequencies.

Returns:

shd_coeffs – posterior mean of the harmonic coefficients

Return type:

complex ndarray of shape (num_real_freqs, num_coeffs)