aspcol.soundfieldestimation.est_ki_rff

aspcol.soundfieldestimation.est_ki_rff(p, seq, pos, pos_eval, samplerate, c, reg_param, kernel_func=None, kernel_args=None)

Estimates the RIR in the frequency domain using random Fourier features Assumes seq is a perfect periodic sequence

Parameters:

  • p (ndarray of shape (M, seq_len)) – sound pressure in time domain at M microphone positions

  • seq (ndarray of shape (seq_len)) – the training signal used for the measurements

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

  • pos_eval (ndarray of shape (num_eval, 3)) – positions of the evaluation points

  • samplerate (int)

  • c (float) – speed of sound

  • reg_param (float) – regularization parameter for kernel interpolation

  • kernel_func (callable) – kernel function to use, defaults to the diffuse Helmholtz kernel.

  • kernel_args (tuple) – additional arguments to pass to the kernel function, defaults to None.

Returns:

est_sound_pressure – estimated RIR per frequency at the evaluation points

Return type:

ndarray of shape (num_real_freqs, num_eval)

References

[uenoKernel2018]