Parametric Homomorphic Deconvolution(Steig​litz–McBride,ARMA)

Parametric homomorphic deconvolution using Steiglitz–McBride (ARMA) to recover time-delay structure from a single-channel signal.

Purpose.

Estimate the propagation (time-delay) response from a single-channel analog-summed microphone signal via homomorphic deconvolution followed by Steiglitz–McBride (ARMA) modeling. Returns the parametric HD magnitude along with ARMA coefficients and poles—compact features ideal for downstream AoA regression.

[z33, vp3, Num3, Den3] = cep_st6(Aord, data0, N, W, MM, novlp, normonoff, plotonoff)

Inputs

• Aord — Steiglitz–McBride ARMA order (both AR and MA set to Aord).
• data0 — Input vector; must be N + novlp*(MM-1) samples.
• N — Frame length (power of two recommended).
• W — Quefrency gate (min delay in samples to pass).
• MM — Ensemble length (frames).
• novlp — Hop size (overlap = N - novlp).
• normonoff — 1 to normalize z33 to unit max.
• plotonoff — 1 to plot HD magnitude and pole/zero diagram.

Outputs

• z33 — [N/2+1 × 1] parametric HD magnitude (DC…Nyquist).
• vp3 — poles.
• Num3, Den3 — ARMA coefficients from stmcb.