FA-91559 / Digital signal filters / Member archive

Polyphase interpolator omits the L gain · case 04

The interpolated signal is L times too quiet because zero-stuffing removed energy.

Member previewVariant 4 · 3 implementations · 7 checks per implementation

Case contract

Input [h, L, samples] (integers). Zero-stuff by L and filter with h, scaled by gain L: y[mL + p] = L sum_j h[jL + p] x[m - j]; return len(samples) * L outputs in time order.

Why this case matters

Interpolators feed DACs and sample-rate converters; gain and phase-indexing slips cause level errors and imaging.

One recorded failure

Sample boundary fixture

This sample comes from the broken implementation of a controlled reproducer.

Boundary fixture	Actual	Expected	Outcome
regression: random interpolator 7	`[9, -12, 9, 3, 3, 13, 0, -4, -8, 5, -7, -1, -19, 14, -17, -3, -10, -15, -5, 5]`	`[36, -48, 36, 12, 12, 52, 0, -16, -32, 20, -28, -4, -76, 56, -68, -12, -40, -60, -20, 20]`	Failed

MEMBER ARCHIVE

The complete case is available to members.

This record includes three runnable implementations, regression fixtures, execution results, and source hashes.

Member access is invitation-based. Sign in with your invited account to inspect the sources.