Struct MultirateMatrixConvolver 

pub struct MultirateMatrixConvolver { /* private fields */ }

Implementations

impl MultirateMatrixConvolver

pub fn new( inputs: usize, outputs: usize, low_block_size: usize, factor: usize, routes: &[FirRoute], ) -> Result<Self, DspError>

Hamming-windowed 64 * factor + 1 tap prototype with cutoff at 0.4 times the low sample rate. Route taps are at that low rate.

pub fn with_prototype( inputs: usize, outputs: usize, low_block_size: usize, factor: usize, routes: &[FirRoute], prototype: &[f64], ) -> Result<Self, DspError>

Use an explicit odd, symmetric anti-alias prototype for both rate converters. Its DC gain is used as supplied; interpolation gains by factor. Symmetry makes the reported integer alignment delay valid.

pub fn process( &mut self, inputs: &[&[f64]], outputs: &mut [&mut [f64]], ) -> Result<(), DspError>

Process one high-rate block without allocation. Decimation computes only retained samples; interpolation visits only nonzero polyphases.

pub fn process_buffers<I: AsRef<[f64]>, O: AsMut<[f64]>>( &mut self, inputs: &[I], outputs: &mut [O], ) -> Result<(), DspError>

Process retained channel buffers without allocating slice-view arrays.

pub fn reset(&mut self)

pub fn block_size(&self) -> usize

pub fn low_block_size(&self) -> usize

pub fn factor(&self) -> usize

pub fn input_count(&self) -> usize

pub fn output_count(&self) -> usize

pub fn latency_samples(&self) -> usize

Two linear-phase prototype delays, in high-rate samples.

pub fn tail_samples(&self) -> usize

Conservative high-rate tail length, including both rate converters.

pub fn transforms_per_block(&self) -> (usize, usize)

Trait Implementations

impl Debug for MultirateMatrixConvolver

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.