Enum Companding 

#[non_exhaustive]
pub enum Companding {
    MuLaw {
        mu: f64,
        peak_db: f64,
    },
    Exponential {
        degree: f64,
    },
}

Companding of the envelope at the transmitter and expansion at the receiver; the compressed signal keeps the unit average power.

The expansion is exact without noise but amplifies the noise where the law is flat: the exponential law of degree 2 squeezes every amplitude above about twice the RMS into a narrow range below its maximum, and expanding them multiplies the noise by 20 to 150 — an error floor. A receiver can instead skip the expansion and treat the compression as a distortion (Bussgang gain plus noise), as for clipping.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

MuLaw

μ-law (Wang, Tjhung & Ng 1999): $c(r) = V \ln(1 + \mu r/V) / \ln(1 + \mu)$, with $V$ at peak_db above the average power, times the gain that keeps the unit average power of a Rayleigh envelope. The gain is computed once per block, by a 64-point Gauss–Legendre rule within $10^{-13}$ (relative) of its integral for $\mu/V \le 10^{5}$.

Fields

mu: f64

Compression parameter $\mu$.

peak_db: f64

Reference amplitude $V$ (dB above the average power).

Exponential

Exponential companding (Jiang, Yang & Song 2005): $c(r) = \left[\alpha (1 - e^{-r^{2}})\right]^{1/d}$ with $\alpha = ((d + 2)/d)^{d/2}$, which turns the Rayleigh envelope into one of fixed distribution (uniform power for $d = 2$, a PAPR of 3 dB).

Fields

degree: f64

Degree $d$.

Implementations

impl Companding

pub fn compress(&self, r: f64) -> f64

Compress an envelope. Each call computes the normalisation of the μ-law (64 points); PaprReduction::reduce computes it once per block.

pub fn expand(&self, r: f64) -> f64

Expand an envelope (the inverse of Companding::compress).

Trait Implementations

impl Clone for Companding

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for Companding

impl Debug for Companding

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

Formats the value using the given formatter.

impl PaprReduction for Companding

fn reduce(&self, ofdm: &Ofdm, data: &[C64]) -> Result<(Vec<C64>, usize)>

The time block (without prefix) of one block of data, and its side information.

fn restore_time(&self, block: &mut [C64])

Undo a memoryless transformation of the time block at the receiver (the expansion of companding); nothing by default.

fn data_len(&self, ofdm: &Ofdm) -> usize

Data symbols per block (all the active subcarriers by default).

fn recover( &self, ofdm: &Ofdm, active: Vec<C64>, _side: usize, ) -> Result<Vec<C64>>

The data of a block from the values of its active subcarriers and its side information (despreading only, by default).

impl PartialEq for Companding

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Companding