Python vs MATLAB Parity#

This page tracks the functional parity between the Python adctoolbox package and the original MATLAB implementation. The goal is to keep both versions feature-equivalent.

Last updated: 2026-06-07

Function Mapping#

The table below maps every MATLAB core function to its Python equivalent. Legacy/deprecated MATLAB wrappers (e.g., sineFit → sinfit) are omitted; only the current MATLAB name is listed.

MATLAB

Python

Status

plotspec

analyze_spectrum

Matched

plotphase

analyze_spectrum_polar

Matched

sinfit

fit_sine_4param

Matched

findbin

find_coherent_frequency

Matched

findfreq

estimate_frequency

Matched

alias

fold_frequency_to_nyquist, fold_bin_to_nyquist

Matched

tomdec

analyze_decomposition_time, analyze_decomposition_polar

Matched

errsin

analyze_error_by_phase, analyze_error_by_value

Matched

inlsin

analyze_inl_from_sine

Matched

wcalsin

calibrate_weight_sine

Matched

bitchk

analyze_overflow

Matched

plotwgt

analyze_weight_radix

Matched

ntfperf

ntfperf / ntf_analyzer

Matched

adcpanel

toolset/generate_aout_dashboard, toolset/generate_dout_dashboard

Matched

cdacwgt

not implemented

MATLAB only

ifilter

ifilter / extract_freq_components

Matched

perfosr

perfosr / sweep_performance_vs_osr

Matched

plotres

not implemented

MATLAB only

MATLAB-Only Functions#

The following two MATLAB functions have no Python equivalent yet:

cdacwgt(cd, cb, cp)

Calculate bit weights for a multi-segment capacitive DAC (CDAC) with bridge capacitors and parasitic capacitances. Returns normalized weights and total capacitance.

plotres(sig, bits, ...)

Plot partial-sum residuals of an ADC bit matrix. Scatter-plots residuals between bit stages to reveal correlations, nonlinearity patterns, and redundancy.

Oversampling Compatibility#

The MATLAB oversampling analysis helpers are available from adctoolbox.oversampling:

ifilter(sigin, passband)

Ideal FFT-based brickwall filter. Retains only the specified normalized frequency bands from an input signal. Operates column-wise on matrices and follows the MATLAB orientation rule for wide inputs.

perfosr(sig, ...)

Sweep ADC performance (SNDR, SFDR, ENOB) versus oversampling ratio (OSR). Returns MATLAB-style outputs osr, sndr, sfdr, enob while sharing the same calculation engine as sweep_performance_vs_osr.

ntfperf(ntf, fl, fh, disp=False)

MATLAB-style wrapper around ntf_analyzer for theoretical NTF in-band noise suppression.

Python-Only Functions#

The following functions exist only in the Python package and have no MATLAB counterpart.

Analog Error Analysis#

  • analyze_error_pdf — Error probability density function via KDE.

  • analyze_error_spectrum — Error spectrum computed from fitting residual.

  • analyze_error_autocorr — Autocorrelation function (ACF) of error signal.

  • analyze_error_envelope_spectrum — Envelope spectrum via Hilbert transform to reveal AM modulation patterns.

  • fit_static_nonlin — Extract static nonlinearity coefficients (k2, k3) from a distorted sinewave.

Digital Output Analysis#

  • analyze_bit_activity — Percentage of 1's per bit (DC offset / clipping detection).

  • analyze_enob_sweep — ENOB vs. number of calibration bits.

  • plot_residual_scatter — Partial-sum residual visualization.

ADC Behavioral Models#

  • sar_convert / sar_reconstruct — SAR forward conversion and analog reconstruction.

  • sar_ideal_weights — Ideal binary or redundant SAR weights.

  • sar_apply_cap_mismatch — Unit-cap/Pelgrom-style CDAC mismatch model.

  • sar_apply_mismatch — Legacy per-weight gaussian perturbation helper.

Time-Interleaved ADC Analysis#

  • deinterleave / interleave — Split and recombine interleaved samples.

  • extract_mismatch_sine — Estimate channel gain, offset, and timing skew from a sine capture.

  • predict_spurs — Predict mismatch spur locations and amplitudes.

  • fractional_delay_fft / fractional_delay_farrow — Fractional-delay correction primitives.

  • calibrate_foreground — Foreground offset/gain/skew correction.

Unit Conversions & Metrics#

These are small utilities that MATLAB users typically write inline:

  • db_to_mag, mag_to_db, db_to_power, power_to_db

  • snr_to_enob, enob_to_snr, snr_to_nsd, nsd_to_snr

  • lsb_to_volts, volts_to_lsb

  • dbm_to_vrms, vrms_to_dbm, dbm_to_mw, mw_to_dbm

  • sine_amplitude_to_power, amplitudes_to_snr

  • bin_to_freq, freq_to_bin

  • calculate_walden_fom, calculate_schreier_fom

  • calculate_thermal_noise_limit, calculate_jitter_limit

Signal Generation#

The Python siggen module provides a Nonidealities class for building realistic ADC test signals with chainable impairments. This has no MATLAB equivalent:

  • Thermal noise, clock jitter, quantization noise

  • Static nonlinearity (polynomial or harmonic-dB specification)

  • Memory effect, incomplete sampling / settling

  • Residue amplifier gain error (static and dynamic)

  • Reference error (settling + droop)

  • AM noise, AM tone, clipping, drift, glitch injection

  • First-order noise shaping

Summary#

Category

MATLAB

Python

Core analysis functions

18

43

Matched across both

14

14

MATLAB-only

4

—

Python-only

—

41

Signal generation module

No

Yes (siggen)

Unit conversion utilities

No (inline)

Yes (20 functions)

SAR behavioral models

No

Yes

Time-interleaved ADC tools

No

Yes