Matlab Model Serdes Systems

MathWorksOfficialCreated by MathWorks

Skill
Tool integration
Science and research
BSD-2-Clause
No issues found

From the author

Model, simulate, and optimize Serializer/Deserializer (SerDes) systems — serial and parallel links — using MATLAB SerDes Toolbox. Design NRZ and PAM-N links (PAM3 through PAM16) — explore equalization architectures (FFE, CTLE, DFE), sweep or optimize parameters with genetic algorithms, and characterize channels from loss models, S-parameter files, or crosstalk scenarios. Process captured waveforms through equalization chains, build eye diagrams, and decompose jitter. Deliver IBIS-AMI models for Tx, Rx, Redriver, or Retimer by exporting to Simulink and compiling .ami/.ibs/.dll/.so files. Covers the full arc from initial design exploration and parameter optimization to compliance testing and compiled model validation, including custom datapath blocks for nonstandard equalization.

— the description by MathWorks

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

This skill provides a comprehensive framework for designing and validating Serializer/Deserializer (SerDes) systems. It covers the full development lifecycle, from initial architecture exploration and channel characterization to parameter optimization and the generation of industry-standard IBIS-AMI models. Users can simulate NRZ and PAM-N links, evaluate equalization architectures like FFE, CTLE, and DFE, and perform compliance testing using statistical and time-domain analysis.

What it does

Models and simulates high-speed serial and parallel links including NRZ and PAM-N signaling.
Characterizes channels using loss models, S-parameter files, and crosstalk scenarios.
Optimizes equalization architectures such as FFE, CTLE, and DFE using genetic algorithms.
Performs statistical and time-domain analysis including eye diagrams and jitter decomposition.
Exports designs to Simulink for IBIS-AMI model generation and compilation.
Validates compiled models against behavioral baselines for compliance and performance.

When to use it

When designing SerDes links for specific data rates and channel loss profiles.
When evaluating and tuning equalization architectures to meet signal integrity targets.
When you need to generate IBIS-AMI models for Tx, Rx, Redriver, or Retimer components.
When performing compliance checks using eye masks and jitter budgets.
When analyzing captured waveforms from oscilloscopes or simulations.

Advantages

Full-arc workflow from design exploration to compiled model delivery.
Supports advanced PAM-N modulation schemes up to PAM16.
Enables programmatic parameter sweeps and optimization for design space exploration.
Provides robust validation tools to ensure compiled models match behavioral references.
Integrates seamlessly with Simulink for complex time-domain system simulations.

Usage examples

“Create a SerdesSystem object for a PAM4 link and run a statistical analysis.”: The agent initializes a SerdesSystem with specified Tx/Rx blocks, configures the channel, and executes the analysis() method to generate eye diagrams and performance metrics.
“Export my current SerDes design to a Simulink model for AMI generation.”: The agent uses exportToSimulink(sys) to generate the Simulink representation and prepares the IbisAmiManager settings for model compilation.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/matlab/matlab-model-serdes-systems/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip matlab-model-serdes-systems.zip -d .agents/skills

Documentation by MathWorks

Modeling and Simulating SerDes Systems

Design, analyze, and deliver high-speed serial link models using SerDes Toolbox. From architecture exploration through IBIS-AMI model generation, covers the full workflow for NRZ and PAM-N links (PAM3 through PAM16).

When to Use

System design and architecture exploration

Designing SerDes links for a target data rate, signaling scheme (NRZ, PAM4, PAM-N), and channel loss
Evaluating equalization architectures (FFE, CTLE, DFE) and optimizing tap settings
Sweeping design parameters to find optimal configurations
Using industry reference designs (PCIe, USB4, DDR5, CEI, UCIe) as starting points
Characterizing Tx/Rx analog effects (parasitic capacitance, rise time, termination impedance)
Building custom datapath blocks for nonstandard equalization

Channel modeling and characterization

Loading S-parameter Touchstone files into SerdesSystem
Modeling channels with loss profiles, crosstalk (FEXT/NEXT), and aggressors
Fitting CTLE transfer functions from measured data via ctlefit

IBIS-AMI model generation

Building IBIS-AMI models for Tx, Rx, Redriver, or Retimer configurations
Exporting to Simulink, configuring AMI parameters, and compiling .ami/.ibs/.dll/.so
Scripting Simulink simulations and parameter sweeps with sim/parsim

Analysis and validation

Running statistical and time-domain simulations
Processing eye diagrams (eye height, eye width, COM, VEC, bathtub curves)
Decomposing jitter (TJ, RJ, DJ, DDJ, DCD, ISI)
Validating compiled AMI models against behavioral baselines
Running compliance checks with eye masks and jitter budgets

When NOT to Use

RF/microwave circuit design, antenna modeling, or baseband DSP filter design
General Simulink model scripting unrelated to SerDes

Must-Follow Rules

System Setup

SymbolTime / SampleInterval must yield an integer SamplesPerSymbol — fractional ratios cause silent errors
TxModel/RxModel are Transmitter/Receiver objects — not cell arrays. Construct with Transmitter('Blocks', {block1, block2}). Transmitter requires single-quoted property names — Transmitter("Blocks", ...) throws an ismember error. Receiver and all other classes accept double quotes
Signal conversion functions require column vectors — impulse2pulse, pulse2stateye, etc. error on row vectors
Include AnalogModel and JitterAndNoise for realistic results — bare Transmitter/Receiver without analog models produce optimistic COM (1-2 dB higher). See reference/equalization-tuning.md for parameter guidance (rise time, parasitic C, termination R)

Equalization

Set WaveType explicitly when using datapath blocks directly in MATLAB — Simulink sets this automatically, but MATLAB defaults to "Sample"
Adapted DFE/CTLE parameters are in results.outparams — NOT on the block object. After analysis(), sys.RxModel.Blocks{k}.TapWeights still holds initial values. In system objects chains, adapted taps are the second output: [y, taps] = dfecdr(x)
DFECDR Mode=0 is passthrough in Sample mode — DFE only applies with Mode≥1. Pre-load adapted taps from outparams with Mode=1 for instant convergence, or use Mode=2 with 10x EqualizationGain (9.6e-04) for self-converging chains
Set Modulation on DFECDR for PAM-N in system objects chains — Simulink inherits it from the model workspace, but MATLAB defaults to 2 (NRZ). Without this, PAM4 DFE adaptation fails silently

Metrics and Waveforms

Metrics.summary.EW is in picoseconds (already scaled) — do NOT multiply by 1e12. EH is in volts. PAM-N returns N-1 values per metric (e.g., PAM4 → 3 eyes, PAM8 → 7 eyes)
Channel impulse from analysis() is in V/s — when using filter() for time-domain convolution, multiply by dt: filter(impulse * dt, 1, wave). Without scaling, amplitudes blow up by ~10^11
pulse2wave operates on the stimulus provided — the output modulation depends on the input pattern (NRZ or PAM-N)

AMI and Simulink

Init-Only models cannot adapt — if DFE taps or CDR converge at runtime, you need a Dual model (both Init and GetWave)
AMI validation requires Signal Integrity Toolbox — serdes.AMI runner and the AMI Simulink block need both SerDes Toolbox and Signal Integrity Toolbox
AMI GetWave: call in a chunked loop — serdes.AMI passes BlockSize (default 1024) to AMI_GetWave, so only BlockSize samples are processed per call. You must call the object in a for loop with BlockSize-length chunks. State is preserved between calls via the DLL memory handle
AMI GetWave: set SkipFirstBlock = false when calling from MATLAB — the default (true) is for Simulink's internal signal buffering and causes the first block to pass through unprocessed
AMI Init: RowSize must match impulse length — serdes.AMI crashes MATLAB (process termination, no error) if RowSize doesn't match numel(impulse)
AMI generation requires Simulink model — use IbisAmiManager GUI or serdes.AMIExport with export() programmatically (see Programmatic AMI Generation)

Workflow

Design Exploration

Most projects start here. The goal is to find the right equalization architecture and settings for your channel.

Design — Create a SerdesSystem with Tx/Rx blocks and channel (loss model or S-parameters)
Analyze — Run analysis for statistical results, plotStatEye for eye diagrams, analysisReport for metrics
Sweep — Vary channel loss, FFE taps, CTLE gain, DFE taps, or jitter to map the design space
Compare — Evaluate architectures (FFE-only vs FFE+CTLE vs FFE+CTLE+DFE) using COM, eye height, eye width
Select — Choose the configuration that meets margin targets, then freeze equalization settings

Use SerdesSystem for programmatic exploration; serdesDesigner for interactive GUI work.

Waveform Processing

When you have a captured or imported waveform (e.g., from an oscilloscope or simulation) and want to equalize and analyze it directly:

Load — Import the waveform and define timing (SampleInterval, SymbolTime)
Equalize — Stream through datapath blocks (FFE, CTLE, DFECDR) with WaveType = "Sample"
Analyze — Build an eye diagram with eyeDiagramSI, extract metrics (eye height, COM, VEC)
Decompose jitter — Run jitter() on the equalized waveform for TJ, RJ, DJ, DDJ, ISI breakdown

DFECDR and DFE require a sample-by-sample for loop in Sample mode; FFE and CTLE accept full vectors. See reference/waveform-processing.md for the Direct Equalization pattern.

IBIS-AMI Model Delivery

When you need compiled models (.ami/.ibs/.dll/.so) for EDA tools or IP delivery:

Export — Call exportToSimulink(sys) to generate a Simulink model from the frozen design
Configure — Set AMI parameters, IBIS component/pin data, and model type via IbisAmiManager or serdes.AMIExport
Generate — Build .ami/.ibs and compile .dll/.so via serdes.AMIExport with export()
Validate — Load compiled DLL/SO with serdes.AMI, compare against behavioral reference (Init for impulse, GetWave for waveform)
Cross-check — The Simulink path (sim with Rx WaveOut) is the preferred time-domain reference. Compare against: statistical analysis(), system objects direct chain, and compiled AMI DLLs. See reference/simulink-serdes-simulation.md
Iterate — Fix discrepancies, re-export, re-validate until all paths agree

AMI Model Types

Choose the model type based on which equalization blocks need to adapt at runtime:

Type	Init_Returns_Impulse	GetWave_Exists	Use For
Init-Only	true	false	LTI equalization (fixed FFE, CTLE). Supports statistical analysis
GetWave-Only	false	true	Time-domain only. No statistical analysis
Dual	true	true	Adaptive equalization (DFE, CDR). All analysis types

Every IBIS-AMI model implements AMI_Init (required), AMI_GetWave (optional), and AMI_Close (required) per the IBIS standard.

Key Classes

Class	Purpose
`SerdesSystem`	Top-level system. Methods: `analysis`, `plotStatEye`, `plotImpulse`, `plotPulse`, `plotAlignedPulse`, `plotWavePattern`, `analysisReport`, `exportToSimulink`
`Transmitter`	Tx container. Construct: `Transmitter('Blocks', {serdes.FFE(...)})`
`Receiver`	Rx container. Construct: `Receiver('Blocks', {serdes.CTLE(...), serdes.DFECDR(...)})`
`ChannelData`	Channel spec. Props: `ChannelLossdB`, `ChannelLossFreq` (default 5 GHz — must override to Nyquist), `ChannelDifferentialImpedance`, or `Impulse`/`dt`
`JitterAndNoise`	IBIS 7.0 jitter/noise. 4 groups: Tx jitter (Rj/Dj/DCD/Sj), Rx jitter, Rx clock recovery (5 params, active with `RxClockMode='clocked'`, `'convolved'`, or `'normal'`), Rx noise. Values in seconds (default) or UI. See `reference/serdes-api-reference.md`
`serdes.AMI`	Run compiled AMI DLLs/SOs. Call: `[waveOut, impulseOut] = ami(waveIn, impulseIn, clockIn)`
`serdes.AMIExport`	Programmatic AMI export (R2026a+). Methods: `export`, `getExportSettings`. Props: `ModelTypeTx`, `DLLFiles`, `LinuxCrossCompile`
`SParameterChannel`	S-parameter to impulse response. Handles `.s4p` through `.s16p` (multi-port returns Nx(K) matrix: col 1=thru, cols 2+=aggressors). Props: `FileName`, `SampleInterval`, `StopTime`, `PortOrder`
`eyeDiagramSI`	Waveform eye diagram (R2024a+). Step: `eyeObj(wave)` — no output. Metrics: `eyeHeight`, `eyeWidth`, `com`, `vec`, `margin`
`ctlefit`	CTLE pole/zero fitter. Import: `ctlefit.readcsv`. Output: `GPZ` matrix for `serdes.CTLE("Specification", "GPZ Matrix")`

Datapath Blocks

Block	Role	Mode Values	Key Properties
`serdes.FFE`	Feed-forward equalizer	0, 1	`TapWeights`, `TapSpacing`, `Normalize`
`serdes.CTLE`	Continuous-time linear EQ	0, 1, 2	`Specification`, `DCGain`, `ACGain`, `PeakingGain`, `GPZ`
`serdes.DFECDR`	DFE + clock recovery	0, 1, 2	`TapWeights`, `CDRMode`, `PhaseDetector`, `Count`
`serdes.DFE`	Standalone DFE	0, 1, 2	`TapWeights`, `EqualizationGain`, `EqualizationStep`
`serdes.CDR`	Standalone CDR	0, 1	`CDRMode`, `Count`, `Step`, `Sensitivity`. Mode is deprecated
`serdes.AGC`	Auto gain control	0, 1	`TargetRMSVoltage`, `MaxGain`, `AveragingLength`
`serdes.VGA`	Variable gain amplifier	0, 1	`Gain`
`serdes.SaturatingAmplifier`	Limiting amplifier	0, 1	`Limit`, `LinearGain`, `Specification`
`serdes.PassThrough`	No-op placeholder	—	—

Mode values: 0 = Fixed (not exported), 1 = Fixed (exported as AMI parameter), 2 = Adaptive (GetWave). Only CTLE, DFECDR, and DFE support Mode=2.

Set CTLE Specification before setting gain properties — using ACGain with the default spec triggers a warning. GPZ Matrix requires Mode=1 for ConfigSelect to switch rows. See reference/equalization-tuning.md for the 4 specification options.

Patterns

Each reference file includes executable code patterns for its topic. Load the relevant reference before writing code — it contains correct calling conventions, constructor arguments, and property names. See the References table in Conventions below for the full routing map.

Common Mistakes

See reference/common-mistakes.md for 37 documented anti-patterns with wrong/correct code pairs covering API misuse, wrong property names, silent failures, and performance pitfalls.

Conventions

Design

Use SerdesSystem for programmatic design; serdesDesigner for interactive exploration. Always run analysis/plotStatEye before exporting
CRITICAL: ChannelLossFreq defaults to 5 GHz — always override to Nyquist (1/(2*SymbolTime)). Without this, 15 dB at 5 GHz becomes 40+ dB at Nyquist and the eye appears closed. Use SParameterChannel for real channels
analysis() does not auto-optimize FFE taps — use canonical de-emphasis with sum(abs(taps)) ≈ 1.0
DFE taps initialized to zero are optimized by the statistical solver automatically
Industry reference designs: serdesDesigner or openExample('serdes/PCIe5TransmitterReceiverIBISAMIModelExample'). See reference/equalization-tuning.md for starting-point configurations

Visualization and Metrics

All SerdesSystem plot methods (plotStatEye, plotImpulse, plotPulse, plotAlignedPulse, plotWavePattern) render into the current axes — use tiledlayout/nexttile or subplot for multi-panel layouts
eyeDiagramSI (R2024a+) for waveform eye diagrams; jitter (R2024b+) for jitter decomposition
DFECDR/DFE in "Sample" WaveType require scalar (sample-by-sample) input; FFE/CTLE accept full vectors
When generating PRBS waveforms via pulse2wave, ask the user for PRBS order and symbol count — default PRBS-10 (1023 symbols) if unspecified
Use serdes.utilities.SignalIntegrityColorMap (not hot/parula) for pulse2stateye plots. For compliance, use margin(eyeObj, eyeMask)
See reference/visualization-and-metrics.md for all plot methods, metrics fields/units, eyeDiagramSI, and jitter

AMI Export and Validation

Use serdes.AMIExport with export() for programmatic AMI generation — not slbuild (legacy, R2025b and earlier only)
Validate compiled DLLs/SOs with serdes.AMI before sending to EDA tools — both Init (impulse correlation) and GetWave (waveform chain)
Simulink is the preferred time-domain path — sim() with Rx WaveOut is the ground truth for adaptive systems. Compare against system objects chain and compiled AMI DLLs as cross-checks. Use the same PRBS stimulus across all time-domain paths
AMI export parameters on Simulink blocks use *AMI suffix ('ModeAMI', 'TapWeightsAMI'); model-level settings go through IbisAmiManager or serdes.AMIExport
No compiled DLLs/SOs ship with the toolbox. Cross-compile for Linux .so via serdes.utilities.createCrossCompiler

Script-First Workflow

For design, analysis, or sweep tasks — write code to .m files on disk, not inline MCP snippets. For quick one-off checks, inline evaluate_matlab_code is fine.

Create a .m file (e.g., serdes_pam4_design.m), run via run_matlab_file, iterate by editing and re-running
Deliver a polished script — %% Parameters block at top, %% Section headers, summary fprintf at end

References

Load when...	Reference
Building or modifying a `SerdesSystem`	`reference/serdes-api-reference.md`
Debugging wrong API names or silent failures	`reference/common-mistakes.md`
Working with S-parameter files or crosstalk	`reference/channel-modeling.md`
Tuning CTLE, FFE, DFE, or fitting transfer functions	`reference/equalization-tuning.md`
Sweeping parameters or running GA optimization	`reference/optimization.md`
Plotting eyes, reading metrics, or decomposing jitter	`reference/visualization-and-metrics.md`
Equalizing captured or imported waveforms directly	`reference/waveform-processing.md`
Exporting to Simulink, compiling AMI DLLs, or validating AMI models	`reference/programmatic-ami-generation.md`
Scripting Simulink simulations or `parsim` sweeps	`reference/simulink-serdes-simulation.md`
Building custom datapath blocks	`reference/custom-datapath-blocks.md`
Looking up utility functions (SNR, ICN, resample)	`reference/serdes-utilities.md`

Copyright 2026 The MathWorks, Inc.

Files

ATTRIBUTION.mdAdded by AutoGineer1.0 KiB
LICENSELicence file added by AutoGineer1.4 KiB
SKILL.md16.4 KiB
manifest.yaml1000 B
reference/channel-modeling.md7.8 KiB
reference/common-mistakes.md28.5 KiB
reference/custom-datapath-blocks.md12.8 KiB
reference/equalization-tuning.md11.4 KiB
reference/optimization.md9.0 KiB
reference/programmatic-ami-generation.md25.4 KiB
reference/serdes-api-reference.md28.4 KiB
reference/serdes-utilities.md10.4 KiB
reference/simulink-serdes-simulation.md19.4 KiB
reference/visualization-and-metrics.md20.0 KiB
reference/waveform-processing.md7.8 KiB

reference/serdes-api-reference.md

SerDes Toolbox Class and Function Reference

SerdesSystem

Top-level design and analysis object.

Construction

sys = SerdesSystem(...
    'TxModel', Transmitter('Blocks', {block1, block2}), ...
    'RxModel', Receiver('Blocks', {block3, block4}), ...
    'SymbolTime', 100e-12, ...
    'SamplesPerSymbol', 16, ...
    'Modulation', 2, ...
    'Signaling', "Differential", ...
    'BERtarget', 1e-6);

Properties

Property	Type	Description
`TxModel`	`Transmitter`	Tx container with `Blocks`, `RiseTime`, `VoltageSwingIdeal`, `AnalogModel`
`RxModel`	`Receiver`	Rx container with `Blocks`, `AnalogModel`
`Name`	string	System name
`SymbolTime`	double	Duration of one symbol (s)
`SamplesPerSymbol`	integer	Samples per symbol (default 16; valid: 8, 16, 32, 64, 128). `dt = SymbolTime / SamplesPerSymbol`
`Modulation`	integer	Modulation levels: 2 (NRZ), 3 (PAM3), 4 (PAM4), up to 16
`Signaling`	string	`"Differential"` or `"Single Ended"`
`BERtarget`	double	Target BER for statistical analysis (default `1e-6`)
`ChannelData`	`ChannelData`	Channel specification (see below)
`JitterAndNoise`	`JitterAndNoise`	Jitter/noise parameters (see below)
`dt`	double	Read-only. Computed sample interval
`ImpulseResponse`	double	Read-only. Channel impulse response after analysis
`Eye`	struct	Read-only. Statistical eye results
`Wave`	struct	Read-only. Equalized/unequalized waveforms
`Metrics`	struct	Read-only. BER eye contours and bathtub data

Choosing BERtarget

BERtarget sets the BER contour at which eye height and eye width are measured. It does not affect COM (which uses its own internal BER assumption). Common values:

Standard	BERtarget	Notes
IEEE 802.3 (Ethernet)	`1e-6`	Default; used for 100G/200G/400G/800G specs
OIF CEI	`1e-15`	More conservative; tighter eye opening
PCIe Gen5/6	`1e-6`	Post-FEC target
USB4	`1e-4`	Relaxed pre-FEC target

Lower BERtarget → smaller reported EH/EW (the eye is measured further into the tails of the distribution). When comparing results across designs, keep BERtarget consistent.

Metrics.summary Fields

After analysis(sys), sys.Metrics.summary contains:

Field	Type	Units	Description
`COMestimate`	scalar	dB	Channel Operating Margin
`EH`	vector	Volts	Eye height at BERtarget. PAM-N returns N-1 values (e.g., PAM4: 3, PAM8: 7)
`bestEH`	vector	Volts	Best-case eye height (optimistic sampling)
`EW`	vector	Picoseconds (already scaled — do NOT multiply by 1e12)	Eye width at BERtarget
`VEC`	scalar	Percent	Vertical eye closure
`eyeAreas`	vector	—	Normalized eye area per eye opening
`eyeLinearity`	scalar	Ratio	PAM-N level linearity (1.0 = perfectly symmetric)

Units warning: EH is in volts (multiply by 1e3 for mV), but EW is already in picoseconds. This is inconsistent with every other timing value in the toolbox (which use seconds).

analysis() outparams — Adapted Block Parameters

results.outparams is a cell array with one entry per datapath block (Tx blocks first, then Rx blocks in order). Each cell is either a double (e.g., FFE output) or a struct keyed by block class name.

Adapted DFE/CTLE parameters live here, NOT on the block object. After analysis(), sys.RxModel.Blocks{k}.TapWeights still holds the initial values. The solver's adapted values are only in outparams.

% With Tx FFE + Rx CTLE + Rx DFECDR:
% results.outparams{1} — FFE (double)
% results.outparams{2} — CTLE (struct with field 'CTLE')
% results.outparams{3} — DFECDR (struct with field 'DFECDR')

% Extract adapted DFE taps
for p = 1:numel(results.outparams)
    if isstruct(results.outparams{p}) && isfield(results.outparams{p}, 'DFECDR')
        adaptedTaps = results.outparams{p}.DFECDR.TapWeights;
        adaptedPhase = results.outparams{p}.DFECDR.Phase;
    end
end

Methods

Method	Returns	Description
`analysis(sys)`	struct with `t1`, `impulse`, `pulse`, `t2`, `wave`, `outparams`	Run statistical analysis
`plotStatEye(sys)`	—	Plot statistical eye diagram
`plotImpulse(sys)`	—	Plot equalized/unequalized impulse
`plotPulse(sys)`	—	Plot equalized/unequalized pulse
`plotAlignedPulse(sys)`	—	Plot aligned pulse response
`plotWavePattern(sys)`	—	Plot PRBS waveform
`analysisReport(sys)`	—	Print analysis summary to command window
`exportToSimulink(sys)`	—	Export to Simulink model
`validateSerdesSystem(sys)`	—	Validate system configuration

ChannelData

Three channel definition modes (mutually exclusive):

Lossy model (default): ChannelLossdB + ChannelLossFreq + ChannelDifferentialImpedance
S-parameter model: via serdesDesigner export
Direct impulse: Impulse + dt

IMPORTANT: To select impulse mode, pass "Impulse" and "dt" as constructor name-value pairs. Assigning sys.ChannelData.Impulse = ... after construction does NOT switch the mode — the loss model silently takes priority.

% CORRECT -- impulse mode via constructor
channel = ChannelData("Impulse", impulseVector, "dt", sampleInterval);
sys = SerdesSystem("ChannelData", channel, ...);

% WRONG -- loss model runs, impulse is ignored
sys.ChannelData.Impulse = impulseVector;
sys.ChannelData.dt = sampleInterval;

Property	Description
`ChannelLossdB`	Channel loss in dB at `ChannelLossFreq`
`ChannelLossFreq`	Frequency for loss specification (Hz). Default: 5 GHz — almost always wrong. Set to Nyquist: `1/(2*SymbolTime)`
`ChannelDifferentialImpedance`	Differential impedance (ohms)
`Impulse`	Direct impulse response (column vector)
`dt`	Sample interval for direct impulse (s)
`EnableCrosstalk`	Enable crosstalk modeling (ICN mode only — multi-column impulse auto-enables)
`CrosstalkSpecification`	`"CEI-28G-SR"` (default), `"CEI-25G-LR"`, `"CEI-28G-VSR"`, `"100GBASE-CR4"`, or `"Custom"`
`FEXTICN`	FEXT integrated crosstalk noise (V RMS, default 0.015). Only with `"Custom"` spec
`NEXTICN`	NEXT integrated crosstalk noise (V RMS, default 0.010). Only with `"Custom"` spec
`fb`	Baud rate for ICN noise shaping (Hz, auto-derived from SymbolTime)
`Aft` / `Ant`	Aggressor amplitude for FEXT/NEXT (Vpp, default 1.2)
`Tft` / `Tnt`	Aggressor rise time for FEXT/NEXT (s, default 9.6e-12)
`OptionSel`	Read-only. Internal mode selector (1=impulse, 3=loss model)

JitterAndNoise

IBIS 7.0 compliant jitter and noise parameters. Each jitter property is a SimpleJitter object with Value, Type, and Include fields. Assign a scalar to set the value directly; access the sub-fields for unit control.

RxClockMode

Controls how the Rx sampling clock is modeled:

Mode	Description
`"ideal"`	Perfect clock — no Rx clock jitter. Rx_Clock_Recovery_* parameters ignored
`"clocked"`	Clock with jitter — Rx_Clock_Recovery_* parameters affect the sampling eye
`"convolved"`	Convolves clock jitter with data jitter (statistical combination)
`"normal"`	Normal clock recovery model

Key rule: Set RxClockMode to "clocked" for Rx clock recovery jitter to affect analysis. With "ideal", all Rx_Clock_Recovery_* parameters are ignored.

Tx Jitter

Property	Default	Convention
`Tx_Rj`	0	Random jitter — standard deviation (not peak-to-peak)
`Tx_Dj`	0	Deterministic jitter — peak-to-peak
`Tx_DCD`	0	Duty cycle distortion — peak-to-peak
`Tx_Sj`	0	Sinusoidal jitter — peak-to-peak amplitude
`Tx_Sj_Frequency`	0	Sinusoidal jitter frequency (Hz, not UI)

Rx Jitter

Requires RxClockMode = "clocked" — Rx jitter is modeled as clock jitter and has NO effect in "ideal" mode. Rx jitter and Rx Clock Recovery jitter of the same type produce identical degradation.

Property	Default	Convention
`Rx_Rj`	0	Random jitter — standard deviation
`Rx_Dj`	0	Deterministic jitter — peak-to-peak
`Rx_DCD`	0	Duty cycle distortion — peak-to-peak
`Rx_Sj`	0	Sinusoidal jitter — peak-to-peak amplitude

Rx Clock Recovery Jitter

Only active when RxClockMode is "clocked", "convolved", or "normal". Ignored when RxClockMode = "ideal".

Property	Default	Convention
`Rx_Clock_Recovery_Mean`	0	CDR mean phase offset. No measurable effect on COM in statistical analysis
`Rx_Clock_Recovery_Rj`	0	CDR random jitter — standard deviation
`Rx_Clock_Recovery_Dj`	0	CDR deterministic jitter — peak-to-peak
`Rx_Clock_Recovery_Sj`	0	CDR sinusoidal jitter — peak-to-peak
`Rx_Clock_Recovery_DCD`	0	CDR duty cycle distortion — peak-to-peak

Rx Noise

Property	Default	Convention
`Rx_Receiver_Sensitivity`	0	Minimum detectable voltage (V). No measurable effect on COM in statistical analysis
`Rx_GaussianNoise`	0	Additive Gaussian noise — RMS (V)
`Rx_UniformNoise`	0	Additive uniform noise — peak (V)

Units: Seconds or UI

Each jitter property wraps a SimpleJitter object with Value, Type, and Include fields. The Type field accepts "Float" (seconds, default) or "UI" (unit intervals). Both work correctly — analysis() converts UI values to seconds using the system's SymbolTime.

Critical: Include defaults to false. Direct scalar assignment (jn.Tx_Rj = 1e-12) auto-sets Include = true and Type = "Float". When constructing SimpleJitter manually, always set Include = true explicitly.

% Direct assignment in seconds (simplest — auto-sets Include=true)
jn = JitterAndNoise('RxClockMode', 'clocked');
jn.Tx_Rj = 1e-12;     % 1 ps RMS (Type="Float", Include=true automatic)
jn.Tx_Dj = 5e-12;     % 5 ps peak-to-peak

% Using UI type via SimpleJitter constructor (Include=true required)
sj = SimpleJitter('Value', 0.05, 'Include', true, 'Type', 'UI');
jn.Tx_DCD = sj;       % 0.05 UI → analysis converts via SymbolTime

% Or convert from UI spec to seconds manually
uiSpec = 0.05;  % 5% UI from a datasheet
jn.Tx_DCD = uiSpec * symbolTime;  % explicit conversion

% Disable a specific jitter component
sj = jn.Rx_Sj;
sj.Include = false;
jn.Rx_Sj = sj;

Construction Example

% Typical jitter budget for a 56 Gbps PAM4 link
jn = JitterAndNoise('RxClockMode', 'clocked');
jn.Tx_Rj = 0.5e-12;                  % 0.5 ps RMS
jn.Tx_Dj = 2e-12;                    % 2 ps p-p
jn.Tx_DCD = 1e-12;                   % 1 ps DCD
jn.Rx_Rj = 0.3e-12;                  % 0.3 ps RMS
jn.Rx_Clock_Recovery_Rj = 0.2e-12;   % 0.2 ps CDR RJ
jn.Rx_Clock_Recovery_Dj = 1e-12;     % 1 ps CDR DJ
jn.Rx_GaussianNoise = 0.5e-3;        % 0.5 mV RMS

sys = SerdesSystem('JitterAndNoise', jn, ...);

serdes.AMI

Run and test compiled IBIS-AMI DLLs/SOs. Requires SerDes Toolbox + Signal Integrity Toolbox.

Calling Convention

[waveOut, impulseOut] = ami(waveIn, impulseIn, clockIn)

waveIn — Input waveform (column vector of BlockSize length for GetWave). For Tx: stimulus. For Rx: channel output.
impulseIn — Channel impulse response (column vector of RowSize length, optionally with aggressors).
clockIn — Clock times vector of BlockSize length ending with -1, or just -1 (scalar) for Init-only mode.
waveOut — Output waveform.
impulseOut — Output impulse response (modified by Init on first call, zeros on subsequent GetWave calls).

GetWave chunked loop: AMI_GetWave processes exactly BlockSize samples per call. When calling from MATLAB with a waveform longer than BlockSize, you must call the object in a for loop with BlockSize-length chunks. The DLL state is preserved between calls via the memory handle — Init fires only on the first call:

nPad = ceil(N / blockSize) * blockSize;
wavePad = [wave; zeros(nPad - N, 1)];
waveOut = zeros(nPad, 1);
clockIn = -ones(blockSize, 1);
for k = 1:nPad/blockSize
    idx = (k-1)*blockSize + (1:blockSize);
    [waveOut(idx), ~] = ami(wavePad(idx), impulse, clockIn);
end
waveOut = waveOut(1:N);

Properties

Property	Default	Description
`LibraryName`	`''`	Name of the AMI DLL/SO (without extension)
`LibraryPath`	`''`	Path to the DLL/SO directory
`InputString`	`''`	AMI parameter input string
`SymbolTime`	`1e-10`	Symbol time (s)
`SampleInterval`	`6.25e-12`	Sample interval (s)
`BlockSize`	`1024`	Number of samples passed to each `AMI_GetWave` call
`RowSize`	`7065`	Impulse response length — must match `numel(impulseIn)` or MATLAB crashes
`Aggressors`	`0`	Number of aggressor channels
`PassThrough`	`false`	Pass-through mode (bypass DLL)
`InitOnly`	`true`	`true` = Init flow only, `false` = Init + GetWave
`SkipFirstBlock`	`true`	Skip first GetWave block. Set `false` when calling from MATLAB — the default `true` is for Simulink's internal signal buffering and causes the first block to pass through unprocessed
`OutputName`	`'AMIOut'`	Workspace variable name (Simulink only, ignored in MATLAB)
`AMIData`	struct	Output data structure from DLL

Static method: serdes.AMI.generateDefaultInputString(amiFilePath) — parse .ami file and return default input string.

serdes.AMIExport (R2026a+)

Programmatic AMI export object. Mirrors the IbisAmiManager GUI.

AMIExport = serdes.AMIExport("modelName")
AMIExport.ModelsToExport = "Both Tx and Rx";
AMIExport.DLLFiles = true;
AMIExport.LinuxCrossCompile = true;
export(AMIExport);

Property	Default	Description
`ModelName`	(read-only)	Simulink model name
`ModelConfiguration`	`"TxandRx"`	System model type
`ModelTypeTx`	`"Dual model"`	Tx export type: `"Dual model"`, `"GetWave only"`, `"Init only"`
`ModelTypeRx`	`"Dual model"`	Rx export type
`ModelsToExport`	`"Both Tx and Rx"`	Which models to export
`DLLFiles`	`true`	Export compiled DLL/SO
`AMIFiles`	`true`	Export .ami files
`IBISFile`	`true`	Export .ibs file
`LinuxCrossCompile`	`false`	Build Linux .so from Windows
`Obfuscate`	`false`	Obfuscate executables
`TargetDir`	string	Output directory

Methods: export(AMIExport), getExportSettings(AMIExport).

SParameterChannel

Convert S-parameter Touchstone files to time-domain impulse responses with analog Tx/Rx termination modeling. Uses rational fitting internally to produce a causal impulse response from frequency-domain S-parameter data.

Construction

spCh = SParameterChannel("FileName", "channel.s4p");

IMPORTANT: The constructor requires "FileName" as a name-value argument. Positional syntax (SParameterChannel("channel.s4p")) does not work.

Properties — Input

Property	Default	Description
`FileName`	`"default.s4p"`	Touchstone file path (`.s2p`, `.s4p`, `.s8p`, etc.)
`PortOrder`	`[]`	Port ordering `[in+ out+ in- out-]`. Empty = auto-detect. For files with convention `1=in+, 2=in-, 3=out+, 4=out-`, use `[1 3 2 4]`
`MaxNumberOfPoles`	`1000`	Maximum poles for rational model fit
`ErrorTolerance`	`-40`	Desired error tolerance (dB) for rational model fit

Properties — Timing

Property	Default	Description
`SampleInterval`	`6.25e-12`	Output sample interval (s). Must match `SerdesSystem` dt
`StopTime`	`1e-7`	Impulse response duration (s). Use >= 60 ns for SerdesSystem integration

Properties — Analog Model

Property	Default	Description
`TxAmplitude`	`1`	Stimulus input amplitude (V, single-ended)
`TxRiseTime`	`1e-11`	20%-80% rise time of stimulus (s)
`TxR`	`50`	Tx termination resistance (ohms, single-ended)
`TxC`	`1e-13`	Tx termination capacitance (F)
`RxR`	`50`	Rx termination resistance (ohms, single-ended)
`RxC`	`2e-13`	Rx termination capacitance (F)

Properties — Read-Only Outputs

Property	Description
`ImpulseResponse`	Channel impulse response (column vector). Continuous-time scaling: `sum(ImpulseResponse) * SampleInterval` gives DC gain
`t`	Time vector (column, s)
`TransferFunction`	System transfer function (column vector, complex)
`ChannelTransferFunction`	Channel-only transfer function (excludes Tx/Rx analog model)
`f`	S-parameter frequency vector (column, Hz)
`RationalResults`	`rational` object with fitting statistics
`PortOrderUsed`	Actual port order used after auto-detection
`RiseTimeUncertainty`	Uncertainty due to rise time discretization (s, negative)

Integration with SerdesSystem

To feed an S-parameter impulse into SerdesSystem, construct ChannelData with "Impulse" and "dt" name-value pairs in the constructor. Do NOT assign sys.ChannelData.Impulse after construction — the internal mode selector stays on loss-model and ignores the impulse.

% CORRECT -- use ChannelData constructor
spCh = SParameterChannel("FileName", "channel.s4p");
spCh.SampleInterval = symbolTime / samplesPerSymbol;
spCh.StopTime = 60e-9;
spCh.PortOrder = [1 3 2 4];

channel = ChannelData( ...
    "Impulse", spCh.ImpulseResponse, ...
    "dt", spCh.SampleInterval);

sys = SerdesSystem( ...
    "ChannelData", channel, ...
    "SymbolTime", symbolTime, ...
    "SamplesPerSymbol", samplesPerSymbol, ...
    ... );

% WRONG -- impulse is silently ignored, loss model runs instead
sys.ChannelData.Impulse = spCh.ImpulseResponse;
sys.ChannelData.dt = spCh.SampleInterval;

Key Gotchas

StopTime: Use >= 60 ns for SerdesSystem integration. Short durations (e.g., 20 ns) produce too few samples and analysis may give incorrect metrics.
SampleInterval: Must match SerdesSystem.dt (SymbolTime / SamplesPerSymbol).
PortOrder: For 4-port files with convention 1=in+, 2=in-, 3=out+, 4=out-, set [1 3 2 4]. Check PortOrderUsed after construction to verify.
Multi-port files: SParameterChannel handles .s8p/.s12p/.s16p directly. ImpulseResponse returns an Nx(K) matrix: column 1 = victim thru, columns 2+ = aggressor coupling. Pass the full matrix to ChannelData("Impulse", ir, "dt", dt) to include crosstalk.

See reference/channel-modeling.md for a complete S-parameter workflow.

Signal Conversion Functions

All functions require column vector inputs.

Function	Signature	Description
`impulse2pulse`	`P = impulse2pulse(I, N, dt)`	Impulse to pulse response
`impulse2step`	`S = impulse2step(I, dt)`	Impulse to step response
`pulse2impulse`	`I = pulse2impulse(P, N, dt)`	Pulse to impulse
`step2impulse`	`I = step2impulse(S, dt)`	Step to impulse
`pulse2wave`	`W = pulse2wave(P, D, N)`	Pulse + data to waveform
`wave2pulse`	`P = wave2pulse(W, D, N)`	Waveform to pulse
`prbs`	`[P, seed] = prbs(O, N, seed, reverse)`	PRBS generator. O = order (5-40), N = length. Returns column vector of 0/1. Optional seed for reproducibility, reverse for time-reversed LFSR
`pulse2stateye`	`[S, V, T] = pulse2stateye(P, N, M)`	Statistical eye. M = modulation levels
`pulse2pda`	`[E, T, D] = pulse2pda(P, N, M)`	Peak distortion analysis eye
`optPulseMetric`	`M = optPulseMetric(P, N, dt, BER)`	Returns struct with fields: `maxEyeHeight`, `maxMeanEyeHeight`, `maxCOM`, `eyeArea`, `eyeWidth`, `centerEyeHeight`, `centerMeanEyeHeight`, `centerCOM`, `SNR`, `usedBER`. LTI-only — does not account for jitter
`pulseRecoverClock`	`C = pulseRecoverClock(P, N)`	Clock recovery (hula-hoop algorithm)
`applyLinearFilter`	`[Y, Xh, Yh] = applyLinearFilter(X, Xh, Yh, B, A)`	Recursive filter (Direct Form I)

Waveform Analysis (R2024a+)

Class/Function	Signature	Description
`eyeDiagramSI`	`eyeObj = eyeDiagramSI; eyeObj(wave)`	System object — builds 2-D eye histogram from waveform. No output from step. Do NOT name the variable `eye` — it shadows MATLAB's built-in `eye()`. Metrics via `eyeHeight`, `eyeWidth`, `com`, `vec`, `eyeArea`, `eyeLinearity`, `bathtub`, `margin`
`jitter`	`J = jitter(y, SampleInterval=s, SymbolTime=t, Plot=false)`	Jitter decomposition. Returns struct: `TJrms`, `TJpkpk`, `RJrms`, `DJpkpk`, `DDJpkpk`, `DCDrms`, `ISIpkpk`, `SJa`/`SJf`/`SJp`

See reference/visualization-and-metrics.md for eyeDiagramSI/jitter property tables and reference/waveform-processing.md for direct equalization patterns.

Fitting Tools

ctlefit

Programmatic interface to the CTLE Fitter. Fits rational pole/zero models to CTLE frequency response data and produces a GPZ matrix for serdes.CTLE.

Construction

[f, H] = ctlefit.readcsv(csvFile);   % Import from CSV
obj = ctlefit('f', f, 'H', H, 'SampleInterval', dt, ...
    'MaxNumberOfPoles', 2, 'ErrorTolerance', -40, ...
    'TendsToZero', 1, 'UseCommonPoles', 0, 'PaddedPole', 1e11);

Import

Static Method	Description
`ctlefit.readcsv(filename)`	Read CTLE frequency response from CSV. Returns `[f, H]`. Header format: `freq,real,imag,real,imag,...` (also accepts `db,phaseDeg`, `db,phaseRad`, `mag,phaseDeg`, `mag,phaseRad`)

Constructor Properties

Property	Type	Description
`f`	double vector	Frequency vector (Hz)
`H`	complex matrix	Complex transfer functions (`nFreq x nSettings`)
`SampleInterval`	double	Sample interval (s). Typically `SymbolTime / SamplesPerSymbol`
`MaxNumberOfPoles`	integer	Maximum number of poles per setting (1, 2, or 3). Default: 2
`ErrorTolerance`	double	Maximum acceptable fit error (dB). Default: -40
`TendsToZero`	logical	`1` = response rolls off at high frequency (typical for CTLE). `0` = flat
`UseCommonPoles`	logical	`1` = force all settings to share pole locations. `0` = independent poles
`PaddedPole`	double	Extra high-frequency pole (Hz) for rolloff. Example: `1e11`

Output

Property	Format	Description
`GPZ`	matrix	Fitted poles/zeros/gains. Format: `[Gain(dB) Pole1(Hz) Zero1(Hz) Pole2(Hz); ...]`. One row per CTLE gain setting. Poles/zeros are negative (left half-plane). Feeds directly into `serdes.CTLE('Specification', "GPZ Matrix", 'GPZ', gpz)`

Preprocessing Methods

Method	Description
`truncateAbove(obj, fMax)`	Remove frequency data above `fMax` (Hz). Use `2 * fNyquist` to cut fitting noise
`truncateBelow(obj, fMin)`	Remove frequency data below `fMin` (Hz)
`resample(obj, fNew)`	Resample transfer functions onto new frequency vector
`removeDelay(obj)`	Remove group delay from transfer functions

Visualization Methods

Method	Description
`plot(obj, format, which, plotType)`	Magnitude response. `format`: `"dB"`, `"mag"`. `which`: `"All"` or index. `plotType`: `"semilogx"`, `"plot"`
`plotPulse(obj, which)`	Pulse response of fitted model
`plotError(obj, which)`	Fit error (deviation between fitted and original)
`plotFitMetric(obj, which)`	Fit quality metric plot
`plotPoleZero(obj, which)`	Pole-zero map for each gain setting

Typical Workflow

[f, H] = ctlefit.readcsv("ctle_data.csv");
obj = ctlefit("f", f, "H", H, "SampleInterval", dt, "MaxNumberOfPoles", 2);
truncateAbove(obj, 2 * fNyquist);
gpz = obj.GPZ;
ctle = serdes.CTLE("Specification", "GPZ Matrix", "GPZ", gpz);

Other Fitting Tools

Tool	Description
`ctlefitter`	CTLE Fitter GUI app (interactive version of `ctlefit`)
`SParameterChannel`	S-parameter fitting (also listed above)
`sParameterFitter`	S-parameter Fitter GUI app

GUI Apps

App	Launch	Purpose
`serdesDesigner`	`serdesDesigner`	Main design app — design, analyze, export
`IbisAmiManager`	From Configuration block	AMI parameter tree, IBIS files, DLL generation
`ctlefitter`	`ctlefitter`	CTLE pole/zero fitting
`sParameterFitter`	`sParameterFitter`	S-parameter to impulse fitting

serdes.utilities Namespace

Signal quality metrics, signal manipulation, and formatting functions. All called as serdes.utilities.<FunctionName>(...). See reference/serdes-utilities.md for full signatures, gotchas, and usage patterns.

Function	Purpose	Key Output
`SER`	Symbol error rate class	`.SymbolErrorRate` after `calculateSER()`
`SNR(P, N)`	Signal-to-noise ratio from pulse	Linear ratio (not dB)
`WaveEyeHeight(wave, N)`	Max eye height by waveform folding	`[maxEH, upper, lower]` in volts
`icn(f, Sxt, fb)`	Integrated crosstalk noise (OIF-CEI)	V RMS
`poleZeroDefine(NV)`	CTLE poles/zeros from peaking specs	`[G, P, Z]` or GPZ matrix
`SoftClipper(gain, limit)`	Cubic soft-clipper curve	`[Nx2]` Vin/Vout
`resampleImpulse(X, dtIn, dtOut, normalize)`	Resample impulse response	Column matrix
`shiftPulse(pulse, N)`	Center pulse for eye diagrams	Shifted pulse + cursor index
`floatingTaps(taps, fixed, groups, size)`	DFE floating tap placement	Constrained tap vector
`dt2Trf` / `trf2dt`	IBIS ↔ LTI rise time conversion	Rise time (s)
`db(x)`	`20*log10(abs(x))`	dB value
`num2prefix(X, unit)`	SI prefix formatting	`"100 ps"`, `"14.1 GHz"`
`SignalIntegrityColorMap`	Standard eye diagram colormap	Colormap matrix
`pulseRecoverClock(P, N)`	Clock recovery (hula-hoop)	Clock index

Simulink SerDes Blocks

System objects used as Simulink blocks in exported SerDes models. See reference/simulink-serdes-simulation.md for scripted simulation workflows.

Block	Purpose	Key Properties
`serdes.Stimulus`	PRBS pattern + jitter	`Specification` (10 options), `Modulation` (2-32), `Dj`/`Rj`/`DCD`/`Sj`, `BinaryToPAMnMapping`
`serdes.ChannelLoss`	Lossy transmission line (IEEE 802.3bj)	`Loss`, `TargetFrequency`, `Zc`, `TxR`/`RxR`/`TxC`/`RxC`, `EnableCrosstalk`, `FEXTICN`/`NEXTICN`
`serdes.ChannelImpulse`	Channel from impulse matrix	`ImpulseMatrix` (Nx1 or Nx2xM), `SampleInterval`
`serdes.EyeAnalyzer`	Waveform eye analysis (MATLAB-only)	`ClockMode`, `IgnoreSymbols`, jitter/noise impairments, `EyeHistogram` (read-only)

Pattern: Programmatic System Design and Analysis

% Create a 56 Gbps PAM4 SerDes system with full analog models
sys = SerdesSystem(...
    'TxModel', Transmitter('Blocks', ...
        {serdes.FFE('TapWeights', [-0.1 0.7 -0.2])}, ...
        'AnalogModel', AnalogModel('R', 50, 'C', 1e-13), ...
        'RiseTime', 1e-11, 'VoltageSwingIdeal', 1), ...
    'RxModel', Receiver('Blocks', ...
        {serdes.CTLE('Specification', "DC Gain and AC Gain", ...
                      'DCGain', 0, 'ACGain', 12, ...
                      'PeakingFrequency', 14e9), ...
         serdes.DFECDR('TapWeights', zeros(1, 5), 'Mode', 2)}, ...
        'AnalogModel', AnalogModel('R', 50, 'C', 2e-13)), ...
    'ChannelData', ChannelData('ChannelLossdB', 15, ...
        'ChannelLossFreq', 14e9, 'ChannelDifferentialImpedance', 100), ...
    'JitterAndNoise', JitterAndNoise('RxClockMode', 'clocked'), ...
    'SymbolTime', 35.71e-12, ...
    'SamplesPerSymbol', 16, ...
    'Modulation', 4, ...
    'Signaling', 'Differential', ...
    'BERtarget', 1e-6);

% Run analysis — returns struct with t1, impulse, pulse, t2, wave, outparams
results = analysis(sys);
plotStatEye(sys);
analysisReport(sys);

% Read metrics — EH is in volts, EW is in picoseconds (already scaled)
m = sys.Metrics.summary;
fprintf("COM: %.2f dB, EH: %.1f mV, EW: %.1f ps\n", m.COMestimate, m.EH(2)*1e3, m.EW(2));

% Read adapted DFE taps from outparams (NOT from the block object)
% outparams is a cell array — one entry per datapath block in Tx/Rx order
for p = 1:numel(results.outparams)
    if isstruct(results.outparams{p}) && isfield(results.outparams{p}, 'DFECDR')
        adaptedTaps = results.outparams{p}.DFECDR.TapWeights;
        fprintf("Adapted DFE taps: %s\n", mat2str(adaptedTaps, 4));
    end
end

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.