702 - MATLAB Integration¶

Control GeckoCIRCUITS remotely from MATLAB or Octave using the Remote Method Invocation (RMI) interface. This enables seamless integration of circuit simulation into MATLAB workflows, Simulink models, and automated analysis pipelines.

Overview¶

MATLAB integration provides:

RMI Interface - Full remote control from MATLAB command line or scripts
Parameter Control - Set and get component parameters programmatically
Simulation Execution - Run simulations and wait for completion from MATLAB
Data Extraction - Retrieve measurements and waveforms directly into MATLAB variables
Simulink Co-Simulation - Embed GeckoCIRCUITS in Simulink models via S-Function blocks

Prerequisites

MATLAB R2018b or later (with Java support)
GeckoCIRCUITS 1.0+ (gecko-1.0-jar-with-dependencies.jar)
Both tools must run on the same machine for RMI

RMI Setup¶

Step 1: Start GeckoCIRCUITS in Remote Mode¶

First, launch GeckoCIRCUITS with the remote flag:

java -Xmx3G -jar gecko-1.0-jar-with-dependencies.jar --remote

The application will start in RMI server mode, listening for MATLAB connections on the default RMI port (1099).

Alternatively, start it from MATLAB:

% Start GeckoCIRCUITS in background (Windows/Linux/macOS)
system('java -Xmx3G -jar gecko-1.0-jar-with-dependencies.jar --remote &');
pause(5);  % Wait for startup and RMI registration

Step 2: Connect from MATLAB¶

Once GeckoCIRCUITS is running, add the JAR to MATLAB's classpath and connect:

% Add GeckoCIRCUITS JAR to Java path
javaaddpath('path/to/gecko-1.0-jar-with-dependencies.jar');

% Import the remote interface class
import gecko.GeckoRemoteInterface;

% Get singleton instance (connects to running GeckoCIRCUITS)
gecko = GeckoRemoteInterface.getInstance();

% Verify connection
if gecko.isConnected()
    disp('Connected to GeckoCIRCUITS');
else
    error('Failed to connect to GeckoCIRCUITS');
end

Port Configuration

By default, GeckoCIRCUITS uses RMI port 1099. If this port is in use, start with:

java -jar gecko.jar --remote --rmi-port 2099

Then connect from MATLAB using the GeckoRemoteInterface constructor with the custom port.

Basic Operations¶

Opening Circuits¶

% Open circuit file
gecko.openFile('buck_converter.ipes');

% Save circuit
gecko.saveFile('modified_buck.ipes');

Setting Parameters¶

% Set component parameter
gecko.setParameter('R_load', 10);
gecko.setParameter('L1.inductance', 47e-6);
gecko.setParameter('PWM1.frequency', 100e3);
gecko.setParameter('PWM1.dutyCycle', 0.5);

Running Simulations¶

% Configure simulation
gecko.setSimulationTime(0.01);  % 10 ms
gecko.setTimeStep(1e-7);        % 100 ns

% Run simulation
gecko.startSimulation();

% Wait for completion
while gecko.isSimulationRunning()
    pause(0.1);
end

Getting Results¶

% Get scope measurements
Vout_mean = gecko.getMean('SCOPE', 'CH1');
Iout_rms = gecko.getRMS('SCOPE', 'CH2');
ripple_pp = gecko.getPeakToPeak('SCOPE', 'CH3');

% Get waveform data
time = gecko.getSignalTime('SCOPE');
voltage = gecko.getSignalData('SCOPE', 'CH1');
current = gecko.getSignalData('SCOPE', 'CH2');

% Plot in MATLAB
figure;
plot(time*1e3, voltage);
xlabel('Time (ms)');
ylabel('Voltage (V)');

Parameter Sweeps¶

Duty Cycle Sweep Example¶

duties = 0.2:0.05:0.8;
Vout = zeros(size(duties));
efficiency = zeros(size(duties));

for i = 1:length(duties)
    gecko.setParameter('PWM1.dutyCycle', duties(i));
    gecko.startSimulation();

    while gecko.isSimulationRunning()
        pause(0.1);
    end

    Vout(i) = gecko.getMean('SCOPE', 'Vout');
    Pin = gecko.getMean('SCOPE', 'Pin');
    Pout = gecko.getMean('SCOPE', 'Pout');
    efficiency(i) = Pout / Pin;
end

% Plot results
figure;
subplot(2,1,1);
plot(duties, Vout);
xlabel('Duty Cycle'); ylabel('Output Voltage (V)');

subplot(2,1,2);
plot(duties, efficiency*100);
xlabel('Duty Cycle'); ylabel('Efficiency (%)');

Frequency Sweep (Bode Plot)¶

frequencies = logspace(1, 5, 50);  % 10 Hz to 100 kHz
magnitude = zeros(size(frequencies));
phase = zeros(size(frequencies));

for i = 1:length(frequencies)
    gecko.setParameter('AC_SOURCE.frequency', frequencies(i));
    gecko.startSimulation();

    while gecko.isSimulationRunning()
        pause(0.1);
    end

    Vin = gecko.getFFTMagnitude('SCOPE', 'Vin', frequencies(i));
    Vout = gecko.getFFTMagnitude('SCOPE', 'Vout', frequencies(i));
    magnitude(i) = 20*log10(Vout/Vin);

    phase(i) = gecko.getFFTPhase('SCOPE', 'Vout', frequencies(i)) - ...
               gecko.getFFTPhase('SCOPE', 'Vin', frequencies(i));
end

% Bode plot
figure;
subplot(2,1,1);
semilogx(frequencies, magnitude);
xlabel('Frequency (Hz)'); ylabel('Magnitude (dB)');
grid on;

subplot(2,1,2);
semilogx(frequencies, phase);
xlabel('Frequency (Hz)'); ylabel('Phase (deg)');
grid on;

Simulink Integration¶

S-Function Block¶

GeckoCIRCUITS can run as Simulink S-Function:

Add GeckoSFunction block to model
Configure:
Circuit file path
Input signals mapping
Output signals mapping
Set sample time to match simulation step

Co-simulation Setup¶

% In MATLAB, before running Simulink model
gecko = GeckoRemoteInterface.getInstance();
gecko.openFile('converter.ipes');
gecko.setSimulinkMode(true);

Memory-Mapped Files (MMF)¶

For high-speed data exchange:

% Start with MMF mode
system('java -jar gecko.jar --mmf &');

% In MATLAB
mmf = memmapfile('gecko_data.bin', 'Format', 'double');
data = mmf.Data;

Error Handling¶

try
    gecko.openFile('circuit.ipes');
    gecko.startSimulation();
catch ME
    fprintf('Error: %s\n', ME.message);
    % Handle error
end

Best Practices¶

Practice	Why	Example
Check simulation completion	Avoid reading stale results	Use `while gecko.isSimulationRunning()`
Batch parameter updates	Minimize RMI overhead	Set all parameters before `startSimulation()`
Use appropriate timestep	Balance accuracy vs speed	Use 1-10 ns for high-frequency switching
Reuse gecko instance	Avoid reconnection delay	Create once, use throughout script

Error Handling

Always wrap RMI calls in try-catch blocks:

try
    gecko.openFile('circuit.ipes');
    gecko.startSimulation();
catch ME
    fprintf('Error: %s\n', ME.message);
    % Handle cleanup
end

Example: Complete Efficiency vs Load Sweep¶

% Full end-to-end example: measure efficiency across load range
gecko = GeckoRemoteInterface.getInstance();
gecko.openFile('buck_converter.ipes');

loads = [2, 5, 10, 20, 50];  % Ohms
efficiency = [];

for R = loads
    gecko.setParameter('R_load', R);
    gecko.setSimulationTime(0.01);
    gecko.setTimeStep(1e-7);

    gecko.startSimulation();
    while gecko.isSimulationRunning()
        pause(0.05);
    end

    Pin = gecko.getMean('SCOPE', 'Pin');
    Pout = gecko.getMean('SCOPE', 'Pout');
    eta = Pout / Pin * 100;

    efficiency = [efficiency; eta];
    fprintf('Load: %g Ohm -> Efficiency: %.1f%%\n', R, eta);
end

% Plot results
figure;
plot(loads, efficiency, 'b-o', 'LineWidth', 2);
xlabel('Load (Ohms)');
ylabel('Efficiency (%)');
grid on;
title('Buck Converter: Efficiency vs Load');

References¶

Tutorial guide: resources/tutorials/7xx_scripting_automation/702_matlab_integration/
Example circuits in resources/examples/
MATLAB documentation: Calling Java from MATLAB