Physicists worldwide rely on MATLAB and Simulink to perform both exploratory and computationally demanding simulations. The matrix-oriented computing environment makes MATLAB a natural choice for rapid code development in search of novel physics and collaborating with the industry. MATLAB and Simulink also provide an integrated approach for hardware code generation, data acquisition, real-time simulation and testing, data analysis, and scalable computations.
Physicists choose MATLAB and Simulink to:
- Integrate AI methods with workflows to analyze and visualize data
- Run particle accelerators
- Process signals received by radio telescopes and gravitational wave detectors
- Control a variety of “small lab” hardware
- Compare simulations with experimental data
- Teach physics and share work with other physicists
“For LIGO, we used MATLAB to analyze the fundamental noises that limit gravitational wave detector performance, calculate the optical response of our interferometers, and verify the entire control chain ...”
Matthew Evans, MIT
Using MATLAB and Simulink for Physics
MATLAB and Simulink for Physics in “Small Labs”
Physicists use MATLAB and Simulink to connect to and control lab hardware, such as custom microscopes, perform various spectroscopic analyses, develop AI-enhanced sensors, and analyze data.
Highly optimized operations on dense and sparse matrices are convenient for rapid code development to simulate classical and quantum many-body systems. Symbolic math enables calculation to arbitrary precision.
Using MATLAB and Simulink, physicists can:
- Automatically generate HDL and C/C++ code for hardware connectivity, and use C/C++ code within MATLAB
- Control and acquire data in real-time from hardware and instruments
- Deploy computations on clusters for big data or demanding calculations
- Share codes using intuitive Live Scripts and GUIs
- Accelerate AI and other computationally intensive analytics on GPUs
- Scale computations to clusters and clouds using MATLAB Parallel Server
- Teach physics using an interactive course curriculum
MATLAB and Simulink for Physics in “Big Labs”
MATLAB and Simulink enable rapid prototyping and modeling real-time control systems for big experiments, such as LIGO. Readability of the code and backward compatibility are particularly attractive features of MATLAB for large, long-term collaborations.
Accelerator physicists use MATLAB to control synchrotrons and Linacs around the globe. With MATLAB they also monitor particle beams and compare beam behavior with simulated versions. MATLAB and some community toolboxes written by accelerator physicists provide a rapid prototyping and deployment system, which has been well-tested world-wide.
MATLAB and Simulink enable physicists and engineers to:
- Design filter modules enhanced with AI for noise suppression and signal processing
- Eliminate unscheduled “big machine” downtime
- Interact with hardware using intuitive scripting and GUI environments
- Use code generation for PLCs, FPGAs, and ASICs
- Design control systems
- Seamlessly transition from desktop simulation to real-time testing with Simulink Real-Time and Speedgoat
Customer Success
- MATLAB Based Toolboxes to Simulate and Control High Energy Particle Accelerators
- Containing the Power of the Sun: The Path to Carbon Independence Through Nuclear Fusion
- Developing Radio Astronomy Instruments with Simulink Libraries
- RF Data Analysis and System Design for Scientific Applications (3 videos)
Community Toolboxes
Learn More
MATLAB for Medical Physics
Medical physicists use MATLAB as a unified platform for treatment planning, which is particularly convenient for education and research. With MATLAB, radiation therapists can use semi-automatic labeling tools that facilitate integrating AI methods in workflows for applications such as image guided radiation therapy.
With MATLAB and community toolboxes, medical physicists can:
- Produce clinically accurate treatment plans
- Plan intensity-modulated radiation therapy for multiple modalities
- Preprocess and export data to train deep learning auto-segmentation models, beginning with DICOM or other file formats
- Train deep learning models using multi-channel images by various transformations and methods of populating image channels
- Perform time domain acoustic and ultrasound simulations in complex and tissue-realistic media