MATLAB

MATLAB
Name	MATLAB
Developer	MathWorks
Released	1984
Latest release	(varies)
Programming language	C, C++, Java
Operating system	Microsoft Windows, macOS, Linux
Genre	Numerical computing, Technical computing

MATLAB is a proprietary numerical computing environment and programming language used for matrix computations, algorithm development, data analysis, visualization, and prototyping. It was created to provide engineers and scientists with high-level tools for linear algebra, signal processing, control systems, and image analysis. Widely adopted in industry, academia, and government, the platform integrates with hardware, simulation environments, and documentation toolchains.

History

MATLAB traces origins to numerical routines developed by Cleve Moler in the 1970s and early 1980s, which built on libraries such as LINPACK and EISPACK and the academic environment of University of New Mexico. The commercial product was co-founded by Jack Little and Steve Bangert when they formed The MathWorks in 1984, paralleling developments in software by vendors like IBM and projects influenced by Fortran and C. Over successive releases, the product absorbed capabilities from competitors and collaborators, integrated with systems like Simulink and toolchains from Microsoft Visual Studio and GNU Compiler Collection, and responded to standards such as IEEE 754 and formats like HDF5. Major milestones include expansions for signal processing, control theory, and support for GPUs and distributed computing influenced by industry trends exemplified by NVIDIA and Amazon Web Services.

Overview and Features

The environment centers on an interactive desktop, plotting engine, and numerical libraries derived from linear algebra packages like LAPACK and sparse solvers used in scientific computing projects such as PETSc. Key features include high-level matrix operations, built-in functions for optimization, statistics, and machine learning influenced by methods from John Tukey and David Cox, and visualization tools comparable to graphics systems used in MATLAB Central community examples. Integration points include connectivity to SQL Server, hardware interfaces for Arduino and Raspberry Pi, and code generation for targets supported by Embedded Coder and standards such as AUTOSAR. The platform supports parallelism on clusters managed by systems like Slurm and cloud services such as Microsoft Azure.

Language and Syntax

The language is matrix-oriented and dynamically typed, borrowing syntax elements and influences from Fortran and C while providing array indexing conventions related to work from Alston Householder. Core constructs include matrix creation, elementwise operators, and function handles; control flow resembles constructs found in Pascal and structured programming advocates like Edsger Dijkstra. Interoperability features allow calling code written in C++, Java, and Python and interfacing with external libraries via foreign function interfaces similar to those used by R extensions. The environment supports object-oriented programming patterns and integrates testing frameworks comparable to unit testing philosophies promoted by Kent Beck.

Toolboxes and Extensions

A large ecosystem of add-ons provides domain-specific functionality: signal processing, control systems, image processing, and machine learning toolboxes resonate with methods from Alan Turing and Norbert Wiener traditions. Specialized toolboxes support communications influenced by standards from 3GPP and IEEE, aerospace workflows related to NASA missions, and automotive toolchains aligned with ISO 26262 and AUTOSAR guidelines. Third-party extensions and community-contributed files are distributed through platforms like File Exchange and collaborations with vendors such as Intel and ARM Holdings. Code generation and hardware deployment tools interoperate with embedded toolchains from Texas Instruments and simulation environments used in projects by Airbus and Boeing.

Applications and Use Cases

The environment is used in research and engineering across domains exemplified by projects at MIT, Stanford University, Caltech, and national laboratories such as Los Alamos National Laboratory. Common applications include control design for systems used by General Motors and Ford Motor Company, signal and image processing in medical devices for organizations like Philips and Siemens, financial modeling for firms in Wall Street and central banking research similar to analyses at Federal Reserve System, and machine learning prototyping adopted by technology companies including Google and Microsoft Corporation. Academic courses on numerical analysis, linear algebra, and computational modeling at universities such as Harvard University and Princeton University frequently use the environment for coursework and labs.

Licensing and Availability

Distributed by The MathWorks, the software is available under proprietary commercial licenses, academic licenses for institutions like University of Oxford and University of Cambridge, and trial or campus-wide site licenses used by consortia such as XSEDE. Licensing models include individual, group, and enterprise agreements, and deployment options cover desktop installations on Microsoft Windows, macOS, and Linux, as well as cloud-hosted offerings through partners like Amazon Web Services and Microsoft Azure. Export controls and compliance follow regulations from agencies such as the Bureau of Industry and Security and licensing arrangements reflect intellectual property practices seen across the software industry.

Category:Numerical software