Related Experiment Video
Updated: Mar 21, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
MIMO transmit scheme based on morphological perceptron with competitive learning.
Raul Ambrozio Valente1, Taufik Abrão1
1Department of Electrical Engineering, State University of Londrina (DEEL-UEL), Rod. Celso Garcia Cid - PR445, s/n, Campus Universitário, P.O. Box 10.011, 86057-970, Londrina, PR, Brazil.
A novel artificial neural network (ANN) MIMO scheme doubles spectral efficiency by decoding two symbols per time slot. While slightly reducing bit-error rate (BER) performance, it offers improved complexity over traditional methods.
Area of Science:
- Wireless communication systems
- Signal processing
- Artificial intelligence in telecommunications
Background:
- Multi-input Multi-Output (MIMO) systems enhance wireless communication capacity.
- Existing MIMO schemes like Alamouti and ML-MIMO have limitations in spectral efficiency and complexity.
- Artificial Neural Networks (ANNs) offer potential for advanced signal processing in MIMO.
Purpose of the Study:
- To introduce a new MIMO transmit scheme utilizing an artificial neural network (ANN).
- To enhance spectral efficiency and optimize detection complexity in MIMO systems.
- To evaluate the performance of the proposed scheme against established MIMO techniques.
Main Methods:
- Deployment of a morphological perceptron with competitive learning (MP/CL) as the MIMO detection decision rule.
- Implementation of a novel MIMO transmit scheme leveraging ANN.
- Comparative performance analysis against Alamouti and Maximum-Likelihood MIMO (ML-MIMO) detectors.
- Evaluation under varying channel information conditions.
Main Results:
- The proposed ANN-aided MIMO scheme achieves double the spectral efficiency compared to the Alamouti scheme.
- The scheme exhibits polynomial complexity relative to modulation order, reducing to linear complexity for longer data streams.
- A slight reduction in bit-error rate (BER) performance was observed due to partial loss of diversity gain under space-time coding.
Conclusions:
- The proposed MP/CL-NN MIMO scheme presents a viable alternative for enhanced spectral efficiency in wireless communications.
- The trade-off between spectral efficiency gains and minor BER performance reduction is a key consideration.
- ANNs demonstrate significant potential for improving MIMO system performance and complexity management.
Related Concept Videos
Associative Learning
Classical conditioning, also known...
Purposive Learning
Observational Learning
Introduction to Learning
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...
Multi-input and Multi-variable systems
In the absence of...
Improving Translational Accuracy
