NEURONAL NETWORK IMPLEMENTATION IN PYTHON FOR HIGHLY NON-LINEAR WIRELESS SYSTEMS (IMPLEMENTACIÓN DE RED NEURONAL EN PYTHON PARA SISTEMAS INALÁMBRICOS ALTAMENTE NO LINEALES)
Resumen
In this work, a system based on a neural network implemented in an open-source programming language is developed in order to operate simple band systems as an alternative to other modeling platforms that require the use of a software license. An artificial neural network implementation in Python language is carried out with multilayer perceptrons, one-dimensional convolutional neural networks, short and long-term memory networks and transformers. In the development of this work, a precision of -17.5249 dB NMSE was obtained by a one-dimentional convolutional neural network for highly non-linear behavior. The objective of developing open-source proposals is to use a free hardware development board that applies to specific wireless systems. The modeling system represents the preliminary stage for linearization and spectral correction processes in the case of RF transmission.Keywords: linearization, modeling, neural networks, wireless transmissions.
Resumen
En este trabajo se desarrolla un sistema basado en una red neuronal implementada en un lenguaje de programación de código abierto con el fin de operar sistemas en banda simple como una alternativa a otras plataformas de modelado que requieren del uso de licencia de software. Se realiza una implementación de redes neuronales artificiales en el lenguaje Python con las arquitecturas de perceptrón multicapa, redes neuronales convolucionales de una dimensión, redes de memoria a corto y largo plazo y transformers. En el desarrollo de este trabajo se obtuvo una precisión de la implementación de la red neuronal convolucional de una dimensión de -17.5249 dB NMSE para un comportamiento altamente no lineal. El objetivo de desarrollar propuestas de código abierto es utilizar tarjetas de desarrollo de hardware libre que apliquen para ciertos sistemas inalámbricos. El sistema de modelado representa la etapa previa para procesos de linealización y corrección espectral en caso de una transmisión de RF.
Palabras Clave: Linealización, modelado, redes neuronales, transmisiones inalámbricas.
Texto completo:
132-145 PDFReferencias
A. Raffo, et al., 2010. Behavioral Modeling of RF front end devices in IEEE Transactions on Microwave Theory and Techniques, Chalmers University of Technology, SE-412 96 Göteborg, Sweeden, 58(4), pp. 710-718, 47.
Bishop, Christopher M. 1995. Neural Networks for Pattern Recognition. Clarendon Press, Oxford University Press.
E. Arabi and S. Ali. 2008. Behavioral Modeling of RF front end devices in Simulink®, Master Thesis, Chalmers University of Technology, SE-412 96 Göteborg, Sweeden, pp. 1-3, 21-26, 47.
Frenzel, Louis E. 2014. Principles of Electronic Communication Systems. Fourth edition, McGraw-Hill, a business unit of The McGraw-Hill Companies, Inc, pp. 272, 590.
Hochreiter, S., and Schmidhuber, J. 1997. Long Short-Term Memory. Neural Computation, 9(8), 1735–1780.
Kiranyaz, Serkan, et al. Apr. 2021. 1D Convolutional Neural Networks and Applications: A Survey. Mechanical Systems and Signal Processing, vol. 151, p. 107398.
Kishore, G. Shyam, and Hemalatha Rallapalli, 2019, Performance Assessment of M-Ary ASK, FSK, PSK, QAM and FQAM in AWGN Channel. 2019 International Conference on Communication and Signal Processing (ICCSP), IEEE, pp. 0273–77.
Rawat, Karun, et al. 2020. Introduction to RF Power Amplifier Design and Architecture. Bandwidth and Efficiency Enhancement in Radio Frequency Power Amplifiers for Wireless Transmitters, by Karun Rawat et al., Springer International Publishing, pp. 1, 7, 76, 85, 340.
Secretaría de Agricultura y Desarrollo Rural, 2019, Plan Nacional de Desarrollo 2019-2024, pp. 61.
Su, Ri-na, et al. Jan. 2019. Digital Pre-Distortion Linearization Modeling Based on Cloud Platform. DEStech Transactions on Computer Science and Engineering, no. msota.
Vaswani, Ashish, et al. June 2017. Attention Is All You Need.
Warren S. Sarle. 1994. Neural Networks and Statistical Models.
Zhang, Yikang, et al. 2019. Vector Decomposition Based Time-Delay Neural Network Behavioral Model for Digital Predistortion of RF Power Amplifiers. IEEE Access, vol. 7, pp. 91559–68.
URL de la licencia: https://creativecommons.org/licenses/by/3.0/deed.es
Pistas Educativas está bajo la Licencia Creative Commons Atribución 3.0 No portada. 
TECNOLÓGICO NACIONAL DE MÉXICO / INSTITUTO TECNOLÓGICO DE CELAYA
Antonio García Cubas Pte #600 esq. Av. Tecnológico, Celaya, Gto. México
Tel. 461 61 17575 Ext 5450 y 5146
pistaseducativas@itcelaya.edu.mx