BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION

Human activities can be recognised through reflections of wireless signals which solve the problem of privacy concerns and restriction of the application environment in vision-based recognition. Spiking Neural Networks (SNNs) for human activity recognition (HAR) using Wi-Fi signals has been proposed...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yee Leong Tan, Yan Chiew Wong, Syafeeza Ahmad Radzi
Formato:	article
Lenguaje:	EN
Publicado:	Scientific Research Support Fund of Jordan (SRSF) and Princess Sumaya University for Technology (PSUT) 2021
Materias:	human activity recognition wi-fi signals spiking neural network Information technology T58.5-58.64 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/4a60100d6d6c44e2828ae181088ace2f
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:4a60100d6d6c44e2828ae181088ace2f
record_format	dspace
spelling	oai:doaj.org-article:4a60100d6d6c44e2828ae181088ace2f2021-12-03T07:32:06ZBRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION2413-935110.5455/jjcit.71-1629096728https://doaj.org/article/4a60100d6d6c44e2828ae181088ace2f2021-12-01T00:00:00Zhttp://www.ejmanager.com/fulltextpdf.php?mno=111344https://doaj.org/toc/2413-9351Human activities can be recognised through reflections of wireless signals which solve the problem of privacy concerns and restriction of the application environment in vision-based recognition. Spiking Neural Networks (SNNs) for human activity recognition (HAR) using Wi-Fi signals has been proposed in this work. SNNs are inspired by information processing in biology and processed in a massively parallel fashion. The proposed method reduces processing resources while still maintaining accuracy through using frail but robust to noise spiking signals for information transfer. The performance of HAR by SNNs is compared with other machine learning (ML) networks such as LSTM, Bi-LSTM and GRU models. Significant reduction in memory usage while still having accuracy that is on a par with other ML networks has been observed. More than 70% saving in memory usage has been achieved in SNNs compared with the other existing ML networks, making SNNs potential solutions for edge computing in industrial revolution 4.0. [JJCIT 2021; 7(4.000): 363-372]Yee Leong TanYan Chiew WongSyafeeza Ahmad RadziScientific Research Support Fund of Jordan (SRSF) and Princess Sumaya University for Technology (PSUT)articlehuman activity recognitionwi-fi signalsspiking neural networkInformation technologyT58.5-58.64Electronic computers. Computer scienceQA75.5-76.95ENJordanian Journal of Computers and Information Technology , Vol 7, Iss 4, Pp 363-372 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	human activity recognition wi-fi signals spiking neural network Information technology T58.5-58.64 Electronic computers. Computer science QA75.5-76.95
spellingShingle	human activity recognition wi-fi signals spiking neural network Information technology T58.5-58.64 Electronic computers. Computer science QA75.5-76.95 Yee Leong Tan Yan Chiew Wong Syafeeza Ahmad Radzi BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
description	Human activities can be recognised through reflections of wireless signals which solve the problem of privacy concerns and restriction of the application environment in vision-based recognition. Spiking Neural Networks (SNNs) for human activity recognition (HAR) using Wi-Fi signals has been proposed in this work. SNNs are inspired by information processing in biology and processed in a massively parallel fashion. The proposed method reduces processing resources while still maintaining accuracy through using frail but robust to noise spiking signals for information transfer. The performance of HAR by SNNs is compared with other machine learning (ML) networks such as LSTM, Bi-LSTM and GRU models. Significant reduction in memory usage while still having accuracy that is on a par with other ML networks has been observed. More than 70% saving in memory usage has been achieved in SNNs compared with the other existing ML networks, making SNNs potential solutions for edge computing in industrial revolution 4.0. [JJCIT 2021; 7(4.000): 363-372]
format	article
author	Yee Leong Tan Yan Chiew Wong Syafeeza Ahmad Radzi
author_facet	Yee Leong Tan Yan Chiew Wong Syafeeza Ahmad Radzi
author_sort	Yee Leong Tan
title	BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
title_short	BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
title_full	BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
title_fullStr	BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
title_full_unstemmed	BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION
title_sort	brain-inspired spiking neural networks for wi-fi based human activity recognition
publisher	Scientific Research Support Fund of Jordan (SRSF) and Princess Sumaya University for Technology (PSUT)
publishDate	2021
url	https://doaj.org/article/4a60100d6d6c44e2828ae181088ace2f
work_keys_str_mv	AT yeeleongtan braininspiredspikingneuralnetworksforwifibasedhumanactivityrecognition AT yanchiewwong braininspiredspikingneuralnetworksforwifibasedhumanactivityrecognition AT syafeezaahmadradzi braininspiredspikingneuralnetworksforwifibasedhumanactivityrecognition
_version_	1718373808835919872

BRAIN-INSPIRED SPIKING NEURAL NETWORKS FOR WI-FI BASED HUMAN ACTIVITY RECOGNITION

Ejemplares similares