TY - JOUR
T1 - FogETex
T2 - Fog Computing Framework for Electronic Textile Applications
AU - Ozlem, Kadir
AU - Atalay, Asli Tuncay
AU - Atalay, Ozgur
AU - Ince, Gokhan
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2024
Y1 - 2024
N2 - Textile products are present in almost every aspect of human life. With the introduction of electronic textiles, textile products have become capable of converting various physiological and environmental stimuli into electrical signals, many of which are of vital importance to humans. Therefore, these products require real-time (low-latency) and robust computing systems. However, due to comfort considerations, they cannot accommodate powerful computing resources. In this study, a novel fog computing-based framework (FogETex) is proposed to meet the needs of e-textile applications. FogETex is a Platform-as-a-Service model that is cross-platform supported, scalable, and operates in real-time. This framework encompasses end-to-end integration of the system including textile-based Internet of Things (T-IoT) device, fog devices, and the cloud. Fog devices consist of a broker that manages the fog node and a worker that handles incoming computation requests. Sensor data is transmitted to the fog node through a mobile application, and system architecture can be monitored through developed user interfaces. Resource usage from broker devices is monitored in real time to prevent worker devices from experiencing overload. For the system case study, a deep learning-based gait phase analysis application using textile-based capacitive sensors is employed. FogETex was evaluated in terms of time characteristics, resource usage, and network bandwidth usage using a mock client to determine the ideal system performance and an actual client to conduct real-world tests. The fog devices outperformed the cloud system in these metrics. Besides being developed primarily for electronic textile applications, FogETex framework can accommodate other IoT devices as well.
AB - Textile products are present in almost every aspect of human life. With the introduction of electronic textiles, textile products have become capable of converting various physiological and environmental stimuli into electrical signals, many of which are of vital importance to humans. Therefore, these products require real-time (low-latency) and robust computing systems. However, due to comfort considerations, they cannot accommodate powerful computing resources. In this study, a novel fog computing-based framework (FogETex) is proposed to meet the needs of e-textile applications. FogETex is a Platform-as-a-Service model that is cross-platform supported, scalable, and operates in real-time. This framework encompasses end-to-end integration of the system including textile-based Internet of Things (T-IoT) device, fog devices, and the cloud. Fog devices consist of a broker that manages the fog node and a worker that handles incoming computation requests. Sensor data is transmitted to the fog node through a mobile application, and system architecture can be monitored through developed user interfaces. Resource usage from broker devices is monitored in real time to prevent worker devices from experiencing overload. For the system case study, a deep learning-based gait phase analysis application using textile-based capacitive sensors is employed. FogETex was evaluated in terms of time characteristics, resource usage, and network bandwidth usage using a mock client to determine the ideal system performance and an actual client to conduct real-world tests. The fog devices outperformed the cloud system in these metrics. Besides being developed primarily for electronic textile applications, FogETex framework can accommodate other IoT devices as well.
KW - Fog computing
KW - deep learning
KW - electronic textile
KW - gait phase analysis
KW - raspberry pi
KW - socket programming
UR - http://www.scopus.com/inward/record.url?scp=85208680892&partnerID=8YFLogxK
U2 - 10.1109/JIOT.2024.3490981
DO - 10.1109/JIOT.2024.3490981
M3 - Article
AN - SCOPUS:85208680892
SN - 2327-4662
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
ER -