Data Driven Stochastic Game Network-Based Smart Home Monitoring System Using IoT-Enabled Edge Computing Environments

Edge computing plays a crucial role in the processing of Consumer Internet of Things (IoT)-enabled latency-sensitive applications. In smart homes, dynamic action strategies based on multiple IoT objects with edge processing can be the best solution for handling adverse events. To overcome these chal...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on consumer electronics 2024-06, p.1-1
Main Authors: Verma, Prabal, Sood, Sandeep K., Kaur, Harkiran, Kumar, Mohit, Wu, Huaming, Gill, Sukhpal Singh
Format: Article
Language:English
Subjects:
Consumer IoT
Data-Driven Modeling
Edge Computing
Game theory
Games
Intelligent sensors
Internet of Things
Sensors
Smart homes
Stochastic Game Net
Stochastic processes
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Edge computing plays a crucial role in the processing of Consumer Internet of Things (IoT)-enabled latency-sensitive applications. In smart homes, dynamic action strategies based on multiple IoT objects with edge processing can be the best solution for handling adverse events. To overcome these challenges, the use of Stochastic Game Net (SGN) forming IoT devices as players with predefined action sets is one of the feasible solutions. Relative to this context, the edge-assisted IoT-enabled data-driven SGN model is proposed to handle various events in the smart home environment. Stochastic Petri Nets (SPNs) and game theory are integrated into our proposed model to build data-driven dynamic SGNs for the smart home environment. Dynamic SGNs for a comprehensive smart home system are generated in real-time through transitions based on sensor data, enhancing interoperability and scalability in smart home environments. We use the Net logo tool and state-of-the-art smart home sensor datasets to generate dynamic SGNs for various events. Experimental results demonstrate the effectiveness of the proposed model within a data-driven smart home environment. It shows that the present work significantly outperforms other state-of-the-art techniques in terms of decision-making at the edge layer. Moreover, using the proposed system the energy efficacy increased to around 39mJ/K nodes, and the average temporal delay for different events was reduced significantly.
ISSN:0098-3063
1558-4127
DOI:10.1109/TCE.2024.3411657