An Improved Data Compression Framework for Wireless Sensor Networks Using Stacked Convolutional Autoencoder (S-CAE)

Data compression is crucial in the networks as there is limited energy which is accessible to sensor nodes in wireless sensor networks (WSNs). The sensor nodes lifetime is extended enormously by reducing the data reception and transmission. We introduce a stacked convolutional RBM auto-encoder (stac...

Full description

Saved in:
Bibliographic Details
Published in:SN computer science Vol. 4; no. 4; p. 419
Main Authors: Kumble, Lithin, Patil, Kiran Kumari
Format: Journal Article
Language:English
Published: Singapore Springer Nature Singapore 01.07.2023
Springer Nature B.V
Subjects:
Accuracy
Advances in Computational Approaches for Image Processing
Cloud Applications and Network Security
Compression ratio
Computer Imaging
Computer Science
Computer Systems Organization and Communication Networks
Data collection
Data compression
Data Structures and Information Theory
Energy consumption
Energy storage
Information Systems and Communication Service
Nodes
Optimization techniques
Original Research
Pattern Recognition and Graphics
Pressure ratio
Reduction
Sensors
Software Engineering/Programming and Operating Systems
Vision
Wireless Networks
Wireless sensor networks
Energy, consumption optimization
Stacked-CAE
Transfer learning
Data compression
ISSN:2661-8907, 2662-995X, 2661-8907
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data compression is crucial in the networks as there is limited energy which is accessible to sensor nodes in wireless sensor networks (WSNs). The sensor nodes lifetime is extended enormously by reducing the data reception and transmission. We introduce a stacked convolutional RBM auto-encoder (stacked CAE) model for compressing sensor data, which is made up of layers: an encode layer and a decode layer, both of which are discussed. The encode layer is used to compress and decompress data from sensor and then, the decode layer is used for reconstruction and compression of the data from sensors. Both encode and decode layers are comprised of four standard restricted Boltzmann machines, which are employed throughout the system. This work focuses on energy reduction technique by reduction of model’s parameters which in turn reduces the model’s calculation and storage energy. The model's effectiveness is evaluated against the Intel Lab data. The average temperature reconstruction inaccuracy is 0.312 °C, the average percentage RMS difference is 9.84%, and the figures imply that the model's compression ratio is 10. Hence, there is a possibility of minimizing the consumption of energy of node communication in WSNs by 92%. The new model attains higher compression proficiency and remark precision while keeping the same pressure ratio when in comparison with the previous method.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2661-8907
2662-995X
2661-8907
DOI:10.1007/s42979-023-01845-7