Data Compression Strategies for Enhancing IoRT Communications over Heterogeneous Terrestrial-Satellite Networks

Data transmission over low-speed networks is a subject of global research interest, with a specific focus on overcoming limitations in satellite communication channels. However, little research has addressed the impact of compression on data transmission efficiency in heterogeneous networks using sa...

Full description

Saved in:
Bibliographic Details
Published in:International Symposium on Problems of Redundancy in Information and Control Systems (Online) pp. 189 - 193
Main Authors: Karnaukhov, Alexander, Idelevich, Artem, Rolich, Alexey, Voskov, Leonid
Format: Conference Proceeding
Language:English
Published: IEEE 24.10.2023
Subjects:
Control systems
Data compression
Data preprocessing
Dictionaries
Heterogeneous networks
Internet of Remote Things
IoRT
JSON
LEO
Protobuf
Redundancy
Satellites
Serialization
Terrestrial-Satellite
ISSN:2694-5215
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data transmission over low-speed networks is a subject of global research interest, with a specific focus on overcoming limitations in satellite communication channels. However, little research has addressed the impact of compression on data transmission efficiency in heterogeneous networks using satellite links, especially in the context of the Internet of Remote Things and under poor terrestrial network infrastructure. This study explores various data compression algorithms customized for tiny data in such scenarios. It identifies effective algorithms when combined with ProtoBuf serialization, achieving compression ratios between 3.5 and 8.2 for JSON messages of 30 to 680 bytes using the Huffman algorithm with an extended dictionary. For messages sized 680 to 2,048 bytes, Protobuf combined with LZ78 achieves compression ratios of 3.5 to 4.5. Moreover, a novel data preprocessing method is introduced, boosting processing performance by up to 27 times for messages under 680 bytes in size. These findings contribute to IoRT data serialization and compression and can potentially enhance existing methods.
ISSN:2694-5215
DOI:10.1109/Redundancy59964.2023.10330191