Home // ICSNC 2019, The Fourteenth International Conference on Systems and Networks Communications // View article

Genetic Algorithm For LoRa Transmission Parameter Selection

Authors:
Aghiles Djoudi
Rafik Zitouni
Nawel Zangar
Laurent George

Keywords: Genetic algorithm; Fuzzy logic; LoRaWAN; Adap- tive Data Rate (ADR).

Abstract:
The exponential growth of Internet of things (IoT ) applications in both industry and academic research raises many questions in wireless sensor networks. Heterogeneous networks of IoT devices strongly depend on the ability of IoT devices to adapt their data transmission parameters to each application requirement. One of the most important problem of the emerging IoT networks is the limitation in terms of energy consumption and computation capability. These limitations could be addressed by using the edge computing to unload IoT devices from additional computation tasks. Our work is motivated by the idea of matching each transmission configuration with a reward and cost values to satisfy applications constraints. Our goal is to make IoT devices able to select the optimal configuration and send their data to the gateway with the QoS required by IoT applications. In this work, we use LoRa network to evaluate the efficiency of our algorithm. Determining the best configuration among 6720 LaRa transmission settings is challenging. The difficulty is mainly due to the lack of tools that could take all applications requirements into account to select the best settings. To address this problem, we use a genetic algorithm in an edge computing to select the transmission parameters needed by the application. Each LoRa configuration represents a feature that needs to be selected to match better the QoS criteria. Particularly, we analyze the impact of selecting one configuration in 3 kinds of applications: text, voice and image transmission by modeling a new adaptive data rate selection process.

Pages: 53 to 54

Copyright: Copyright (c) IARIA, 2019

Publication date: November 24, 2019

Published in: conference

ISSN: 2163-9027

ISBN: 978-1-61208-753-5

Location: Valencia, Spain

Dates: from November 24, 2019 to November 28, 2019