Simulation and Performance Analysis of a Weighted Routing Algorithm for Industrial Wireless Sensor Networks

Industrial Wireless Sensor Networks such as Wire-lessHART and ISA SP100.11a are attractive for automation and process control applications. In those standards, routing is a relevant aspect that affects network performance. Routing algorithms have been proposed over the last years for the constructio...

Full description

Saved in:
Bibliographic Details
Published in2018 VIII Brazilian Symposium on Computing Systems Engineering (SBESC) pp. 24 - 30
Main Authors Kunzel, Gustavo, Pedroso Cainelli, Gustavo, Feldman, Max, Muller, Ivan, Pereira, Carlos Eduardo
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2018
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Industrial Wireless Sensor Networks such as Wire-lessHART and ISA SP100.11a are attractive for automation and process control applications. In those standards, routing is a relevant aspect that affects network performance. Routing algorithms have been proposed over the last years for the construction of routes and graphs with path redundancy to increase reliability. Those routing algorithms are evaluated with different parameters, scenarios and operating conditions that make difficult to compare them. They also use simulators with protocols that have little similarity to the current standard protocols. This work aims to analyze the performance of an algorithm that uses weights to define routes and compare it with other state-of-the-art algorithms. Using a simulation environment with the complete implementation of the WirelessHART stack, the algorithm was subjected to random topologies representing recurrent IWSNs application scenarios to assess how latency, network lifetime, packet error rate and graph reliability were affected by changes in the weights and also to compare its performance over other algorithms. The results show that, within the scenarios tested, some sets of weights are able to balance parameters such as latency and network lifetime, and that the empirical adjustment of weights may be unsuitable for practical applications, and new approaches must be used to aid in those adjustments.
ISSN:2324-7894
DOI:10.1109/SBESC.2018.00013