On the performance of data-driven approaches for energy efficiency on WiFi and LoRa-based sensors: an experimental study

Most research on energy efficiency in wireless sensor networks considers that the communication subsystem consumes significantly more energy than the sensing and computing ones. In order to verify this widely adopted premise, an experimental study has been conducted on a Pysense sensor shield that utilizes WiFi and LoRa. This paper compares the energy consumption of each subsystem and two data-driven energy conservation algorithms that employ different strategies. The findings of this work indicate that lowering the energy consumption of the communication subsystem is only advantageous when using WiFi but was less effective and promising when using LoRa. Additionally, it demonstrates the importance of simultaneously optimizing the activation of multiple subsystems to minimize energy consumption. The findings of this study, as well as the source code, are available on Github: https://github.com/BouTayehGaby/WSN-energy-consumption-benchmark.