A CPN-based model for assessing energy consumption of IoT networks

Diogo Lages, Eric Borba, Eduardo Tavares, Andson M. Balieiro, Erica Souza

Abstract: Wireless sensor networks (WSN) commonly have energy consumption as a remarkable issue. Due to the popularity of Internet of Things (IoT) systems, attention has been devoted to low-power wide-area networks (LPWAN), concerning the influence of transceivers on energy utilization. For instance, the energy for a communication transceiver to send one single bit can be 1500 to 2000 times higher than the energy required to execute a single instruction. Therefore, evaluating the energy consumption of network physical layer for IoT systems is prominent for system design. This work presents a Colored Petri nets (CPN)-based framework for evaluating the energy consumption of LPWAN-based IoT systems, focusing on CPU and communication transceivers. The proposed model has been validated using a hardware platform with ARM7DMI and LoRa, and results indicate an error less than 1.4% and 0.14% for the CPU and network, respectively, and show the feasibility of the proposed model to estimate the impact of packet losses and timeout on the system performance.