Design and Implementation of Multithreaded CAD-Enabled LoRa Simulator for Massive IoT

An increasing number of smart devices have made wireless sensor networks denser, thus increasing traffic in the network. Massive IoT (mIoT) faces the major problems of collision and retransmission. Long Range (LoRa) network is specially designed to consume low power for long range. With the number of devices, collision and retransmission increase, leading to more power consumption. Carrier sensing using channel activity detection (CAD) is introduced to improve these challenges. Various scheduling algorithms are designed to minimize latency and energy. There is a need for LoRa simulators that can form a testbed for the initial evaluation of these algorithms. Existing simulators do not provide the simulation of features like CAD. This paper introduces a multithreaded simulator, LoRaCAD, specially designed to evaluate CAD sensing algorithms. Multithreading has enabled us to simulate a real-time environment where devices independently communicate with the LoRa gateway. We further evaluate the existing reinforcement learning-based algorithm LoRa-RL [1] on LoRaSim and LoRaCAD based on our primary studies in [2], find how energy consumption differs, and highlight how carrier sensing improves massive IoT connectivity.