Improving LoRa Network Simulator for a More Realistic Approach on LoRaWAN

Nowadays, the need for information is rapidly increasing, urging the development of technology to obtain reliable information. Under the scope of Smart Cities, Low Power Wide Area Networks (LPWANs) have been seen as one of the enabling technologies for the data gathering process but, the evaluation of new methods and solutions in large-scale, real world-scenarios, is a challenging and sometimes an impossible task. Thus, the need to use tools enabling the simulation of such networks arose, triggering the development of different network simulators. In this work we focus on the simulation of LoRa networks. Resorting on the LoRa network simulator mostly used in the literature, the LoRaSim, we study the impact of different propagation and the capture-effect models in the performance of a LoRa network. In addition, an extension to LoraSim is implemented, offering the capability to tweak the network on a more detailed fashion, by supporting distinguished parameters (spreading factor, bandwidth, code rate and payload) on the end-devices. Results show that when applying a well-known propagation model (Okumura-Hata) and a more realistic capture-effect on a network that emphasizes the use of orthogonality among all devices, this heterogeneity greatly increases its efficiency and quality.