Multi-class random access wireless network

This paper presents new analytical results for evaluating the ALOHA-like multi-class random access wireless network’s performance. The proposed model is motivated by the growth of low-power wireless networks that employ random access protocols. In particular we compare our analytical formulation with system-level simulations of Long Range (LoRa) technology. We show that the proposed formulation provides an accurate approximation of LoRaWAN performance capturing its main trade-offs. The main contributions are (i) an extensive analysis of the impact of different LoRa spreading factors (SFs) allocation strategies including area intersection among SFs which is little explored in the literature and represents the optimal approach under some conditions; and (ii) the optimal proportion of users that maximizes the network throughput for each class and for each allocation strategy considered in the paper.