Toward Improving the Scalability of LR-FHSS: Transmission Policies and Performance Analysis

Low power wide area networks (LPWANs) have revolutionized IoT connectivity by enabling long-range communication with low cost end devices and minimal power consumption. LoRaWAN is the prominent LPWAN technology, while long range-frequency hopping spread spectrum (LR-FHSS) is a newly introduced modulation scheme within the LoRaWAN framework designed to offer improved scalability than the conventional LoRa modulation. There remains a need for research to mature the technology further. Particularly, in massive networks, ensuring reliability remains a challenge. This study examines three message transmission techniques, concatenation, embedded retransmission, and replication, in LR-FHSS networks under different traffic loads, using both theoretical and simulation-based approaches, while measuring success probability, delay, and energy consumption. Concatenation merges multiple messages into a single payload, reducing the number of transmissions, packet overhead, and alleviating network congestion. However, it increases latency, making it more suitable for delay-tolerant applications. Embedded retransmission attaches copies of previous messages to subsequent packets, balancing reliability and delay, and performing well in low to moderate-density networks. Replication, which involves transmitting independent copies of the same packet, enhances reliability but can contribute to network congestion, limiting its effectiveness in high-density environments. Overall, the results indicate that concatenation offers a clear advantage over the other investigated techniques in the analyzed scenarios, while achieving up to 44% higher message success probability compared to regular LR-FHSS at 50,000 end devices.