Fragment-Level Macro-Diversity Reception in LoRaWAN Networks with LR-FHSS
1 st Samer Lahoud, 2 nd Kinda Khawam · 2025
Read Fragment-Level Macro-Diversity Reception in LoRaWAN Networks with LR-FHSS by 1 st Samer Lahoud, 2 nd Kinda Khawam (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-01689.txt.
Introduction
Samer Lahoud and Kinda Khawam propose a smarter way to receive long-range radio signals in a paper from two thousand twenty-five. A frequency-hopping variant of LoRaWAN spreads packets into fragments for capacity and resilience. Today, though, one gateway must collect a whole packet alone.
Their macro-diversity scheme lets several gateways pool the fragments each of them hears. A stochastic geometry model captures header repetition, fragmentation, and coding redundancy. Closed-form formulas give success rates under interference. Simulations show large capacity gains over single-gateway reception. The authors argue fragment sharing could scale future low-power wide-area networks. The method uses gateways already deployed, turning overlap from waste into strength.
Catalog record
| CREATOR | 1 st Samer Lahoud, 2 nd Kinda Khawam |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-01689
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Fragment-Level Macro-Diversity Reception in LoRaWAN Networks with LR-FHSS 1 st Samer Lahoud, 2 nd Kinda Khawam Abstract The rapid expansion of Internet of Things (IoT) deployments demands wireless protocols that combine high scalability with robust performance. Long Range–Frequency Hopping Spread Spectrum (LR-FHSS) extends LoRaWAN by increasing capacity and resilience through frequency hopping and redundancy. However, current deployments require packet reconstruction at a single gateway, limiting the benefits of LR-FHSS. This paper proposes a macro-diversity reception strategy where multiple gateways collectively receive and combine payload fragments. We develop a stochastic geometry-based analytical model that captures the impact of header repetition, payload fragmentation, and coding redundancy. Closed-form expressions quantify success probabilities under interference, and numerical evaluations demonstrate significant capacity gains over nearest-gateway reception. These results highlight the potential of fragment-level macro-diversity to improve scalability and reliability in future LPWAN deployments. I Introduction The Internet of Things (IoT) continues to expand rapidly, with applications ranging from smart agriculture and industrial monitoring to infrastructure management and environmental sensing. Supporting this growth requires communication protocols that can sustain long-range connectivity, tolerate interference, and operate at minimal energy cost. Low Power Wide Area Networks (LPWANs), and in particular LoRaWAN, have become foundational to these deployments due to their low power operation and extended coverage in unlicensed sub-GHz bands. However, as IoT networks scale to thousands or millions of devices, LoRaWAN faces increasing performance bottlenecks. The original Chirp Spread Spectrum (CSS)-based LoRa modulation suffers from limited spectral efficiency and high collision rates under dense deployments, largely due to its reliance on pure ALOHA channel access and fixed spreading factors. To address these limitations, Long Range–Frequency Hopping Spread Spectrum (LR-FHSS) has been introduced as a physical layer enhancement for LoRaWAN [ 1 ] . LR-FHSS enables devices to transmit using intra-packet frequency hopping over narrowband subcarriers, improving resilience to interference and achieving compliance with duty-cycle and dwell-time constraints through spectral spreading. LR-FHSS introduces two key mechanisms that differentiate it from LoRaWAN networks using LoRa modulation: (i) header repetition, where multiple replicas of the packet header are transmitted to increase the chance of synchronization; and (ii) payload fragmentation, where the packet is split into smaller fragments transmitted over different frequencies. Payload decoding relies on erasure coding, enabling successful recovery even if some fragments are lost. While LR-FHSS shows promising improvements in empirical studies, existing analytical evaluations remain limited in scope. Most models assume that a packet must be fully reconstructed at a single gateway, thereby underutilizing the inherent redundancy of LR-FHSS. Moreover, several critical aspects are inadequately addressed: interference is often simplified without accounting for the hopping structure; device and gateway distributions are commonly modeled using grids instead of stochastic processes; and macro-diversity reception, where multiple gateways collaborate to decode a packet, is never considered. These limitations obscure the LR-FHSS performance bounds, especially in dense networks. This work proposes a fragment-level macro-diversity reception strategy for LR-FHSS-enabled LoRaWAN, where multiple gateways jointly collect and forward payload fragments, enabling distributed packet reconstruction. To evaluate this strategy, we develop a stochastic geometry-based analytical framework that models interference, spatial randomness, and fragment redundancy. We derive closed-form expressions for the packet success probability under macro-diversity, accounting for key LR-FHSS parameters such as header repetition, payload fragmentation, and coding rate. We validate our model through numerical evaluations comparing macro-diversity with traditional nearest-gateway reception, across different data rate configurations. The results quantify how macro-diversity improves success probability and goodput, and highlight design trade-offs between redundancy, capacity, and interference tolerance. This analysis lays the theoretical foundation for scalable LPWAN deployments using LR-FHSS. II Related Works The introduction of LR-FHSS by Semtech [ 1 ] significantly extends LoRaWAN capabilities by improving spectral efficiency, increasing link budget, and ensuring compliance with regional duty-cycle regulations. Simulation-based evaluations such as [ 2 ] have demonstrated the potential of LR-FHSS to improve network capacity, though under simplified assumptions where collisions are modeled only by time and frequency overlap. More recent works [ 3 , 4 , 5 , 6 , 7 ] have examined aspects such as sequence design, transceiver implementations, energy consumption, and behavior under real-world trace analysis. These studies deepen the empirical understanding of LR-FHSS, particularly in direct-to-satellite and terrestrial IoT deployments. However, most existing evaluations rely on simulation or measurement campaigns, with limited theoretical modeling. Early analytical attempts [ 8 , 9 ] addressed delivery and outage probabilities, but either ignored aggregate interference or approximated it using simplified time-on-air metrics, neglecting fading and spatial randomness. To rigorously model interference in large-scale deployments, stochastic geometry has become essential. Although widely used in cellular network analysis [ 10 ] , its use in LPWAN and LR-FHSS contexts remains limited. The work in [ 11 ] modeled distributed reception in a cell-free IoT network, offering early insights into multi-gateway decoding. Similarly, [ 12 ] investigated the decoding of headerless LR-FHSS frames, illustrating the role of fragment-level behavior in reception success. Macro-diversity reception, in which multiple gateways jointly decode transmissions, has been studied in the context of ALOHA-based LPWANs [ 13 ] , showing improved reliability through spatial diversity. However, these studies assume full-packet reception at each gateway and do not account for the specific features of LR-FHSS such as erasure coding, header repetition, and intra-packet frequency hopping. Consequently, the redundancy mechanisms and fragment-level diversity inherent to LR-FHSS remain underexploited in current models. This work addresses this gap by proposing a macro-diversity reception strategy tailored to LR-FHSS. It enables distributed fragment collection across gateways and supports packet reconstruction via centralized coordination. We develop a stochastic geometry-based framework that captures spatial randomness, interference, and key LR-FHSS design parameters. To the best of our knowledge, this is the first theoretical capacity analysis of fragment-level macro-diversity reception in LR-FHSS-enabled LoRaWAN networks. III LR-FHSS Modulation This section describes the LR-FHSS modulation scheme as deployed in the 902–928 MHz ISM band, which is authorized in Canada and other regions governed by FCC regulations. This configuration is adopted as a representative setting for the analysis. Figure 1 illustrates the time-frequency structure of an LR-FHSS transmission. Fig. 1: Spectrogram of an LR-FHSS packet showing sequential header replicas followed by pseudo-randomly hopped payload fragments. III-A Data Rates and Coding In the FCC band, LR-FHSS supports two standardized data rates: DR5 and DR6. DR5 provides a physical-layer bitrate of 162 bps, uses a convolutional coding rate of 1/3 , and transmits three header replicas. DR6 increases the bitrate to 325 bps with a coding rate of 2/3 and transmits two header replicas. DR5 offers stronger redundancy for robustness under interference, while DR6 reduces time-on-air and improves spectral efficiency. In both cases, the payload is convolutionally encoded and decoded using a Viterbi decoder to enhance resilience to channel impairments. III-B Packet Structure Each LR-FHSS transmission consists of a SyncWord, one or more physical-layer header replicas, and a fragmented payload. The header, transmitted at a fixed rate regardless of the data rate, includes metadata such as the hopping sequence seed, payload length, data rate index, number of header repetitions, and coding rate. Header replicas are transmitted sequentially and enable synchronization with the hopping sequence. The payload is divided into L fragments, each assigned to a subcarrier according to a pseudo-random hopping sequence. Successful decoding requires that at least \mu L fragments be correctly received,