# Experimental Evaluation of LoRaWAN Connectivity Reliability in Remote Rural Areas of Mozambique

## TL;DR

LoRaWAN shows reliable connectivity in rural Mozambique, with performance affected by terrain and elevation, offering a low-cost solution for remote areas.

## Contribution

Empirical validation of LoRaWAN's reliability in a remote, infrastructure-limited environment with no electricity or mobile network.

## Key findings

- LoRaWAN achieved over 89% packet delivery reliability in a rural Mozambican setting.
- Signal strength and quality degraded significantly at lower altitudes due to terrain effects.
- Gateways installed on natural elevations improve coverage in uneven rural topography.

## Abstract

What are the main findings?
LoRaWAN demonstrated high packet delivery reliability (>89%) in a rural Mozambican environment with no electricity or mobile network.Signal strength and quality degraded significantly in lower altitudes, confirming the strong influence of terrain on connectivity.

LoRaWAN demonstrated high packet delivery reliability (>89%) in a rural Mozambican environment with no electricity or mobile network.

Signal strength and quality degraded significantly in lower altitudes, confirming the strong influence of terrain on connectivity.

What is the implication of the main finding?
Installing gateways on natural elevations improves LoRaWAN coverage in rural areas with uneven topography.LoRaWAN is a viable, low-cost, and scalable solution for smart agriculture and digital inclusion in disconnected regions.

Installing gateways on natural elevations improves LoRaWAN coverage in rural areas with uneven topography.

LoRaWAN is a viable, low-cost, and scalable solution for smart agriculture and digital inclusion in disconnected regions.

This paper presents an experimental evaluation of the connectivity reliability of a LoRaWAN (Long Range Wide Area Network), deployed in a rural area of Mozambique, focusing on the influence of distance and relative altitude between end nodes and the gateway. The absence of telecommunications and power infrastructure in the study region provided a realistic and challenging scenario to assess LoRaWAN’s feasibility as a low-cost, low-power solution for remote sensing in disconnected environments. Field trials were conducted using an Arduino-based node (with 2 dBi antenna) powered by a 2200 mAh power bank, with no GPS or cellular support. Data were collected at four georeferenced points along a 1 km path, capturing Received Signal Strength Indicator (RSSI), Signal-to-Noise Ratio (SNR), and Packet Delivery Rate (PDR). Results confirmed that both distance and terrain elevation strongly affect performance, with significantly degraded metrics when the end nodes were located at lower altitudes relative to the gateway. Despite operational constraints, such as the need for manual firmware resets and lack of real-time monitoring, the network consistently achieved PDR above 89% and remained operational autonomously for over 24 h. The study highlights the effectiveness of installing gateways on natural elevations to improve coverage and demonstrates that even with basic hardware, LoRaWAN (Low Power Wide Area Network), is a viable and scalable option for rural connectivity. These findings offer valuable empirical evidence to promote national digital inclusion policies and future LPWAN deployments.