Low-cost telemetry for environmental monitoring
Between 2000 and 2019, over one million people died due to approximately 7,000 disasters worldwide, resulting in economic losses estimated at around 3 trillion USD. Hydrological disasters caused more than 7 million deaths and affected about 3.8 billion people globally between 1900 and 2020. In recent years, extreme events have intensified and become more frequent, largely due to climate change, increasing risks to critical infrastructure such as dams. In Brazil alone, from 2000 to 2024, hydrological disasters caused nearly 4,000 deaths and impacted 12 million people. Continuous environmental monitoring is crucial to mitigate the effects of natural and anthropogenic disasters; however, commercial monitoring systems are often prohibitively expensive, limiting their deployment in vulnerable regions.
Advances in low-cost sensors and IoT technologies—especially those based on LoRa and MQTT protocols—enable affordable, real-time data acquisition and transmission even in remote areas without electrical power or Wi-Fi connectivity. This study presents the development of a low-cost telemetry system using these technologies for remote environmental monitoring, equipped with sensors for river level, air, and water temperature, along with communication range testing, demonstrating a scalable solution for improving disaster risk management and environmental monitoring.
The developed system consists of an operational setup for environmental monitoring structured into three distinct stations: a monitoring station, an intermediate station, and a central station, all connected via a LoRa module (Figure 1). The monitoring station continuously measures environmental parameters and can include various peripherals for monitoring or system control. In this example, the monitoring station is equipped with temperature, humidity, and distance sensors and is configured to operate in deep sleep mode to optimize energy consumption.
The process begins when the intermediate station wakes up from deep sleep and sends a request to the monitoring station via LoRa communication. Upon receiving the request, the monitoring station also wakes from deep sleep, collects environmental data, and packages it in a standardized JSON structure for transmission. After the intermediate station receives the data, it forwards the collected information to the central station, again using LoRa communication. Once the data transmission is complete, the intermediate station returns to deep sleep mode until the next request cycle. Finally, the central station receives the data for storage and further processing, making it available via MQTT through services such as HiveMQ.
Experiments were conducted to evaluate the signal range performance of the LoRa EBYTE E32 433T20D module using different antenna types, in two distinct scenarios: an indoor environment (inside a laboratory) and an open-field outdoor setting. In the indoor tests, the standard bidirectional SMA antenna achieved a maximum range of 129 meters. In contrast, replacing it with a fiberglass antenna under the same conditions extended the range to 250 meters — an increase of approximately 94%. Overall, the fiberglass antenna consistently demonstrated superior performance, with an average range improvement of 59% compared to the SMA antenna across all tested conditions. Additionally, two different LoRa modules were compared: the EBYTE E32 433T20D and the EBYTE E32 433T33D. The results did not show a consistent advantage of one module over the other, as their performance varied with the specific combination of antenna type and environment. For example, the 20D module performed better indoors when paired with the fiberglass antenna (16% improvement) and outdoors with the SMA antenna (15%). Meanwhile, the 33D module showed slightly better results indoors with the SMA (12%) and outdoors with the fiberglass antenna (19%). Despite the 33D's higher transmission power (30 dBm vs. 20 dBm), both modules use the same core chip, the Semtech SX1278, known for enabling long-range communication in LoRa systems.
The field tests were carried out in non-line-of-sight (NLoS) conditions, with the indoor environment experiencing greater interference due to physical structures. In these conditions, the maximum ranges recorded were 250 meters indoors and 502 meters in the open field, demonstrating an average improvement of 119% in range in outdoor settings. These findings align with prior studies that highlight how signal propagation is significantly affected by environmental obstructions. In ideal line-of-sight (LoS) conditions, such as elevated antenna placement and unobstructed terrain, LoRa communication can exceed distances of 15 km, but such scenarios are rarely applicable in real-world monitoring deployments. In urban environments with dense buildings, the effective range may be drastically reduced; for example, other studies reported a maximum of 700 meters in high-density areas, compared to 7 km in rural settings. To address range limitations, expanding the monitoring network with intermediate nodes is a viable strategy. These nodes can act both as local data collectors and signal relays, forming a multi-hop communication network that increases system coverage. However, while effective in localized deployments, the scalability of LoRa networks is limited. Research suggests that good performance is maintained with up to 120 nodes over areas of approximately 38,000 m², but further expansion may lead to increased interference and decreased data reliability due to lower packet delivery rates.<\p>
