Arif, M. (2026) Implementation of an IoT-Based Rainfall Prediction System Using LoRa and Extreme Gradient Boosting (XGBoost). Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (cover)
21081010199.-cover.pdf Download (1MB) | Preview |
|
|
Text (bab 1)
21081010199.-bab1.pdf Download (81kB) | Preview |
|
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (bab 2)
21081010199.-bab2.pdf Restricted to Repository staff only until 25 May 2028. Download (419kB) |
|
|
Text (bab 3)
21081010199.-bab3.pdf Restricted to Repository staff only until 25 May 2028. Download (585kB) |
|
|
Text (bab 4)
21081010199.-bab4.pdf Restricted to Repository staff only until 25 May 2028. Download (971kB) |
|
|
Text (bab 5)
21081010199.-bab5.pdf Download (8kB) | Preview |
|
|
Text (daftar pustaka)
21081010199.-daftarpustaka.pdf Download (130kB) | Preview |
|
|
Text (lampiran)
21081010199.-lampiran.pdf Restricted to Repository staff only until 25 May 2028. Download (4MB) |
Abstract
Unpredictable weather conditions caused by climate variability present major challenges for the agricultural sector, particularly in determining farming activities such as irrigation and pesticide spraying. Limited access to real-time weather information in rural areas further increases the risk of agricultural losses. Therefore, this study proposes an Internet of Things (IoT)-based rainfall prediction system integrated with LoRa communication and the Extreme Gradient Boosting (XGBoost) algorithm. The developed system consists of an Arduino Uno R3, SHT31 sensor, anemometer, LoRa SX1278 module, ESP32 gateway, and AWS EC2 cloud server using the MQTT protocol. InfluxDB and Grafana are utilized for data storage and monitoring. The prediction model is trained using historical BMKG weather data from 2020–2025 with input features consisting of TN, TX, TAVG, RH_AVG, FF_X, and FF_AVG. Rainfall is categorized into four classes: No Rain, Light Rain, Moderate Rain, and Heavy Rain. Testing results show that the LoRa communication system operated reliably with RSSI values ranging from −81.88 dBm to −74.58 dBm and SNR values ranging from 9.70 dB to 9.96 dB at a testing distance of approximately 483 meters. The Packet Loss Rate (PLR) remained relatively low, indicating stable data transmission. The XGBoost model achieved an accuracy of 74% with weighted precision, recall, and F1-score values of 0.70, 0.74, and 0.71, respectively. Overall, the developed system supports realtime environmental monitoring and rainfall prediction for agricultural applications.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | M. Arif | ||||||||||||
| Date Deposited: | 29 May 2026 01:27 | ||||||||||||
| Last Modified: | 29 May 2026 01:27 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/52679 |
Actions (login required)
 |
View Item |