# Cognitive Radio–Based Ionospheric Scintillation Detection: A Low-Cost Framework for GNSS Detection and Monitoring in Equatorial Regions

## TL;DR

A low-cost system using cognitive radio and machine learning effectively detects and monitors ionospheric scintillation in equatorial regions, improving GNSS reliability.

## Contribution

A novel low-cost framework integrating cognitive radio and ML for real-time ionospheric scintillation detection in equatorial regions.

## Key findings

- The LCSL system successfully detected and monitored ionospheric scintillation in Bogotá, Cartagena, and Santa Marta.
- DIMSUMnet outperformed Adapt4 XG1 in GNSS signal stability and interference detection under scintillation conditions.
- The system validated its accuracy by comparing data with historical records from the Geophysical Institute of Peru (IGP).

## Abstract

What are the main findings?
The proposed Low-Cost Scintillation Laboratory (LCSL), integrating Software-Defined Radio, Cognitive Radio, and Machine Learning, demonstrated effective real-time detection and monitoring of ionospheric scintillation in equatorial and low-latitude regions, with successful validation using data from Bogotá, Cartagena, and Santa Marta.The comparative evaluation of cognitive radio algorithms showed that DIMSUMnet provides superior GNSS signal stability, higher C/N0, and faster interference detection under scintillation conditions, outperforming Adapt4 XG1 in environments affected by equatorial plasma bubbles.

The proposed Low-Cost Scintillation Laboratory (LCSL), integrating Software-Defined Radio, Cognitive Radio, and Machine Learning, demonstrated effective real-time detection and monitoring of ionospheric scintillation in equatorial and low-latitude regions, with successful validation using data from Bogotá, Cartagena, and Santa Marta.

The comparative evaluation of cognitive radio algorithms showed that DIMSUMnet provides superior GNSS signal stability, higher C/N0, and faster interference detection under scintillation conditions, outperforming Adapt4 XG1 in environments affected by equatorial plasma bubbles.

What are the implications of the main findings?
The results confirm that low-cost, cognitive radio–based infrastructures can significantly enhance GNSS resilience in countries with limited space weather monitoring capabilities, reducing dependence on expensive commercial scintillation receivers.The demonstrated adaptability of cognitive radio and ML-based approaches supports their use as a scalable foundation for national and regional ionospheric monitoring networks, with direct benefits for aviation, navigation, and satellite-dependent critical systems in equatorial regions.

The results confirm that low-cost, cognitive radio–based infrastructures can significantly enhance GNSS resilience in countries with limited space weather monitoring capabilities, reducing dependence on expensive commercial scintillation receivers.

The demonstrated adaptability of cognitive radio and ML-based approaches supports their use as a scalable foundation for national and regional ionospheric monitoring networks, with direct benefits for aviation, navigation, and satellite-dependent critical systems in equatorial regions.

Global Navigation Satellite Systems (GNSS) are highly affected in equatorial regions, especially due to the formation of Equatorial Plasma Bubbles (EPBs), which cause disturbances in the ionosphere resulting in different forms of signal degradation. Despite Colombia’s privileged geographic position, its limited monitoring infrastructure hinders the detection and mitigation of these effects. This study proposes the development of a Low-Cost Scintillation Laboratory (LCSL) using a cognitive radio–based approach for real-time scintillation monitoring, aimed at improving GNSS reliability. The system was designed following a Systems Engineering methodology, defining functional architectures and constraints. A communication system model was developed to account for EPBs’ effects on GNSS signals, while cognitive radio algorithms within a Software-Defined Radio (SDR) framework enabled real-time detection, monitoring, and alert generation. To implement this approach, monitoring stations were deployed in Bogotá, Cartagena, and Santa Marta utilized low-cost GNSS receivers integrated with Machine Learning (ML) algorithms for the automatic classification of scintillation events. Additionally, the system’s accuracy was validated by comparing experimental data with historical records from the Geophysical Institute of Peru (IGP). The results demonstrated that the integration of cognitive radio and ML-based detection enhanced precision and adaptability compared to traditional methods. The network of monitoring stations effectively validated the system’s performance, providing valuable insights into equatorial ionospheric dynamics. This study contributes to the advancement of monitoring methodologies and highlights the importance of accessible infrastructure for mitigating EPB effects on GNSS, ultimately fostering more resilient navigation and communication systems.