Applying Machine Learning to Equatorial Plasma Bubble Now Casting on CubeSats - Conference Paper

Aug 5, 2026

Upload by Oliver

Equatorial plasma bubbles are low-density regions in Earth’s ionosphere that form near the equator after sunset, expand upward, and persist through the night, stretching along magnetic field lines. They can cause radio scintillations that degrade navigation systems such as GPS, creating a need for near real-time detection. The work applies machine learning to CubeSat-based plasma density time series from the SPORT instrument to locate bubbles, comparing ARIMA, Random Forests, Gradient Boosting Machines, and LSTM networks using confusion-matrix metrics and computational complexity.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Abstract

1 Introduction

2 Background

SSC24-WV-06 Applying Machine Learning to Equatorial Plasma Bubble Now Casting on CubeSats Benjamin Lewis, Shawn Jones, Charles Swenson, Kevin Moon, Mario Harper Utah State University [Address (same for all)]; 435-557-1682 [benjamin.lewis@usu.edu](benjamin.lewis@usu.edu) ABSTRACT Equatorial plasma bubbles are a space weather phenomenon that occur at low latitudes within the Earth’s ionosphere. These bubbles are regions of low-density plasma that form at the base of the ionosphere and expand upward through the peak and into the topside. They form in the early evening and persist through the nighttime, stretching north and south along magnetic field lines and effecting a sector of longitudes a few degrees wide. These bubbles cause scintillations on radio signals that pass through them, disrupting the performance of systems, such as GPS, throughout the night in regions around the Earth. Due to their potential social impact, there is a desire to know in real time if equatorial plasma bubbles are occurring. This can be achieved using a small satellite in orbit with automated processing to locate bubbles by processing data from in-situ sensors. Inter-satellite communications to a LEO communications constellation allows the possibility of real time detection of equatorial plasma bubbles from such a satellite. Langmuir probes, impedance probes, and ion drift meters are all instruments capable of detecting plasma bubbles. Sensors such as the Scintillation Prediction Observations Research Task (SPORT) Brazil/US CubeSat mission routinely detect equatorial plasma bubbles. This work investigates the application of various machine learning techniques for the detection of plasma bubbles on a CubeSat using time series plasma density data from the SPORT instrument. This paper reviews four machine learning approaches: Auto-regressive Integrated Moving Average (ARIMA) models, Random Forests, Gradient Boosting Machines, and Recurrent Neural Networks (RNN) such as Long Short-Term Memory (LSTM) networks. The models are evaluated using common metrics derived from the confusion matrix (e.g. accuracy, sensitivity, positive predictive value, and F1 score), as well as the computational complexity of the model. Building off of this work, USU will implement a plasma bubble detection algorithm on a low-power edge computing device built for a CubeSat to provide near real-time plasma bubble reporting. USU’s Low-power Array for CubeSat Edge Computing Architecture, Algorithms and Applications (LACE-C3A) project is being funded by NASA’s University SmallSat Technology Partnership Program to develop an FPGA-based compute board that will be capable of performing machine learning algorithms on a CubeSat scale. The near real-time plasma bubble reporting will be one of the applications USU will demonstrate on LACE-C3A. 1 Introduction Space weather phenomena affect the performance of space based communication and navigation systems. In particular, equatorial plasma bubbles are associated with scintillation on RF signals pass through the ionosphere where bubbles are present. These bubbles are depletions in the density of the plasma that propagates from the bottom edge through the layers of the ionosphere. They form near the Earth’s equator in the evening after sunset and their trails persist till after midnight. The sheet like bubbles rapidly rise near the Earth’s magnetic equator being a tens to hundreds of km wide in the east-westward directions but expanding 1000’sof km north-southward along the magnetic field. In- struments, such as Space Weather Probes (SWP) on the joint Brazil-NASA Scintillation Prediction Observations Research Task (SPORT) mission provide in-situ measurements of ionospheric density while crossing these bubbles. Within these along track satellite observations the bubbles are primarily identified as sudden and significant drops in density of an order of magnitude or more. SPORT, a 6U CubeSat, was deployed into a mid inclinati

FAQs

Why is real-time detection of equatorial plasma bubbles important?

Equatorial plasma bubbles produce ionospheric scintillations that disrupt radio signals and can degrade systems such as GPS during the night.

What data and sensors does the approach rely on?

The study uses time series plasma density data from the SPORT CubeSat instrument, which detects plasma bubbles using in-situ measurements.

Which machine learning models are evaluated for bubble detection?

The paper evaluates ARIMA, Random Forests, Gradient Boosting Machines, and recurrent neural networks such as LSTM, comparing them with metrics derived from the confusion matrix and model computational complexity.

Aug 5, 2026

Upload by Oliver