Beyond Bandpass: Time-Varying Interference Mitigation on In-Situ Magnetic Field Measurements using Singular Spectrum Analysis

Matthew Finley, PhD.; University of Iowa

High-fidelity magnetic field data is critical to advance our understanding of various space physics phenomena near Earth, such as Alfvén waves and field-aligned currents. However, despite attempts at spacecraft magnetic cleanliness, magnetometer measurements are often contaminated by stray magnetic fields produced by other onboard subsystems. Several techniques have been proposed to mitigate this magnetic interference; one common method is to employ a band-stop filter at the frequencies associated with the interfering subsystems. Although effective, this technique is not suitable if the dominant source of magnetic contamination varies in frequency over short intervals of time. This presentation will discuss a method for the mitigation of time-varying magnetic interference when a pair of magnetometers is available or in the case where only a single magnetic field sensor is present. Specifically, a statistical technique called Singular Spectrum Analysis (SSA) will be discussed. This tool enables the decomposition of magnetic field data into physically meaningful components without input from any physics-based models. Then, statistical analysis can be performed on the decomposed components to label them as interference or residual fields. Removing the interference terms allows for reconstruction of the magnetic field data with increased fidelity, without required hand-tuned filtering at each interval under observation. This presentation will provide several case studies from the Swarm-Echo and Parker Solar Probe missions.