Work

Imaging the Lower Ionosphere Using Network Measurements of VLF Transmitter Signals and a Particle Filter Algorithm

Year: 2025

Type: article

Abstract: The D-region ionosphere is critical to radio wave propagation, communication and navigation, but remotely sensing this region is challenging due to its height and low electron density. The technique using measurements of Very-low-frequency (VLF) signals remains one of the most reliable methods to probe the D-region, and the Kalman filter has been utilized to measure the D-region ionosphere over large area. However, the inversion of D-region ionosphere using VLF measurements is inherently nonlinear, and the noise is distributed in a non-Gaussian manner. To overcome this limitation, in this study we have proposed a technique to image the D-region ionosphere over large area using the Gaussian mixture extension of local particle filter (LPFGM) algorithm with a network of VLF receivers. On basis of numerical experiments, we have examined the performance of different filters. For typical daytime experiments, the error in ionospheric reference height (h') is found to be 1.44 km and 1.18 km for the local ensemble transform Kalman filter (LETKF) and LPFGM, respectively. The corresponding error in the ionospheric sharpness parameter (β) is 0.036 km-1and 0.028 km-1. For typical nighttime experiments, the error in h' is 1.39 km and 0.90 km for LETKF and LPFGM, respectively, with the corresponding error in β as 0.031 km-1 and 0.026 km-1. Therefore, the particle filter can improve largely the accuracy in the inversion problem of D-region ionosphere using VLF measurements, since it is particularly designed for nonlinear problems and the Gaussian mixture helps resolve the issue of sparse measurements. Moreover, our proposed LPFGM method converges notably faster than LETKF, and thus has the potential to remotely sense the D-region ionosphere in real time.

Source: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

DOI: 10.1109/jstars.2025.3645493