Parameter Estimation of Ionospheric Echo Signals Using an Electromagnetic Vector Array
Shujuan Ding, Jie Wang, Jinzhi Bi, Na Wei, Rongqiu Zheng, Junbo Yang · IEEE Access · 2026
With the use of the non-stationarity and waveform information of the ionospheric echo signals, an approach to the joint estimation of angles and polarization states is proposed by using an array of electromagnetic vector antennas. To address the issues of time-varying covariance matrix and overlapping time-frequency distributions of signals, the two key steps of the approach are 1) beam perturbation; and 2) identification of the non-stationary time-frequency signal subspace without eigen-decomposition. The first step is for the suppression of the cross time-frequency energy terms. The second step can be viewed as a non-stationary extension of the traditional stationary multi-stage Wiener filtering scheme. Simulation results show that the proposed approach is superior to the existing ones in computation efficiency and estimation accuracy for processing multiple ionospheric echo signals with overlapping time-frequency distributions. Without eigen-decomposition, the cost reduce for subspace construction is nearly $O(L_{0}^{3} -ML_{0}^{2})$ , where $L_{0} $ is the dimension of the time-frequency matrix used for parameter estimation, and $M$ is the number of echo signals. Under the tested scenario (the signal to noise ratio is between 0dB and 20dB, and the number of snapshots is 800), the highest improvement in the root mean square error of angle estimation is approximately 0.2 degree.