New Denoising Method based on the Contribution Value Parameters of the Medium Frequency Radar Antenna

The wind field of the Mesosphere and Lower Thermosphere (MLT) is an important parameter for studying the dynamics of the middle and upper atmosphere. Medium Frequency Radar (MF Radar) is one of the important techniques for observing the atmospheric wind field in the MLT region. Full Correlation Analysis (FCA) is a commonly used inversion algorithm for MF radar wind fields, but it is highly sensitive to noise. Effective noise processing can significantly enhance the accuracy of the FCA method in MF radar wind field inversion. Currently, the polynomial fitting denoising method widely used in MF radar exhibits inconsistent performance in different noise environments, leading to a reduction in the amount of valid wind field data that MF radar can observe, which restricts its application prospects for wind field observation in the MLT region. To address this challenge, this study introduces the antenna contribution parameter into MF radar for the first time and proposes the MF-AH algorithm, which uses the antenna contribution parameter as the evaluation criterion to effectively reduce noise in the correlation functions generated from MF radar signals. Compared with the conventional polynomial fitting method, the MF-AH algorithm, validated by both simulation and experimental data, reduces zonal and meridional wind speed errors by approximately 20% under low Signal-to-Noise Ratio (SNR) conditions. Moreover, the algorithm eliminates reliance on noise as the primary evaluation metric, significantly enhancing the effectiveness and richness of wind field data.