Machine Learning–Driven Prediction and Detection of 5G-Induced Nonlinear Interference in Radio Altimeters

View at publisher Summarise with VIRAT
Authors: Gurram Sunitha, Sivakumar Depuru, Sontimalli Sreeja, T Hethan Krishna, Urri Surendra Kumar, Varakuti Yashwanth Kumar
Cited by: 0
References: 16

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Systems coexisting with mid-band 5G with aeronautical radio altimeters (RAs) have brought about a substantial safety issue by adjacent-band electromagnetic interference that has the potential to cause nonlinear distortion, compress gain on the IF, and cause errors in altitude measurements. These mechanisms are found to be described by conventional analytical and simulation-based studies, which do not have the ability to predict the severity of interference quantitatively, nor do they have the ability to detect the degradation of performance in real time. In this paper, a machine learning-based scheme will be suggested to detect and forecast the effect of$\mathbf{5 G}$on RAs due to interference. Classification and regression models were built using the features obtained by using the IF spectral responses and experimental injection measurements that were used to estimate the state of interference, height-error ratio, and the onset of nonlinear saturation. The offered classifier was found to be 98.4 percent accurate, and the regression model was able to predict height-error ratios with a ratio error rate of 0.37. The model found the critical interference threshold (e.g., −15 dBm at 4000 MHz) accurately, and the sensitivity of predicting the NCD conditions was 95.1 %. The outcomes prove that ML is an effective tool in improving the assessment of RA interference and assurance of safety.

Fields of Study & Subjects

Radar Systems and Signal Processing Electromagnetic Compatibility and Noise Suppression Air Traffic Management and Optimization

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
696–702
Publication date
21 January 2026
Language
EN