@article{MAE4D51C5,
title = "Deep Model-Based Optimization of Jamming Effectiveness under Aircraft AESA Radar Operational Environments",
journal = "The Journal of Korean Institute of Communications and Information Sciences",
year = "2025",
issn = "1226-4717",
doi = "10.7840/kics.2025.50.11.1647",
author = "Hanseul Cho, Baekrok Shin, Chaewon Moon, Sang-Geun Hong, U-Ju Byoun, Jin-Yong Sung, Chulhee Yun",
keywords = "AESA Radar, Jamming Effectiveness, Modeling and Simulation, Model-Based Optimization, , Neural Network",
abstract = "We propose a deep learning algorithm to find effective jamming parameters in the aircraft AESA radar operational environment, based on a model-based optimization technique called RoMA. To represent a series of measurements obtained under the operational environment as a single number, we design jamming effectiveness by combining ranging failure rate and average range error. Next, we collect a jamming effectiveness dataset for various radar/jammer parameter combinations by repeatedly running the simulation. Our algorithm consists of two stages: the first is to pre-train a neural network that robustly approximates the function from radar/jammer parameters to jamming effectiveness; the second is to estimate the optimal jamming parameters by exploiting our model. As a result, the proposed method improved jamming effectiveness by an average of 41.2% and up to 80.3% compared to random search, and consistently outperformed other baseline models."
}