Project P2.20 is a collaboration between DEWC Systems and the Institute for Intelligent Systems Research and Innovation (IISRII), Deakin University. The main project objective is to design and develop
intelligent data-based models for classification of radar signals with potential distributed radio frequency (RF) processing capabilities for deployment across space-based platforms. The project leverages advancements in artificial intelligence (AI) to develop cutting edge intelligent data analytic solutions for radar monitoring capabilities. This initial collaboration is part of a multi-phase AI driven RF monitoring system roadmap of DEWC systems to advance sovereign RF technologies and capabilities in Australia, along with the support of SmartSat CRC. The collaboration team brings together researchers from Deakin University and domain experts from DEWC Systems to analyse and address radar signal analytics, leading to detection and classification of conventional and low-probability-of-intercept (LPI) radar signals.

The eight-month duration of the project has resulted in a proof-of-concept deep learning-based framework capable of detecting and classifying synthetic RF signals tested in a lab environment with simulated noise. A variety of data sets with different signal-to-noise ratios (SNRs) from different target radar signals have been generated for evaluation. The obtained accuracy scores across different scenarios are promising, yielding close to perfect classification performance and demonstrating efficacy of the developed DL framework for RF signal detection and classification.