Explainable AI-Driven Radar Tracking: Clutter Filtering and Feature Analysis with LSTMS

In this work, we introduce an approach to enhance radar target tracking by filtering clutter using Long Short-Term Memory (LSTM) networks, augmented with explainable AI techniques. Our approach combines LSTM networks with SHAP (SHapley Additive exPlanations) and Permutation Importance to provide insights into feature contributions while maintaining high performance. The model achieved 98.42% accuracy on the test set, with SHAP and Permutation analyses revealing radial velocity and range as the most critical features for clutter filtering. This integration of LSTM and explainable AI techniques provides a robust, interpretable framework for radar signal processing, with potential applications across various signal processing domains.