Adaptive LPD Radar Waveform Design with Generative Deep Learning

2024-03-18

IEEE Transactions on Radar Systems

AI Summary

Addressing the fundamental trade-off between covertness and sensing performance in low-probability-of-detection (LPD) radar waveform design, this paper proposes a novel waveform synthesis framework integrating adversarial generative learning with ambiguity-function-based constraints. Specifically, it introduces the first unsupervised generative adversarial network (GAN) tailored for radar waveforms, guided by a custom loss function explicitly designed to shape the ambiguity function—thereby enabling adaptive waveform generation that matches the ambient RF background distribution while jointly optimizing detection concealment and range resolution. Experimental results demonstrate that, compared to conventional LPD waveforms, the proposed method reduces single-pulse detectability by 90%, narrows the ambiguity function’s mainlobe width by 23%, and lowers peak sidelobe level by 18 dB. These improvements significantly enhance range resolution and interference resilience. The framework establishes a new paradigm for intelligent, environment-aware LPD radar waveform design.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Adversarial Learning & Robustness
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Security and Privacy: Large-scale security measurements

Abstract

We propose a novel, learning-based method for adaptively generating low probability of detection (LPD) radar waveforms that blend into their operating environment. Our waveforms are designed to follow a distribution that is indistinguishable from the ambient radio frequency (RF) background -- while still being effective at ranging and sensing. To do so, we use an unsupervised, adversarial learning framework; our generator network produces waveforms designed to confuse a critic network, which is optimized to differentiate generated waveforms from the background. To ensure our generated waveforms are still effective for sensing, we introduce and minimize an ambiguity function-based loss on the generated waveforms. We evaluate the performance of our method by comparing the single-pulse detectability of our generated waveforms with traditional LPD waveforms using a separately trained detection neural network. We find that our method can generate LPD waveforms that reduce detectability by up to 90% while simultaneously offering improved ambiguity function (sensing) characteristics. Our framework also provides a mechanism to trade-off detectability and sensing performance.

Problem

• Adaptive LPD radar waveform design
• Blending into RF background
• Trade-off detectability and sensing performance

Innovation

• Generative deep learning for LPD radar
• Unsupervised adversarial learning framework
• Ambiguity function-based loss minimization