Robust Secure ISAC Beamforming for Integrated HAPS-LEO Maritime Networks

Maritime integrated sensing and communication (ISAC) systems face dual challenges: achieving precise target estimation under heavy sea clutter while preventing eavesdropping in open wireless channels. This paper addresses secure beamforming design for high-altitude platform stations (HAPS) and low Earth orbit (LEO) satellites serving maritime regions. We formulate a joint optimization problem that balances sensing accuracy, measured by the Bayesian Cramér-Rao bound (BCRB) under spherically invariant random process (SIRP) clutter, against worst-case secrecy rate (SR) over an eavesdropper angular sector. We optimize this sensing-secrecy trade-off by jointly designing the beamforming vector, Bob-null artificial noise (AN) covariance, and pulse compression filter via an alternating optimization algorithm with provable convergence guarantees. For LEO scenarios, we extend the framework through per-slot parameterization that absorbs Doppler-induced inter-carrier interference (ICI) into effective noise terms, enabling warm-started optimization across orbital passes without altering the core algorithm. Simulations demonstrate smooth trade-offs among sensing and secrecy, robustness to sea state variations, and stable performance across LEO time slots under maritime conditions.