Abstract
In this paper, sensing-assisted secure communication in a multi-user multi-eavesdropper integrated sensing and communication (ISAC) system is investigated. Confidential communication signals and dedicated sensing signals are jointly transmitted by a base station (BS) to simultaneously serve users and sense aerial eavesdroppers (AEs). A sum rate maximization problem is formulated under AEs' Signal-to-Interference-plus-Noise Ratio (SINR) and sensing Signal-to-Clutter-plus-Noise Ratio (SCNR) constraints. A fractional-programming-based alternating optimization algorithm is developed to solve this problem for fully digital arrays, where successive convex approximation (SCA) and semidefinite relaxation (SDR) are leveraged to handle non-convex constraints. Furthermore, the minimum number of dedicated sensing beams is analyzed via a worst-case rank bound, upon which the proposed beamforming design is further extended to the hybrid analog-digital (HAD) array architecture, where the unit-modulus constraint is addressed by manifold optimization. Simulation results demonstrate that only a small number of sensing beams are sufficient for both sensing and jamming AEs, and the proposed designs consistently outperform strong baselines while also revealing the communication-sensing trade-off.
Introduction
In the era of 5G and beyond, Unmanned Aerial Vehicles (UAVs) present a unique security challenge. While they offer flexible connectivity, they can also function as "aerial eavesdroppers" (AEs). Because these AEs are passive and non-cooperative, obtaining their Channel State Information (CSI)—a prerequisite for traditional physical layer security—is notoriously difficult.
Integrated Sensing and Communication (ISAC) offers a solution by allowing the base station (BS) to "see" eavesdroppers by analyzing radar echoes. However, beamforming energy toward an AE for sensing inadvertently increases the risk of information leakage. This research addresses this by designing beamforming strategies that maximize the legitimate communication rate while ensuring AEs are accurately sensed and effectively jammed using dedicated sensing signals that act as artificial noise (AN).
System Architectures and Optimization
The study investigates two primary hardware architectures: Fully Digital Arrays, where every antenna has a dedicated radio frequency (RF) chain, and Hybrid Analog-Digital (HAD) Arrays, which are more cost-effective but introduce complex optimization constraints.
- Problem Formulation: The objective is to maximize the sum rate for legitimate users while maintaining the AEs' SINR below a security threshold and the sensing SCNR above a required accuracy level.
- Fully Digital Design: The authors utilize Fractional Programming (FP) to transform the objective function, alongside Successive Convex Approximation (SCA) and Semidefinite Relaxation (SDR) to manage non-convex constraints.
- HAD Optimization: For the HAD architecture, Manifold Optimization is employed to address the "unit-modulus" constraint of phase shifters, treating the hardware limitations as a geometric problem on a complex circle manifold.
Theoretical Analysis: The Rank Bound
A key contribution of this work is the mathematical proof of a worst-case rank bound for the sensing covariance matrix. Contrary to previous assumptions that sensing requires a high-rank signal (many beams), this paper proves that the optimal sensing covariance matrix has a rank no higher than $\sqrt{2T + 1}$, where $T$ is the number of eavesdroppers. For instance, to handle 8 eavesdroppers, as few as 4 dedicated sensing beams may suffice, significantly reducing the hardware requirements for RF chains.
Key Insights and Results
- Dual-Purpose Waveforms: The dedicated sensing signal serves a dual role: providing the energy for radar echoes (sensing) while simultaneously acting as jamming AN to scramble eavesdropper reception.
- Communication-Sensing Trade-off: Simulations reveal a clear trade-off; sensing accuracy (SCNR) can be increased up to a certain point (approximately 20 dB in tests) without significantly impacting the communication rate. Beyond this "knee," power diversion to sensing causes the communication rate to drop sharply.
- Performance and Efficiency: The proposed designs consistently outperform "Communication-Only" beamforming, which lacks dedicated jamming power. Furthermore, the HAD design closely approaches the performance of the more expensive fully digital architecture, even with limited RF chains. The optimization algorithms demonstrate high efficiency, typically converging within 5 to 10 iterations.
Conclusion and Future Work
This research establishes that a small number of sensing beams are sufficient for both tracking and jamming aerial eavesdroppers in ISAC systems. While the current model assumes stable AE locations within a coherent time block, future work will explore high-mobility scenarios using Extended Kalman Filtering (EKF) and investigate multi-base station coordination to manage rogue UAVs across cellular networks.
Summary updated