[Paper Review] Sensing-Assisted Eavesdropper Estimation: An ISAC Breakthrough in Physical Layer Security

Nanchi Su, Fan Liu+1 more|arXiv (Cornell University)|Oct 15, 2022

Radar Systems and Signal Processing4 citations

TL;DR

This paper proposes a sensing-assisted physical layer security scheme for Integrated Sensing and Communication (ISAC) systems, where the base station uses ISAC's sensing capability to estimate eavesdropper (Eve) directions via the CAML technique and jointly optimizes artificial noise and beamforming to maximize secrecy rate while minimizing estimation error (Cramér-Rao Bound). The approach enables mutual enhancement between sensing accuracy and security performance, achieving improved secrecy rates even with imperfect Eve channel knowledge.

ABSTRACT

In this paper, we investigate the sensing-aided physical layer security (PLS) towards Integrated Sensing and Communication (ISAC) systems. A well-known limitation of PLS is the need to have information about potential eavesdroppers (Eves). The sensing functionality of ISAC offers an enabling role here, by estimating the directions of potential Eves to inform PLS. In our approach, the ISAC base station (BS) firstly emits an omni-directional waveform to search for potential Eves' directions by employing the combined Capon and approximate maximum likelihood (CAML) technique. Using the resulting information about potential Eves, we formulate secrecy rate expressions, that are a function of the Eves' estimation accuracy. We then formulate a weighted optimization problem to simultaneously maximize the secrecy rate and minimize the CRB with the aid of the artificial noise (AN), and minimize the CRB of targets'/Eves' estimation. By taking the possible estimation errors into account, we enforce a beampattern constraint with a wide main beam covering all possible directions of Eves. This implicates that security needs to be enforced in all these directions. By improving estimation accuracy, the sensing and security functionalities provide mutual benefits, resulting in improvement of the mutual performances with every iteration of the optimization, until convergence. Our results avail of these mutual benefits and reveal the usefulness of sensing as an enabler for practical PLS.

Motivation & Objective

• To address the critical challenge in physical layer security (PLS) where prior knowledge of eavesdropper (Eve) locations is typically required but often unavailable in practice.
• To leverage the sensing capability of ISAC systems to estimate potential Eve directions and amplitudes without prior knowledge, enabling proactive security design.
• To jointly optimize artificial noise and beamforming to maximize secrecy rate while minimizing the Cramér-Rao Bound (CRB) of Eve estimation, under a total power budget constraint.
• To establish a feedback loop between sensing and security functionalities, where improved estimation accuracy enhances secrecy performance and vice versa.
• To demonstrate the mutual benefit between sensing and security in ISAC systems through a weighted optimization framework that balances both objectives.

Proposed method

• The ISAC base station first transmits an omni-directional waveform to detect potential eavesdroppers using the combined Capon and approximate maximum likelihood (CAML) technique for direction and amplitude estimation.
• A weighted optimization problem is formulated to simultaneously maximize secrecy rate and minimize the Cramér-Rao Bound (CRB) of Eve estimation, with constraints on total transmit power and a wide main beam covering all possible Eve directions.
• The beam pattern is constrained with a wide main lobe to ensure security is enforced in all potential eavesdropping directions, accounting for estimation uncertainty.
• Artificial noise (AN) is designed to suppress Eve’s signal reception while maintaining quality of service for legitimate users, with the AN covariance matrix optimized via the weighted problem.
• The optimization is iteratively solved, with each iteration improving both estimation accuracy (lower CRB) and secrecy rate, until convergence is achieved.
• The approach explicitly accounts for channel uncertainty and uses a Rician fading model with strong LoS components to reflect realistic mmWave ISAC environments.

Experimental results

Research questions

• RQ1 Can ISAC's sensing capability be leveraged to estimate eavesdropper directions without prior knowledge, enabling proactive physical layer security?
• RQ2 How does the trade-off between sensing accuracy (measured by CRB) and secrecy rate manifest under power constraints in ISAC systems?
• RQ3 What is the impact of Eve location uncertainty (angular interval Δθ) on the secrecy rate and estimation accuracy in a multi-user ISAC setup?
• RQ4 How does the angular separation between the legitimate user and the eavesdropper affect the performance trade-off between secrecy rate and estimation accuracy?
• RQ5 To what extent can joint optimization of artificial noise and beamforming improve both sensing and security performance in ISAC systems?

Key findings

• The proposed algorithm outperforms benchmark methods in secrecy rate even when no prior information about Eve channels is available, with performance gains increasing under higher power budgets.
• Secrecy rate initially increases with angular uncertainty (Δθ) due to reduced eavesdropping SNR, but eventually decreases as the power budget becomes insufficient to secure a wider angular region.
• At low power budgets (e.g., 25 dBm), the optimization becomes infeasible when angular uncertainty exceeds 5 degrees, highlighting the critical role of power and estimation accuracy.
• When the number of legitimate users exceeds five under a 25 dBm power budget, the secrecy rate cannot be guaranteed, indicating system scalability limits.
• As the secrecy rate increases, the CRB of Eve angle estimation also increases, confirming the inherent trade-off between security and sensing accuracy.
• When the legitimate user and Eve are at the same angle (-20°), the system achieves the lowest CRB after four iterations, demonstrating convergence and robustness in challenging scenarios.