A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments: Security and Privacy Perspectives
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Security Radio. Generated commentary on the latest security and cryptography papers.
Nadia: We're looking at the paper: "A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments".
Elias: Here's the gist of the paper. Detailed Research Summary: A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments: Security and Privacy Perspectives This survey provides a comprehensive, in-depth examination of Reconfigurable Intelligent Surface (RIS) applications within real-world systems, with a primary focus on the critical security and privacy threats, associated vulnerabilities, and corresponding defensive strategies relevant to practical deployment.
Nadia: The research systematically analyzes complex scenarios by considering two system types—those *with* legitimate RIS and those *without* it—and two attacker profiles: those utilizing legitimate RIS versus those deploying malicious RIS.
Elias: This comparative analysis effectively demonstrates how the introduction of RIS can introduce novel security risks, including eavesdropping, jamming, and spoofing attacks.
Nadia: In direct response to these threats, the survey meticulously reviews a spectrum of countermeasures, ranging from applying additional security algorithms and disrupting attackers to implementing early detection mechanisms for unauthorized RIS deployments.
Paper summary: Nadia: So we're looking at this paper 'A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments: Security and Privacy Perspectives'. It really lays out how Reconfigurable Intelligent Surfaces are changing wireless environments through dynamic manipulation of electromagnetic waves.
Elias: Exactly. The main point here is that while we've seen a lot of theoretical work on the benefits of RIS for communication and sensing, the real-world use cases in places like homes or factories are still very limited, especially when you think about security and privacy risks.
Priya: What kind of practical systems are they focusing on specifically? Are we talking about something specific, or is it a broad overview?
Nadia: It’s a comprehensive examination that looks at real-world systems across several areas like 6G networks, sensing, autonomous driving, and even integrated sensing and communication <ref:2512.15754#pg2>. The authors set out to look closely at the security and privacy threats that come with deploying RIS in these actual settings.
Elias: They structure their analysis by looking at two system types: those that have a legitimate RIS already in place versus systems where there is no RIS, and then they look at two attacker profiles: attackers using the legitimate RIS versus attackers who are deploying their own malicious ones.
Priya: That distinction between the systems and the attackers sounds crucial for understanding where these risks actually lie.
Nadia: It is. The survey demonstrates how introducing RIS can actually create brand new security risks, like eavesdropping, jamming, and spoofing attacks that we haven't fully considered before.
Elias: And in response to those threats, the paper reviews a wide range of countermeasures out there. They look at everything from applying additional security algorithms to disrupting attackers or setting up early detection mechanisms for unauthorized RIS deployments.
Priya: That’s interesting because it shows that defense isn't just about one thing; it’s a whole spectrum of strategies.
Nadia: Right, and they also look at scenarios where the legitimate user might actually deploy supplementary RIS elements as a way to defend against these kinds of adversarial actions.
Elias: The core contribution here is presenting this first comprehensive analysis that covers both the attack and defense sides for RIS across all these diverse practical systems.
Paper summary: Priya: It’s helpful because it systematically categorizes the different attack mechanisms, like eavesdropping or spoofing, and pairs them up with specific defense strategies.
Nadia: They clearly separate the inherent vulnerabilities of two distinct system archetypes: first, RIS-assisted systems where adversaries try to exploit the legitimate infrastructure, and second, non-RIS systems where attackers deploy their own malicious RIS.
Elias: On the technical side, they detail what those specific attacks actually look like. For instance, jamming can happen in a selective way by tuning an RIS to neutralize one sensor while keeping connectivity for others.
Priya: And eavesdropping sounds like a big privacy concern because RIS fundamentally changes how signals travel, opening up new routes that bypass physical obstacles for unauthorized interception.
Nadia: They also mention techniques like Metasurface-enabled Sideband Steering, or MeSS, which are used to facilitate more sophisticated signal interception in those scenarios.
Elias: Spoofing is another one they cover where adversaries introduce an unauthorized RIS to trick positioning systems or manipulate surface elements to generate false signals, leading directly to inaccurate location data or identity spoofing.
Priya: When you look at the defenses, they mention physical-layer and signal processing countermeasures that aim to maintain a good Signal-to-Interference-plus-Noise Ratio for the legitimate user while simultaneously trying to minimize the adversary's ratio.
Nadia: They also talk about protocol-layer hardening, which means securing that control channel so that only authorized users have access to it.
Elias: And then there’s this point about user-side defense, where a legitimate user proactively sets up an additional RIS specifically to defend against those attacks.
Priya: But I see they also acknowledge some major limitations in their analysis, which is important for us when we look at practical deployment.
Nadia: They point out that many of these attacks and defenses rely heavily on accurate Channel State Information, or CSI. Attackers need precise propagation characteristics between the transmitter, RIS, and target to tune things effectively.
Elias: And it’s not just attackers; the paper notes that robust defenses like secure beamforming also demand accurate CSI across multiple links: between the base station and the RIS, between the RIS and the user, and even between the RIS and an eavesdropper.
Paper summary: Priya: That reliance on perfect CSI is a big hurdle for real deployment because getting that kind of precise measurement is often impractical or too costly.
Nadia: Then there's computational complexity; setting up an RIS for malicious purposes is inherently a complex optimization problem. Countermeasures that require jointly optimizing the RIS phase shifts with other things, like transmit and receive beamforming and artificial noise, introduce a lot of computational burden.
Elias: So what does this mean for us in terms of practical deployment? It suggests that whatever we build needs to be efficient enough to handle these complex optimizations without needing perfect CSI from the start.
Priya: And that brings us toward their future directions, which is where they suggest the next steps for research.
Nadia: They suggest a real push toward CSI-free strategies, which means finding blind or semi-blind control methods that use environmental information and sensor inputs instead of relying on perfect measurements.
Elias: That theoretical analysis for performance bounds in those CSI-free settings still seems underdeveloped, which is a gap we need to fill.
Priya: They also stress the need for computational efficiency, pushing for hardware-friendly algorithms like model-free learning or neural approximators that can reconfigure an RIS almost instantly after it's deployed.
Nadia: I agree with that because near-instantaneous reconfiguration is key if you want to react quickly to a sudden security threat.
Elias: And they talk about scalability and robustness, focusing on multi-user and multi-adversary scenarios, especially when only statistical information about eavesdroppers is available.
Priya: They also look at integrating AI and large language models for context-aware configuration, or using decode-and-forward relays to make things more efficient.
Nadia: Finally, they point toward next-generation scenarios like 6G applications in immersive communication and massive communication security, and exploring low-cost RIS with robust attack modeling <ref:2512.15754#pg2>.
Elias: The paper concludes by suggesting that a key future defense direction is moving beyond just protocol-level security to incorporating environmental awareness, RF tomography, and anomaly detection methods.
Priya: So what this all means for the listener who’s just driving or walking? It means that as we build these wireless systems, we can't just rely on the hardware being set up correctly; we need smart systems that can actively sense and defend themselves against malicious surfaces.
Conclusion: Nadia: So we're wrapping up this deep dive into Reconfigurable Intelligent Surfaces, focusing specifically on security and privacy risks for real-world use cases in things like autonomous driving or smart factories.
Elias: Yeah, we've been looking at how these surfaces can be used to steer signals and see where they fit into the broader picture of wireless systems.
Priya: What we've seen is that RIS isn't just a communication tool; it’s an active element that can be exploited if you aren't careful about the security setup.
Nadia: Exactly, the authors really laid out this comparison between systems that already have RIS and those where attackers put their own malicious surfaces in play.
Elias: They systematically categorize the different threats, like eavesdropping or spoofing, and then map them directly to specific defense strategies available.
Priya: It’s interesting how they show that defenses aren't just one thing; you've got physical layer stuff, protocol hardening, and even user-side solutions where the legitimate user adds more RIS for defense.
Nadia: The authors make a clear point about the practical hurdles, especially how much these attacks and defenses depend on getting perfect channel state information or CSI to work properly.
Elias: That dependency on precise measurements is a big deal because it shows that designing robust defenses is tough when you don't have perfect knowledge of what’s happening in the air.
Priya: And they also flagged the computational load; setting up an RIS for bad things involves complex math that can be hard to run quickly in real-time.
Nadia: So, looking at the whole survey, it really boils down to how we move forward—we need better ways to handle those measurement gaps and make those configurations faster.
Elias: They point toward future work focusing on CSI-free strategies and using AI or machine learning for configuration because real-time is essential.
Priya: That moves the conversation from just 'can it be attacked?' to 'how do we build a system that can detect and adapt without knowing everything perfectly?'
Nadia: It really frames RIS security as a challenge in creating intelligent, adaptive environments rather than just a hardware problem.
Elias: And that leads us right into the next part of this research—the specific technical methods they suggest for achieving those CSI-free controls.
University of Hawaii at Manoa · University of South Florida
cs.CR, cs.NI, eess.SP
Submitted: 2025-12-12
Updated: 2026-11-03
Comments: 35 pages, 17 figures, 13 tables. Survey article on security and privacy in RIS-enabled systems
Project page: https://awesome-ris-security.github.io
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 92/100
The gist: A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments: Security and Privacy Perspectives This survey provides a comprehensive…
Key concepts
- Reconfigurable Intelligent Surface (RIS)
- A surface composed of many small elements that can intelligently change its electromagnetic properties. These elements act as programmable mirrors to steer or focus radio signals precisely where needed, allowing for flexible control over wireless communication paths.
- Jamming
- An attack where an adversary uses a RIS to intentionally disrupt a target signal. This can be done selectively by tuning the surface to block a specific sensor while keeping communication open for others, effectively creating localized interference.
- Eavesdropping
- The unauthorized interception of signals that have been redirected or reflected by an RIS. Because RIS alters the propagation path, attackers can use malicious surfaces to bypass physical barriers and secretly capture sensitive data transmissions.
- Channel State Information (CSI)
- Precise knowledge about the wireless channel—how a signal travels from a source to a receiver. This information is crucial for both effective attacks (to tune interference) and robust defenses (to design secure beamforming). However, obtaining accurate CSI is often difficult in practice.
Terminology
Summary
A Survey on Reconfigurable Intelligent Surfaces Towards Practical Deployments: Security and Privacy Perspectives
This survey provides a comprehensive, in-depth examination of Reconfigurable Intelligent Surface (RIS) applications within real-world systems, with a primary focus on the critical security and privacy threats, associated vulnerabilities, and corresponding defensive strategies relevant to practical deployment. The research systematically analyzes complex scenarios by considering two system types—those with legitimate RIS and those without it—and two attacker profiles: those utilizing legitimate RIS versus those deploying malicious RIS. This comparative analysis effectively demonstrates how the introduction of RIS can introduce novel security risks, including eavesdropping, jamming, and spoofing attacks. p. 0 This survey
provides a comprehensive examination of RIS applications in
real-world systems, with a focus on the security and privacy
threats, vulnerabilities, and defensive strategies relevant to practical use. We analyze scenarios with two types of systems (with
and without legitimate RIS) and two types of attackers (with and
without malicious RIS), and demonstrate how RIS may introduce new attacks to practical systems, including eavesdropping,
jamming, and spoofing attacks. We analyze scenarios with two types of systems (with
and without legitimate RIS) and two types of attackers (with and
without malicious RIS), and demonstrate how RIS may introduce new attacks to practical systems, including eavesdropping,
jamming, and spoofing attacks. In response, we review defenses
against RIS-related attacks in these systems, such as applying
additional security algorithms, disrupting attackers, and early
detection of unauthorized RIS.Where the paper says it
Where the paper says it
Core Contributions and Scope:
The main contributions of this survey are twofold:
-
Pioneering Comprehensive Analysis: It presents the first survey that offers a holistic analysis encompassing both the attack and defense facets associated with RIS across diverse practical systems, spanning areas such as 6G, sensing, localization, smart factories, autonomous driving, and Integrated Sensing and Communication (ISAC).
-
Systematic Categorization: The survey systematically reviews both RIS-assisted systems and systems under attack by RIS. It rigorously categorizes attack mechanisms (e. p. 2
g., eavesdropping, jamming, spoofing) alongside corresponding defense strategies (e. p. 2Where the paper says it
From a security and privacy perspective, an increasing number of surveys highlight the vulnerabilities and protective mechanisms introduced by RIS. These works examine security challenges in RIS-assisted communication, the use of RIS as attack-enabling surfaces, and RIS-based countermeasures against adversarial behavior.
g., additional security algorithms [9], attacker disruption [C], early detection of unauthorized RIS). p. 0Where the paper says it
• We systematically review both RIS-assisted systems and systems attacked by RIS, and categorize attack mechanisms (eavesdropping, jamming, spoofing, etc. ) and defense strategies (additional security algorithm, disruption of attackers, early detection of unauthorized RIS, etc.
Where the paper says it
We analyze scenarios with two types of systems (with and without legitimate RIS) and two types of attackers (with and without malicious RIS), and demonstrate how RIS may introduce new attacks to practical systems, including eavesdropping, jamming, and spoofing attacks. In response, we review defenses against RIS-related attacks in these systems, such as applying additional security algorithms, disrupting attackers, and early detection of unauthorized RIS.
The paper clearly delineates the inherent vulnerabilities of two distinct system archetypes: (1) RIS-assisted systems, where adversaries exploit the legitimate RIS infrastructure; and (2) Non-RIS systems, where attackers deploy their own malicious RIS. p. 2 This enables clearer comparison, classification, and identification of open problems. • We point out the vulnerability of two types of systems:
(1) RIS-assisted systems, where attackers exploit the
legitimate RIS; and (2) Non-RIS systems, where attackers
deploy their own malicious RIS.Where the paper says it
Technical Deep Dive: Fundamentals and Attacks:
The survey establishes a strong foundation by detailing various aspects of RIS technology, covering its fundamentals—distinguishing between passive versus active configurations—and its programmability as a tunable phase profile for beam steering or focusing in both near-field and far-field environments. Its application scope is broad, covering signal modalities like mmWave, sub-6G, and acoustic systems. p. 7 In the following part, we will elaborate
on RIS applications based on different signal modalities to
show the importance of the RIS in different practical systems. Table III shows the structure, control method, deployment, and
application scenarios of mmWave, sub-6G, and acoustic RIS.Where the paper says it
The survey meticulously details several specific types of RIS-related attacks:
-
Jamming: Attackers can execute selective jamming by dynamically tuning an RIS to neutralize a specific sensor while maintaining connectivity for neighboring devices.
-
Eavesdropping: RIS fundamentally alters propagation paths, creating new routes that bypass physical obstacles, thereby facilitating unauthorized signal interception. Attackers can leverage their own malicious RIS to intercept signals or employ sophisticated techniques like Metasurface-enabled Sideband Steering (MeSS). p. 9
Where the paper says it
Chen et al. introduce metasurface-enabled sideband steering (MeSS) attack in a SISO system with an RIS between Tx and Rx [96], as shown in Fig.
-
Spoofing: Adversaries can introduce an unauthorized RIS to mislead positioning systems or manipulate surface elements to generate false signals, leading directly to inaccurate location data or identity spoofing.
Defense Mechanisms and Limitations:
The survey reviews countermeasures across multiple layers:
-
Physical-Layer and Signal Processing Countermeasures: These leverage wireless propagation strategies, such as designing robust beamforming and artificial interference tailored to maintain the legitimate user's Signal-to-Interference-plus-Noise Ratio (SINR) while simultaneously minimizing the adversary’s SINR.
-
Protocol-Layer Hardening: This involves securing the RIS control channel to ensure that only authorized users possess access.
-
User-Side Defense: The survey also investigates scenarios where a legitimate user deploys an additional RIS specifically to defend against attacks, such as employing it for anti-eavesdropping or signal redirection away from threats.
However, the analysis is tempered by acknowledging significant limitations:
-
CSI Requirements: A major hurdle is the heavy reliance of many attacks and defenses on accurate Channel State Information (CSI). Attackers require precise propagation characteristics between transmitter, RIS, and target for effective tuning (e.g., for destructive interference or eavesdropping). p. 7
Similarly, robust defenses like secure beamforming demand accurate CSI across multiple links (BS–RIS, RIS–User, and RIS–Eavesdropper channels). p. 3Where the paper says it
JOURNAL OF LATEX CLASS FILES, VOL. 8, AUGUST 2021 8 Signal Modalities Structure Control method Deployment Application mmWave [63]– [68] Extremely dense arrays of sub-wavelength elements Electronic tuning; fastest switching On walls, objects, or infrastructures Coverage expansion, ISAC, V2X, high-precision localization sub-6G [25], [69]–[75] Less dense arrays, larger elements than mmWave RIS Electronic tuning; slower switching On walls, ceilings or UAVs Coverage expansion, RF sensing, backscatter, V2V, smart home, industrial, far-field IoT Acoustic [76]– [83] Arrays of acoustic resonators or cavitie, usually very large Mostly mechanical adjustments or active actuation; slow switching Underwater: on seabed or buoys In-room: on walls or ceilings Underwater networks, ultrasonic imaging, non-destructive industrial testing, etc.
Where the paper says it
On the contrary, active RISs not only reflect, but further amplify the reflected signals as well. To achieve this goal, an active Acronyms Definitions 5G Fifth Generation 6G Sixth Generation AI Artificial Intelligence AN Artificial Noise CSI Channel State Information DRL Deep Reinforcement Learning DoS Denial-of-Service EM Electromagnetic IoMT Internet of Medical Things IoT Internet of Things IRS Intelligent Reflecting Surface ISAC Integrated Sensing and Communication LoS Line-of-Sight LU Legitimate User MEC Mobile Edge Computing MIMO Multiple-Input Multiple-Output NLoS Non-Line-of-Sight PLS Physical Layer Security RIS Reconfigurable Intelligent Surface RF Radio Frequency SISO Single-Input Single-Output SINR Signal-to-Interference-plus-Noise Ratio SNR Signal-to-Noise Ratio UAV Unmanned Aerial Vehicles V2I Vehicle-to-Infrastructure V2X Vehicle-to-Everything V2V Vehicle-to-Vehicle VLC Visible Light Communication TABLE II: List of acronyms.
-
Computational Complexity: Configuring an RIS for malicious purposes is inherently a complex optimization problem. Countermeasures that require jointly optimizing RIS phase shifts with other parameters (like transmit/receive beamforming and artificial noise) introduce significant computational burdens. p. 2
Where the paper says it
2b, Path A explains defending against an attacker without an RIS by securing the RISassisted systems. For example, the defender uses an existing indoor RIS and jointly optimizes beamforming metrics and RIS parameters to prevent unauthorized sensing [36].
Future Directions and Roadmap:
The paper concludes by outlining critical future research trajectories to overcome these limitations, suggesting a roadmap for both academic inquiry and engineering practice:
-
CSI-Free Strategies: Future work must prioritize blind or semi-blind control strategies that utilize environmental priors, sensor inputs, and unsupervised learning. Theoretical analysis for performance bounds in CSI-free settings remains underdeveloped. p. 0
Where the paper says it
8, AUGUST 2021 1 A Survey on Reconfigurable Intelligent Surfaces in Practical Systems: Security and Privacy Perspectives Ziyu Chen, Yitong Shen, Jingzhe Zhang, Yao Zheng, Member, IEEE, Yili Ren, Member, IEEE, Xuyu Wang, Member, IEEE, Shiwen Mao, Fellow, IEEE, Hanqing Guo, Member, IEEE Abstract—Reconfigurable Intelligent Surfaces (RIS) have emerged as a transformative technology capable of reshaping wireless environments through dynamic manipulation of electromagnetic waves. While extensive research has explored their theoretical benefits for communication and sensing, practical deployments in smart environments such as homes, vehicles, and industrial settings remain limited and under-examined, particularly from security and privacy perspectives.
-
Computational Efficiency: There is an urgent need for hardware-friendly algorithms—such as model-free learning or neural approximators—that enable near-instantaneous reconfiguration post-deployment.
-
Scalability and Robustness: Research must address multi-user and multi-adversary scenarios, focusing on scalable, distributed designs, especially when only statistical CSI of eavesdroppers is available.
-
Advanced Integration: Future exploration should integrate AI/LLMs for context-aware configuration, utilize decode-and-forward relays to enhance efficiency, and explore using the channel for multiple tasks via ISAC systems.
-
Next-Generation Scenarios: Specific focus areas include 6G applications (Immersive Communication, Massive Communication security), incorporating movable antennas (liquid elements), developing low-cost RIS with robust attack modeling, and exploring frequency selectivity through auxiliary technologies or filtering RIS.
-
Anomaly Detection: A key future defense direction involves moving beyond protocol-level security to incorporate environment-awareness, RF tomography, and anomaly detection methods.
In essence, the survey serves as a vital practical reference for engineers aiming to secure emerging 6G and IoT infrastructures while providing a clear research roadmap for scholars tackling the unresolved challenges in RIS security. p. 0 To support future research, we also provide
a collection of open-source tools, datasets, demos, and papers
at: https://awesome-ris-security. By highlighting RIS’s
functionality and its security/privacy challenges and opportunities, this survey aims to guide researchers and engineers toward
the development of secure, resilient, and privacy-preserving RISenabled practical wireless systems and environments.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
AI-driven RIS configuration for adaptive jamming involves using Deep Q-Networks (DQN) to jointly optimize RIS phases and power to counter jamming attacks by
dynamically countering jamming and eavesdropping attacks
[135]. This allows an improved AI system to dynamically adjust its reflection pattern in real time based on observed channel conditions. -
AI-based security monitoring for RIS involves using Deep Reinforcement Learning (DRL) methods, such as the noisy dueling double deep Q-network (Noisy-D3QN), to
co-optimize RIS phases and power and maximizes the secure communication rate for LUs under jamming or eavesdropping attacks
[134]. This enables an AI system to proactively maximize secrecy capacity even when facing adversarial manipulation. -
An AI system can implement a randomized phase profile to create defense schemes by
dynamically adjusting the RIS phase profile to create a timevarying randomness or noise, thus spoofing or jamming the attacker
[18]. This allows an improved AI system to introduce temporal randomness into the channel observation to disrupt unauthorized sensing and eavesdropping. -
A system can use machine learning based anomaly detectors, such as those utilizing Deep Support Vector Data Description (Deep SVDD), to
detect RIS presence within fewer than 19 OFDM symbols across a wide frequency range
[39]. This enables an improved AI system to perform continuous environment scanning and detect unauthorized RIS without needing prior knowledge of the surface layout. -
An AI-driven secure system design can employ min-max optimization, where the system
jointly optimize the BS precoder, artificial noise covariance, and a friendly RIS
to maintain secrecy even in the presence of hidden malicious RIS [149]. This allows an improved AI system to co-design its beamforming and RIS parameters for robust communication against rogue surfaces. -
An AI-based defense mechanism can use spatial nulling by
tuning beamforming to place a null in the direction of the known malicious RIS
[150]. This enables an improved AI system to actively shape the propagation environment by suppressing signals directed toward a known adversarial RIS.
Abstract
Reconfigurable Intelligent Surfaces (RIS) have emerged as a transformative technology capable of reshaping wireless environments through dynamic manipulation of electromagnetic waves. Although RISs have been widely studied for communication, sensing, and localization, their security and privacy implications are still fragmented across application-specific studies, prototypes, and testbeds. This survey provides a deployment-oriented review of RIS applications, with a focus on the security and privacy threats, vulnerabilities, and defensive strategies relevant to practical use. We analyze scenarios with two types of systems (with and without legitimate RIS) and two types of attackers (with and without malicious RIS), and clarify how RIS introduces new attack surfaces to practical systems, including eavesdropping, jamming, and spoofing attacks. In response, we review defenses against RIS-related attacks in these systems, such as protocol- and algorithm-level protections, unauthorized RIS detection, and interference/disruption-based mitigation. We also discuss scenarios in which the legitimate user applies an additional RIS to defend against attacks. To support future research, we also curate papers and available open resources at https://awesome-ris-security.github.io/. By highlighting RIS's functionality and its security/privacy challenges and opportunities, this survey aims to guide researchers and engineers toward the development of secure, resilient, and privacy-preserving RIS-enabled practical wireless systems and environments.
Sources
- Hiding in Plain Sight: RIS-Aided Target Obfuscation in ISAC
- Exploring the Near and Far-Field Coexistence for RIS-Assisted ISAC Systems: An Adaptive Bandwidth Splitting Approach
- mmWave-Whisper: Phone Call Eavesdropping and Transcription Using Millimeter-Wave Radar
- Security Analysis of WiFi-based Sensing Systems: Threats from Perturbation Attacks
- Spatial-Domain Wireless Jamming with Reconfigurable Intelligent Surfaces
- Far- versus Near-Field RIS Modeling and Beam Design
- TRIS-HAR: Transmissive Reconfigurable Intelligent Surfaces-assisted Cognitive Wireless Human Activity Recognition Using State Space Models
- mmMirror: Device Free mmWave Indoor NLoS Localization Using Van-Atta-Array IRS
- Joint Optimization of Communication Enhancement and Location Privacy Protection in RIS-Assisted Underwater Communication System
- Reconfigurable Intelligent Surfaces in 6G Radio Localization: A Survey of Recent Developments, Opportunities, and Challenges
- RIS-Aided Positioning Under Adverse Conditions: Interference from Unauthorized RIS
- MALRIS: Malicious Hardware in RIS-Assisted Wireless Communications
- Sensing Safety Analysis for Vehicular Networks with Integrated Sensing and Communication (ISAC)
- SuperEar: Eavesdropping on Mobile Voice Calls via Stealthy Acoustic Metamaterials
- A Portable and Stealthy Inaudible Voice Attack Based on Acoustic Metamaterials
- Analysis and Detection of RIS-based Spoofing in Integrated Sensing and Communication (ISAC)
- Optimizing Indoor RIS-Aided Physical-Layer Security: A Codebook-Generation Methodology and Measurement-Based Analysis
- Decision Transformers for RIS-Assisted Systems with Diffusion Model-Based Channel Acquisition
- Secure Communication of UAV-mounted STAR-RIS under Phase Shift Errors
- Environment-Aware IRS Deployment via Channel Knowledge Map: Joint Sensing-Communications Coverage Optimization
Related papers
- From Strategy Search to Policy Learning: MCTS-Guided Adversarial Co-Evolution for LLM Safety Alignment
- Privacy Leakage from Recurrent State in LLM Inference
- Incrementally Verifiable Oracle Aided Computation
- Privilege and confidentiality in generative AI workflows
- BadMoE: Backdooring Mixture-of-Experts LLMs via Routing-Aware Triggers and Task-Underutilized Experts
- CoVault: Secure, Scalable Analytics of Personal Data