Srdjan Capkun

Last Name

Capkun

First Name

Srdjan

ORCID

0000-0001-5727-3673

Organisational unit

03755 - Capkun, Srdan / Capkun, Srdan

Show all metadata

Search Results

Publications1 - 10 of 51

Confidential Computing with Heterogeneous Devices at Cloud-Scale
Dhar, Aritra; Sridhara, Supraja; Shinde, Shweta; et al. (2024)
2024 Annual Computer Security Applications Conference (ACSAC)
Cloud-centric workloads increasingly leverage domain-specific accelerators (DSAs) such as GPU, NPU, FPGA, etc., to achieve massive speedup over general-purpose CPUs. These workloads compute sensitive data; furthermore, the programs can be proprietary business secrets such as high-performance AI models. Therefore, several confidential cloud solutions have recently emerged to protect against the attacker-controlled software stack (OS/VMM) and the cloud service providers or CSPs themselves. CPU-centric trusted execution environments, or TEEs, have been around for decades and are deployed commercially. However, despite some recent proposals, most nodes lack TEE capability and, therefore, are unprotected against malicious CSP and software stack. We address this gap by proposing a new dedicated hardware module, the security controller (SC), that acts as the TEE proxy for the legacy non-TEE DSA nodes in a data center across racks. SC enforces access control and attestation mechanisms and protects the non-TEE nodes even from a physical attacker. This way, SC enables new-generation TEE-enabled nodes and legacy non-TEE nodes to be used in a data center simultaneously while ensuring security. We implement and synthesize SC hardware and evaluate it with real-world cloud-centric workloads with heterogeneous DSAs. Our evaluation shows that, on average, SC introduces 1.5-5% overhead while running AI, Redis, and file system workloads and scales well with an increasing number of DSA nodes (up to 2236 concurrent NPUs running CNNs).
Privacy-preserving outsourcing of brute-force key searches
Karame, Ghassan O.; Capkun, Srdjan; Maurer, Ueli (2010)
D-INFK Technical Report
In this work, we investigate the privacy-preserving properties of encryption algorithms in the special case where encrypted data might be brute-force decrypted in a distributed setting. For that purpose, we consider a problem where a supervisor holds a ciphertext and wants to search for the corresponding key assisted by a set of helper nodes, without the nodes learning any information about the plaintext or the decryption key. We call this a privacy-preserving cryptographic key search. We provide a model for privacy-preserving cryptographic searches and we introduce two types of privacy-preserving key search problems: plaintext-hiding and key-hiding cryptographic search. We show that a number of private-key and public-key encryp- tion schemes enable the construction of efficient privacy- preserving solvers for plaintext hiding searches. We also discuss possible constructions of privacy-preserving solvers for key-hiding cryptographic searches. Our results highlight the need to consider the property of enabling efficient privacy-preserving solvers as an additional criterion for choosing which cryptographic algorithm to use.
Physical-layer Identification: Secure or not?
Danev, Boris; Spindler, Alexandre de; Luecken, Heinrich; et al. (2009)
D-INFK Technical Report
Physical-layer identification of wireless devices, commonly referred to as Radio Frequency (RF) fingerprinting, consists of identifying a device based on imperfections exhibited by its radio transceiver during transmission. Such identifica tion can enhance access control protection, prevent device cloning, and/or complement message authentication proto cols. In this paper, we investigate the feasibility of per forming impersonation attacks by reproducing device finger prints of two most prominent physical-layer identification techniques, namely modulation-based and transient-based. We show that modulation-based fingerprints are vulnerable to impersonation and replay attacks. We also show that tran sient-based fingerprints are also reproducible, but more dif ficult to impersonate due to wireless channel effects. We validate our proposed attacks by extensive measurements as well as simulations on the measured data. Finally, we dis cuss the implications of our findings.
Empowering Data Centers for Next Generation Trusted Computing
Dhar, Aritra; Sridhara, Supraja; Shinde, Shweta; et al. (2022)
arXiv
Modern data centers have grown beyond CPU nodes to provide domain-specific accelerators such as GPUs and FPGAs to their customers. From a security standpoint, cloud customers want to protect their data. They are willing to pay additional costs for trusted execution environments such as enclaves provided by Intel SGX and AMD SEV. Unfortunately, the customers have to make a critical choice -- either use domain-specific accelerators for speed or use CPU-based confidential computing solutions. To bridge this gap, we aim to enable data-center scale confidential computing that expands across CPUs and accelerators. We argue that having wide-scale TEE-support for accelerators presents a technically easier solution, but is far away from being a reality. Instead, our hybrid design provides enclaved execution guarantees for computation distributed over multiple CPU nodes and devices with/without TEE support. Our solution scales gracefully in two dimensions -- it can handle a large number of heterogeneous nodes and it can accommodate TEE-enabled devices as and when they are available in the future. We observe marginal overheads of 0.42--8% on real-world AI data center workloads that are independent of the number of nodes in the data center. We add custom TEE support to two accelerators (AI and storage) and integrate it into our solution, thus demonstrating that it can cater to future TEE devices.
EdgeTDC: On the Security of Time Difference of Arrival Measurements in CAN Bus Systems
Roeschlin, Marc; Camurati, Giovanni; Brunner, Pascal; et al. (2023)
Proceedings 2023 Network and Distributed System Security Symposium
A Controller Area Network (CAN bus) is a message-based protocol for intra-vehicle communication designed mainly with robustness and safety in mind. In real-world deployments, CAN bus does not offer common security features such as message authentication. Due to the fact that automotive suppliers need to guarantee interoperability, most manufacturers rely on a decade-old standard (ISO 11898) and changing the format by introducing MACs is impractical. Research has therefore suggested to address this lack of authentication with CAN bus Intrusion Detection Systems (IDSs) that augment the bus with separate modules. IDSs attribute messages to the respective sender by measuring physical-layer features of the transmitted frame. Those features are based on timings, voltage levels, transients—and, as of recently, Time Difference of Arrival (TDoA) measurements. In this work, we show that TDoA-based approaches presented in prior art are vulnerable to novel spoofing and poisoning attacks. We describe how those proposals can be fixed and present our own method called EdgeTDC. Unlike existing methods, EdgeTDC does not rely on Analog-to-digital converters (ADCs) with high sampling rate and high dynamic range to capture the signals at sample level granularity. Our method uses time-to-digital converters (TDCs) to detect the edges and measure their timings. Despite being inexpensive to implement, TDCs offer low latency, high location precision and the ability to measure every single edge (rising and falling) in a frame. Measuring each edge makes analog sampling redundant and allows the calculation of statistics that can even detect tampering with parts of a message. Through extensive experimentation, we show that EdgeTDC can successfully thwart masquerading attacks in the CAN system of modern vehicles.
Modeling and Verifying Physical Properties of Security Protocols for Wireless Networks
Schaller, Patrick; Schmidt, Benedikt; Basin, David; et al. (2009)
2009 22nd IEEE Computer Security Foundations Symposium
We present a formal model for modeling and reasoning about security protocols. Our model extends standard, inductive, trace-based, symbolic approaches with a formalization of physical properties of the environment, namely communication, location, and time. In particular, communication is subject to physical constraints, for example, message transmission takes time determined by the communication medium used and the distance traveled. All agents, including intruders, are subject to these constraints and this results in a distributed intruder with restricted, but more realistic, communication capabilities than those of the standard Dolev-Yao intruder. We have formalized our model in Isabelle/HOL and used it to verify protocols for authenticated ranging, distance bounding, and broadcast authentication based on delayed key disclosure.
AdaptOver: Adaptive Overshadowing Attacks in Cellular Networks
Erni, Simon; Kotuliak, Martin; Leu, Patrick; et al. (2021)
arXiv
In cellular networks, attacks on the communication link between a mobile device and the core network significantly impact privacy and availability. Up until now, fake base stations have been required to execute such attacks. Since they require a continuously high output power to attract victims, they are limited in range and can be easily detected both by operators and dedicated apps on users' smartphones. This paper introduces AdaptOver -- a MITM attack system designed for cellular networks, specifically for LTE and 5G-NSA. AdaptOver allows an adversary to decode, overshadow (replace) and inject arbitrary messages over the air in either direction between the network and the mobile device. Using overshadowing, AdaptOver can cause a persistent ($\geq$ 12h) DoS or a privacy leak by triggering a UE to transmit its persistent identifier (IMSI) in plain text. These attacks can be launched against all users within a cell or specifically target a victim based on its phone number. We implement AdaptOver using a software-defined radio and a low-cost amplification setup. We demonstrate the effects and practicality of the attacks on a live operational LTE and 5G-NSA network with a wide range of smartphones. Our experiments show that AdaptOver can launch an attack on a victim more than 3.8km away from the attacker. Given its practicability and efficiency, AdaptOver shows that existing countermeasures that are focused on fake base stations are no longer sufficient, marking a paradigm shift for designing security mechanisms in cellular networks.
LEO-Range: Physical Layer Design for Secure Ranging with Low Earth Orbiting Satellites
Coppola, Daniele; Mumtaz, Arslan; Camurati, Giovanni; et al. (2025)
Proceedings of the 34th USENIX Security Symposium
We propose LEO-Range, a novel physical layer design for secure ranging between Low Earth Orbiting ( LEO) satellites and devices. LEO-Range 1) is compatible with Orthogonal Frequency Division Multiplexing (OFDM ) modulation scheme which is widely used by high-bandwidth satellite communications, 2) it provides accurate distance measurements (within the limits imposed by the available bandwidth) , and 3) it is provably secure and reliable across a range of common satellite channels. The design is based on a novel verification scheme in the frequency domain. We provide a security proof that bounds the probability of a distance-reduction attack for arbitrary physical layer attack strategies. We implement a prototype of LEO-Range and we test it with a hardware satellite channel emulator. In common line of sight 3GPP channels with SNRs between 8.8dB and 12dB (worst case scenario at low elevation) in a single ranging from a single satellite the adversary has a probability of less than 2−20 to successfully reduce the LEO-Range measured distance by more than 117 meters. These results already significantly limit spoofing, which typically can be done even across continents, and if distances are measured consecutively and from different satellites or ground stations, the overall distance and location spoofing will be even further limited, pointing to the practical viability of LEO-Range.
FAST: Fast and Accurate Security Testing of HRP UWB Chips
Aad, Graciana; Camurati, Giovanni; Dell’Amico, Matteo; et al. (2024)
High-Rate Pulse (HRP) Ultra-Wide Band (UWB) technology is used for secure distance measurement and was standardized by IEEE 802.15.4z in 2020. This standard is currently implemented in chips deployed in consumer devices such as Apple iPhones and Samsung smartphones. However, due to the use of proprietary algorithms and closed implementation, evaluating the security of such chips analytically is challenging. In this work, we therefore investigate how to evaluate the security of HRP UWB chips empirically. We propose FAST, a generic and efficient testing methodology that we use to accurately characterize the security of HRP UWB chips against distance reduction attacks. FAST relies on importance sampling and can accurately estimate very low success rates, for example, much smaller than 2^-10, using a small and practical number of tests. Using FAST, we characterize the security of a Qorvo DWM3000EVB chip across different settings and attack conditions. FAST revealed that different chip configurations affect the success rates (2^-10 to 2^-128) and that a well-designed attack signal can bypass the additional consistency checks offered by the chip.
Secure Neighborhood Discovery: A Fundamental Element for Mobile Ad Hoc Networking
Papadimitratos, Panos; Poturalski, Marcin; Schaller, Patrick; et al. (2008)
IEEE Communications Magazine
Pervasive computing systems will likely be deployed in the near future, with the proliferation of wireless devices and the emergence of ad hoc networking as key enablers. Coping with mobility and the volatility of wireless communications in such systems is critical. Neighborhood discovery (ND) - the discovery of devices directly reachable for communication or in physical proximity - becomes a fundamental requirement and building block for various applications. However, the very nature of wireless mobile networks makes it easy to abuse ND and thereby compromise the overlying protocols and applications. Thus, providing methods to mitigate this vulnerability and secure ND is crucial. In this article we focus on this problem and provide definitions of neighborhood types and ND protocol properties, as well as a broad classification of attacks. Our ND literature survey reveals that securing ND is indeed a difficult and largely open problem. Moreover, given the severity of the problem, we advocate the need to formally model neighborhoods and analyze ND schemes.

Publications1 - 10 of 51

Srdjan Capkun

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results