Article
Computer Science, Artificial Intelligence
Aleksa Damljanovic, Annachiara Ruospo, Ernesto Sanchez, Giovanni Squillero
Summary: This paper presents a technique for detecting hardware trojans using pre-silicon simulation and machine learning algorithms. The validity of the approach has been demonstrated on an industrial grade automotive benchmark suite, showing high accuracy.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Hardware & Architecture
Kerem Arikan, Alessandro Palumbo, Luca Cassano, Pedro Reviriego, Salvatore Pontarelli, Giuseppe Bianchi, Oguz Ergin, Marco Ottavi
Summary: The article presents a new approach based on count-min sketches for detecting microarchitectural attacks on microprocessors featured by embedded systems. The method is effective in detecting various attacks without introducing interference to the system.
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
(2022)
Article
Chemistry, Analytical
Cosmin-Andrei Popovici, Andrei Stan
Summary: This paper proposes a real-time RISC-V-based CAN-FD bus diagnosis tool, named RiscDiag CanFd, as an open-source alternative. This tool enables researchers and students to develop distributed embedded systems using low-cost and reliable setups. In addition to the extended RISC-V core, the proposed SoC provides UDP communication via Ethernet for connecting to a PC. Furthermore, a GUI application is developed for accessing and using the hardware solution deployed in an FPGA. The proposed solution is evaluated and shows a 0% frame loss rate, a timestamp error under 0.001%, and an acquisition cycle jitter under 10 ns.
Article
Computer Science, Hardware & Architecture
Bassam J. Mohd, Sa'ed Abed, Thaier Hayajneh, Mohammad H. Alshayeji
Summary: This article presents a low-power, low-energy and trusted design for lightweight ciphers in resource-constrained devices, leveraging a smart runtime monitoring algorithm to achieve hardware trojan detection and trust establishment.
IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING
(2021)
Article
Chemistry, Multidisciplinary
Zeyu Li, Junjie Wang, Zhao Huang, Nan Luo, Quan Wang
Summary: This study investigates the feasibility of using evolvable hardware (EHW) platforms to mitigate Hardware Trojans (HTs) in CGRA, proposing a method to autonomously reconfigure circuit structures based on an evolutionary algorithm (EA) to avoid using HT-infected (HT-I) processing elements (PEs). Experimental results in a Python simulation environment demonstrate the effectiveness of this approach for hardware platforms supporting coarse-grained reconfiguration and pure combinatorial circuits.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Marcello Barbirotta, Abdallah Cheikh, Antonio Mastrandrea, Francesco Menichelli, Marco Angioli, Saeid Jamili, Mauro Olivieri
Summary: High-performance embedded systems are driving the growth of the IoT through powerful processors, specialized hardware accelerators, and advanced software techniques. By combining hardware and software techniques, it is possible to design embedded architectures that can continue to function correctly even in the event of a failure or malfunction, thus increasing overall reliability and safety.
Article
Computer Science, Hardware & Architecture
Rana Elnaggar, Lorenzo Servadei, Shubham Mathur, Robert Wille, Wolfgang Ecker, Krishnendu Chakrabarty
Summary: Malware applications pose a significant threat to computing systems. Existing defense mechanisms are insufficient against new malware families. This study proposes the use of hardware performance counters (HPCs) and the XGBoost classifier to improve malware detection performance by at least 15%. Additionally, it demonstrates the early detection capability and low false positive rates of a multitemporal classification model. The XGBoost classifier also shows high resilience against adversarial attacks.
IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Shi Wei Yuan, Lei Li, Ji Yang, Yuanhang He, Wan Ting Zhou, Jin Li
Summary: This paper presented a novel method to detect Hardware Trojans attacks induced by General-Purpose Registers by comparing the states of GPRs with an embedded reference model in real time. The experimental results showed that all randomly inserted HT attacks can be detected in real time with a latency of two clock cycles.
IEICE ELECTRONICS EXPRESS
(2021)
Article
Computer Science, Hardware & Architecture
Lilas Alrahis, Satwik Patnaik, Muhammad Abdullah Hanif, Muhammad Shafique, Ozgur Sinanoglu
Summary: The article highlights the success of graph neural networks (GNNs) in detecting intellectual property (IP) piracy and hardware Trojans (HTs), while also addressing the susceptibility of GNNs to data poisoning attacks. The authors propose PoisonedGNN as the first backdoor attack on GNNs in the context of hardware design and demonstrate its effectiveness through experiments.
IEEE TRANSACTIONS ON COMPUTERS
(2023)
Article
Computer Science, Information Systems
Donghyun Kwon, Dongil Hwang, Yunheung Paek
Summary: The article discusses methods for securing the operating system kernel and introduces a new hardware-based monitoring platform, RiskiM. Through the hardware interface architecture PEMI, RiskiM can monitor the internal state of the host system to ensure the integrity of the kernel. Empirical validation shows that RiskiM successfully protects the host kernel with virtually no side effects.
Article
Computer Science, Hardware & Architecture
Luca Cassano, Mattia Iamundo, Tomas Antonio Lopez, Alessandro Nazzari, Giorgio Di Natale
Summary: Hardware Trojan Horses (HTHs) pose a serious threat to academia and industry due to their complexity and dangerousness. This paper proposes DETON, an automatic methodology for software manipulation to protect microprocessor-based systems against information stealing HTHs. DETON aims to produce an obfuscated version of the program to be executed on an untrusted CPU-based system, reducing the exposure of sensitive information to attackers. The DETON framework is compatible with the underlying target HW platform and incorporates guidelines and metrics for assessing software obfuscation against information-stealing HTHs.
JOURNAL OF SYSTEMS ARCHITECTURE
(2022)
Article
Computer Science, Hardware & Architecture
Vaibhav Verma, Tommy Tracy II, Mircea R. Stan
Summary: EXTREM-EDGE is a hardware/software co-design approach to add custom extensions to the RISC-V processor architecture for creating a scalable and flexible ML processor architecture. It satisfies the requirements of current AI/ML workloads and allows for quick adaptation to hardware and software changes.
SUSTAINABLE COMPUTING-INFORMATICS & SYSTEMS
(2022)
Article
Computer Science, Hardware & Architecture
Tiancheng Yang, Ankit Mittal, Yunsi Fei, Aatmesh Shrivastava
Summary: This article introduces a new type of delay-based analog Trojan circuits that can launch fabrication-time hardware attacks by interfacing with digital and analog macros. These Trojans have large delays, operate across multiple on-chip power domains, and are difficult to detect.
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
(2021)
Article
Computer Science, Theory & Methods
Subodha Charles, Prabhat Mishra
Summary: With the advancement in technology, the widespread use of Network-on-Chip (NoC) in multicore System-on-Chip (SoC) architectures has made it a potential target for security attacks. Therefore, a comprehensive investigation and discussion on security vulnerabilities and countermeasures for NoC are necessary.
ACM COMPUTING SURVEYS
(2021)
Article
Computer Science, Information Systems
Pietro Nannipieri, Stefano Di Matteo, Luca Zulberti, Francesco Albicocchi, Sergio Saponara, Luca Fanucci
Summary: Public-key cryptography is essential in digital infrastructures and faces a threat from quantum computers. Post-quantum cryptography aims to find algorithms that can resist attacks from quantum computers. This paper explores a hardware acceleration solution for lattice-based algorithms in order to improve performance and reduce energy consumption.
Article
Computer Science, Hardware & Architecture
Cristiana Bolchini, Giacomo Boracchi, Luca Cassano, Antonio Miele, Diego Stucchi
Summary: This article presents a novel fault-detection approach for image filtering, using an approximate computing strategy combined with image quality assessment metrics and machine learning models to assess the visual impact of faults in a lightweight manner. Experimental results demonstrate the effectiveness of this approach.
IEEE TRANSACTIONS ON COMPUTERS
(2022)
Article
Computer Science, Hardware & Architecture
Mariangela Rapisarda, Jose Alberto Hernandez, Alberto Gatto, Paola Parolari, Pierpaolo Boffi, Michela Svaluto Moreolo, Josep Maria Fabrega, Laia Nadal, Ricardo Martinez, Victor Lopez, Juan-Pedro Fernandez-Palacios, Gabriel Otero, David Larrabeiti
Summary: This paper proposes an alternative metropolitan area network architecture based on sliceable bandwidth/bitrate variable transceivers. The approach enables traffic aggregation from multiple edges and cost reduction through IP offloading. Simulation results demonstrate the viability and applicability of this technique in large metro networks.
JOURNAL OF OPTICAL COMMUNICATIONS AND NETWORKING
(2022)
Article
Computer Science, Hardware & Architecture
Luca Cassano, Mattia Iamundo, Tomas Antonio Lopez, Alessandro Nazzari, Giorgio Di Natale
Summary: Hardware Trojan Horses (HTHs) pose a serious threat to academia and industry due to their complexity and dangerousness. This paper proposes DETON, an automatic methodology for software manipulation to protect microprocessor-based systems against information stealing HTHs. DETON aims to produce an obfuscated version of the program to be executed on an untrusted CPU-based system, reducing the exposure of sensitive information to attackers. The DETON framework is compatible with the underlying target HW platform and incorporates guidelines and metrics for assessing software obfuscation against information-stealing HTHs.
JOURNAL OF SYSTEMS ARCHITECTURE
(2022)
Article
Engineering, Electrical & Electronic
Saurabh Khandelwal, Marco Ottavi, Eugenio Martinelli, Abusaleh Jabir
Summary: Based on memristive devices, we propose a novel gas sensor architecture that can predictably reduce the impact of process and parametric variations on sensing accuracy, while significantly improving energy efficiency and power consumption. Additionally, this architecture offers advantages in terms of multifunctional logic operations and resource reusability.
JOURNAL OF COMPUTATIONAL ELECTRONICS
(2022)
Article
Computer Science, Hardware & Architecture
Luca Cassano, Antonio Miele, Francesco Mione, Nicola Tonellotto, Carlo Vallati
Summary: This article proposes a novel system-level design approach for cyber-physical systems, considering the placement of processing, sensing, and actuation, as well as incorporating fault-tolerance mechanisms. Experimental results show that the approach reduces solution costs and ensures resource requirements and fault tolerance.
IEEE EMBEDDED SYSTEMS LETTERS
(2022)
Article
Computer Science, Hardware & Architecture
Kerem Arikan, Alessandro Palumbo, Luca Cassano, Pedro Reviriego, Salvatore Pontarelli, Giuseppe Bianchi, Oguz Ergin, Marco Ottavi
Summary: The article presents a new approach based on count-min sketches for detecting microarchitectural attacks on microprocessors featured by embedded systems. The method is effective in detecting various attacks without introducing interference to the system.
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
(2022)
Article
Computer Science, Hardware & Architecture
Zhenwei Dai, Anshumali Shrivastava, Pedro Reviriego, Jose Alberto Hernandez
Summary: The learned Bloom filter (LBF) improves the false positive rate (FPR) for a given memory budget by combining a machine learning model with a traditional Bloom filter. By treating the machine learning model as a variable and selecting the one with the lowest FPR, this approach achieves better performance compared to existing designs.
IEEE EMBEDDED SYSTEMS LETTERS
(2022)
Article
Engineering, Electrical & Electronic
Marcello Barbirotta, Abdallah Cheikh, Antonio Mastrandrea, Francesco Menichelli, Marco Ottavi, Mauro Olivieri
Summary: Functional safety is crucial in many applications involving microprocessors. Redundancy techniques like DMR and TMR are used to protect circuits against SEU faults. RISC-V is interesting for its extendability and open-source microarchitecture designs. In this work, a novel approach using IMT microprocessor architecture is presented, allowing for dynamic switching between DMR and TMR. Extensive RTL fault-injection simulations proved the effectiveness of this approach for a specific RISC-V core family.
JOURNAL OF LOW POWER ELECTRONICS AND APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
David Larrabeiti, Luis M. Contreras, Gabriel Otero, Jose Alberto Hernandez, Juan P. Fernandez-Palacios
Summary: 5G network slicing allows operators to deploy virtual connectivity services tailored for specific purposes on top of the same underlying physical infrastructure. For some 5G services, real-time information of the end-to-end Quality of Service for a particular slice is needed. This paper discusses the challenges of achieving this, presents monitoring techniques based on common open interfaces, and provides practical rules for planning end-to-end latency for slices. It also reviews latency engineering approaches for fronthaul traffic.
OPTICAL FIBER TECHNOLOGY
(2023)
Article
Computer Science, Hardware & Architecture
Marc Ruiz, Jose Alberto Hernandez, Marco Quagliotti, Emilio Hugues Salas, Emilio Riccardi, Albert Rafel, Luis Velasco, Oscar Gonzalez de Dios
Summary: The ITU-T Focus Group Network 2030 is assisting network operators in identifying network requirements for the short, medium, and long term. A traffic analysis methodology is introduced to estimate the expected traffic demand. Multi-band optical networks are expected to be implemented to meet the large bandwidth requirements.
JOURNAL OF OPTICAL COMMUNICATIONS AND NETWORKING
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
D. Larrabeiti, G. Otero, J. P. Fernandez-Palacios, L. M. Contreras, J. A. Hernandez
Summary: 5G presents significant challenges in latency management, particularly in fronthaul and backhaul traffic transport. Operators are seeking a unified architecture that incorporates virtualization, programmability, and performance control.
2022 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION (OFC)
(2022)
Article
Computer Science, Information Systems
Antonio Miele, Henry Zarate, Luca Cassano, Cristiana Bolchini, Jorge E. Ortiz
Summary: This article proposes a runtime resource management and provisioning middleware for the dynamic distribution of applications on processing resources in the Fog Computing paradigm. The proposed middleware consists of a two-level hierarchy: a global Fog Orchestrator monitoring the architecture status, and a Local Agent on each node performing fine-grain tuning of its resources. The co-operation between these components allows dynamic adaptation and exploitation of fine-grain nodes view for fulfilling system-level goals.
ACM TRANSACTIONS ON INTERNET OF THINGS
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Anu Bala, Saurabh Khandelwal, Abusaleh Jabir, Marco Ottavi
Summary: This paper evaluates the yield of a memristor-based crossbar array of artificial neural networks in the presence of stuck-at-faults (SAFs). A technique based on Markov chains is used to estimate the yield in the presence of stuck-at-faults. The Poisson distribution is also used for analysis and comparison, considering all repairable fault patterns. The results show that yield can be improved with redundancies and a higher repairable stuck-at-fault ratio.
2022 IEEE 28TH INTERNATIONAL SYMPOSIUM ON ON-LINE TESTING AND ROBUST SYSTEM DESIGN (IOLTS 2022)
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Alperen Bolat, Seyyid Hikmet Celik, Ataberk Olgun, Oguz Ergin, Marco Ottavi
Summary: This article introduces the software obfuscation framework ERIC, which uses physical unclonable functions and unique device identifiers as secret keys to encrypt software executables, protecting software security by preventing static and dynamic analysis.
2022 52ND ANNUAL IEEE/IFIP INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS (DSN 2022)
(2022)