The semiconductor industry has adopted heterogeneous integration (HI), incorporating modular intellectual property (IP) blocks (chiplets) into a unified system-in-package (SiP) to overcome the slowdown in Moore’s Law and Dennard scaling and to respond to the increasing demand for advanced integrated circuits (ICs). Despite the manifold benefits of HI, such as enhanced performance, reduced area overhead, and improved yield, this transformation has also led to security vulnerabilities in the SiP supply chain and in-field operations, ranging from chiplet piracy and SiP reverse engineering (RE) to information leakage. Although conventional countermeasures provide the desired robustness for monolithic ICs, they are insufficient for addressing these challenges in the context of HI. To address these concerns, this paper presents a novel root-of-trust architecture, augmenting the process of integration using a centralized chiplet hardware security module (CHSM), aiming to provide comprehensive and robust protection throughout the SiP supply chain and in-field operations. Also, the proposed architecture equipped with the CHSM effectively addresses potential security breaches while providing robust protection against zero-day attacks through its reconfigurable capabilities. Throughout five detailed case studies, this paper performs a comprehensive security analysis to illustrate the resilience of CHSM against contemporary attack scenarios in the HI domain.
5 December 2023 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
The Internet-of-Thing (IoT) era inspires a surge of networked embedded devices in the real world. However, cyber-attacks such as malware intrusions pose severe concerns about the security of the entire IoT space by hijacking the devices, altering the application’s execution, and/or causing a denial of services. Traditional operating-system-level and built-in hardware detection solutions either induce drastic performance degradation or incur high overhead to the system, failing to provide protection in a timely and affordable fashion. On the other hand, external side-channel-based security monitoring becomes an attractive scheme for addressing the dilemma because the observable physical properties of a running electronic device, such as power consumption and electromagnetic (EM) emanations, can deliver a good amount of information for the underlying operations. In this article, we present TrustGuard, a standalone security monitoring framework integrating both power side-channel capturing and machine learning (ML)-based malware detection capabilities on the same field-programmable gate array (FPGA) fabric. We eliminate the need for dedicated sampling equipment like an oscilloscope with an ON-chip configurable sensor FPGA analog-to-digital converter (ADC) and enable agile prototyping of ML accelerator for attack detection through the high-level synthesis (HLS) technique. We deploy the outcome hardware sensor and security monitor on the Xilinx ZCU104 platform to target the prevalent BeagleBone Black (BBB) board by profiling the behaviors of uncompromised benchmark applications and discovering the anomalies introduced by the attack vectors, including malware infections, code injection, and code reuse. The experimental results demonstrate the performance and effectiveness of TrustGuard by achieving more than 90% malware detection accuracy.
16 October 2023 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
As Moore’s law comes to a crawl, advanced package and integration techniques become increasingly crucial by allowing for the combination of fabricated silicon dies, so-called chiplet, to constitute system-in-package (SiP) achieving a much better yield and time-to-market. However, due to inherent security concerns within the convoluted semiconductor supply chain and in-field environment, hostile attacks targeting software and hardware applications can present a formidable challenge to ensuring the security of SiP. Even worse, the immanent black-box nature of product chiplets renders most conventional security inspection and testing solutions less useful. Therefore, we present our SiPGuard in this article to enable the security monitoring capability during run time to noninvasively track the application-level behaviors of target chiplets and detect any deviations potentially induced by underlying malicious intrusions. The security monitoring mechanism utilizes information-bearing system-level power noise variation and machine learning (ML) techniques. Specifically, we utilize a trusted field-programmable gate array (FPGA) chiplet as our trust anchor to implement the lightweight power sensor and on-chip ML inference engine for near-sensor analysis. We prototype our solution on a 2.5-D chiplet-based FPGA device and demonstrate the effectiveness against threats at software/hardware levels by identifying the consequent power anomalies of malicious activities.
Field-programmable gate array (FPGA) bitstream reverse engineering and counterfeiting is a pertinent challenge in the modern hardware supply chain. To this end, this article proposes a blockchain-based technology to foster authenticity and integrity of the FPGA supply chain for trustworthy traceability. The proposed approach is transformative in being able to detect counterfeit FPGA chips and bitstreams using state-of-the-art blockchain technologies.—Kanad Basu, The University of Texas at Dallas
Physical side-channel attacks utilize power, electromagnetic (EM), or timing signatures from cryptographic implementations during operation to retrieve sensitive information from security-critical devices. This paper provides a comprehensive review of these potent attacks against cryptographic hardware implementations, with a particular emphasis on pre-silicon leakage assessment methodologies. We explore the intricacies of cryptographic algorithms, various side-channel attacks, and the latest mitigation techniques. Although leakage assessment techniques are widely adopted in the post-silicon phase, pre-silicon leakage assessment is an emerging field that addresses the inherent limitations of its post-silicon counterpart. We scrutinize established post-silicon techniques and provide a detailed comparative analysis of pre-silicon leakage assessment across different abstraction levels in the hardware design and verification flow. Furthermore, we categorize and discuss existing pre-silicon power and electromagnetic modeling techniques for leakage detection and mitigation that can be integrated with electronic design automation (EDA) tools to automate security assessments. Lastly, we offer insights into the future trajectory of physical side-channel leakage assessment techniques in the pre-silicon stages, highlighting the need for further research and development in this critical area of cybersecurity.
25 March 2024 in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
The ease and inexpensive setup of injecting timing faults in a hardware design make it vulnerable to adversaries, resulting in confidentiality or integrity violations. The state-of-the-art fault-injection attack assessment frameworks do not consider significant timing variations during layout generation from a gate-level design when assessing security threats of timing faults. Additionally, existing mitigation methods focus on higher design abstractions (e.g., register transfer level (RTL) and gate level), resulting in substantial area, power consumption, and latency overhead. To address these limitations, we propose our layout-aware and security property-assisted timing fault-injection attack assessment (FLAT) framework that automatically assesses the feasibility of injecting controlled timing faults into the layout of a design using clock glitches and quantifies its vulnerability concerning security properties. If the design is vulnerable, FLAT modifies the layout to tune the fan-in path delays of the security-critical registers as local countermeasures. Unlike system-wide mitigation approaches, these countermeasures incur minimal overheads at an IP or system-on-chip (SoC) design regarding power, performance, and area while ensuring security against timing faults. To demonstrate the effectiveness of FLAT, we perform security assessments on the postlayout designs of various benchmarks [e.g., advanced encryption standard (AES), rivest-shamir-adleman (RSA), and floating-point unit (FPU)] by targeting major fault injection attack vectors and deploying local countermeasures. These assessments indicate that the FLAT framework adeptly evaluates each design’s susceptibility to timing faults and implements the countermeasures to mitigate this susceptibility to the desired level.