Writing a Good Security Paper for ISSCC (2025)

Utsav Banerjee, Chiraag Juvekar, Yong Ki Lee, Leibo Liu, Sanu Mathew, Thomas Poeppelmann, Shreyas Sen, Takeshi Sugawara, Ingrid Verbauwhede, Rabia Tugce Yazicigil · 2025

Read Writing a Good Security Paper for ISSCC (2025) by Utsav Banerjee, Chiraag Juvekar, Yong Ki Lee, Leibo Liu, Sanu Mathew, Thomas Poeppelmann, Shreyas Sen, Takeshi Sugawara, Ingrid Verbauwhede, Rabia Tugce Yazicigil (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-12700.txt.

Introduction

Utsav Banerjee and co-authors, members of a top chip conference's security subcommittee, coach researchers on writing strong security papers, in this 2025 guide. Hardware security became a formal track at the International Solid-State Circuits Conference only recently. The authors spent two years reviewing submissions in that track and saw the same weaknesses recur. Their aim is to raise submission quality and grow the young field. The paper distills what reviewers reward, from framing a threat model to presenting silicon results.

Catalog record

CREATORUtsav Banerjee, Chiraag Juvekar, Yong Ki Lee, Leibo Liu, Sanu Mathew, Thomas Poeppelmann, Shreyas Sen, Takeshi Sugawara, Ingrid Verbauwhede, Rabia Tugce Yazicigil
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE19 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-12700

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Writing a Good Security Paper for ISSCC (2025) Utsav Banerjee, Chiraag Juvekar, Yong Ki Lee, Leibo Liu, Sanu Mathew, Thomas Poeppelmann, Shreyas Sen, Takeshi Sugawara, Ingrid Verbauwhede, Rabia Tugce Yazicigil Abstract I Introduction Security is increasingly more important in designing chips and systems based on them, and the International Solid-State Circuits Conference (ISSCC), the leading conference for presenting advances in solid-state circuits and semiconductor technology, is committed to hardware security by establishing the security subcommittee since 2024. In the past two years, the authors of this paper reviewed submissions as members of the Security Subcommittee, a part of International Technical Program Committee (ITPC). This paper aims to encourage high-quality submissions to grow this field in the overall scope of the ISSCC. The first purpose of the document is to share our knowledge in writing good security papers. Hardware security is an interdisciplinary field, and writing papers in this area requires special attention, and we observed common and repeated mistakes. Although we are making an effort to improve the situation by providing review feedback to authors, the change made in this way occurs only gradually, and this paper is intended to accelerate the process. The first two sections are devoted to providing insight into how we will review and rate papers and how to avoid common mistakes. Section II begins with the general hints applicable to most papers. In Section III , we define specific requirements for papers that fall into specific subcategories, namely (i) implementations of fully homomorphic encryption (FHE) and post-quantum cryptography (PQC), (ii) evaluation of physically unclonable function (PUFs) and random number generator (RNG), and (iii) countermeasures against side-channel attacks (SCAs) and fault-injection attacks (FIAs). The second purpose of this document is to expand the scope of potential submissions. In the past two years, many papers have been accepted from the aforementioned three categories (i)–(iii). Researchers typically select conferences for submission based on previously accepted papers, and we are afraid of a situation where only similar types of papers continue to be submitted. We are open to more diverse papers, and Section IV describes the other categories that we would like to see submitted in the future. I-A Disclaimer This document captures the perspectives of the authors as of May 2025, noting that these opinions may evolve with time or community input. It compiles the personal viewpoints of the members and does not represent the official stance of their respective organizations or the ISSCC in general. II General Hints This part provides general recommendations likely to help most security papers. For additional tips, the readers may also consult the guide by Van der Spiegel and Smith [ 2 ] . II-A Innovation and Significance Innovation and Significance as defined below are the two main criteria in reviewing papers [ 3 ] : Innovation. The conference attendees want to see new ideas leading to new research directions, and this criterion values new designs, methods, products, solutions, applications, or new problem to solve with immediate or potential value, such as novel circuits and system architectures. Significance. Attendees also want to see the state-of-the-art, especially those from industry, and this criterion covers a positive impact on the system, product or SSCS community, including industry firsts, technology firsts, high-volume product deployment, and state-of-the-art advancements. Although a paper with outstanding innovation and significance is ideal, the ISSCC ITPC aims to accept many papers with either outstanding innovation or outstanding significance. For example, although a highly innovative paper may not have the best figure-of-merit (FoM), these contributions explore new circuit and system architectures that may approach problems from a unique perspective, challenge fundamental tradeoffs, or even open up new directions for future research and products. In contrast, a highly significant paper, describing an industry first or state-of-the-art advancements in ICs and SoCs, may be based on prior works. When innovative ideas are eventually integrated into products and/or advance the field, this is an important validation of the underlying design techniques. ISSCC ITPC members are instructed to review each paper based on the innovation and significance of the work. II-B Evaluation Benchmark. Papers on a circuit, building block, architecture, or system should provide comparisons with state-of-the-art publications on industry-recognized benchmarking methods and metrics. These benchmarks and comparisons should be conducted in a similar level (e.g., system, chip, subsystem, etc.), and if not, these distinctions should be clearly indicated and clarified. Performance Details. Circuit performance is affected by multiple factors, and some articles lack critical details to separate the improvement based on the claimed innovation from those based on other factors. Thus, make sure to provide enough information that the claimed innovation actually leads to an improvement in an important key performance indicator (KPI). Here are more concrete suggestions: • Provide measurement conditions such as voltage, frequency, and temperature. • Clarify what blocks or functions are included (or omitted) in the evaluation, especially for power consumption, area, and latency measurements. In case of omissions, e.g., off-chip memory accesses, these need to be explained sufficiently. • The evaluation of the area should include details such as the technology node and the circuit size in mm 2 . For papers aiming at technology-independent innovation, e.g., computing architecture, technology-independent metrics should be provided, such as the number of gates normalized with a 2-way NAND (GEs), the number of FFs, the area and the capacity of SRAMs, and other hard macros used. If possible, provide a breakdown of the circuit into sub-blocks to allow for a detailed evaluation. • If runtime is measured, it should be provided in absolute numbers (e.g., milliseconds) and clock cycles. • Provide sufficient details on power consumption, especially when it is claimed as the main KPI. Discriminate active and passive power consumption and provide information on peak power consumption. • Provide measurements across different voltage and frequency settings, especially when the chip claims a wide operating condition or features dynamic voltage and frequency scaling. Clarification of Silicon Boundary. The measured and simulated results should be clearly discriminated. In general, the main results should be based on real measurements, and simulation should be limited for alternative evaluations, such as (i) covering scenarios that are not implemented, (ii) initial design-space exploration, and (iii) presilicon validation that is later compared with real measurement. Out-of-Scope Components. The essence of security design can lie in the details, but authors can omit or simplify components that are not the main focus of the paper. For example, a key register that allows external read and write access is an immediate security problem, but such a design is acceptable unless such access control is the main subject of a paper. An LFSR is unacceptable as a cryptographic random number generator, but it can be used for some evaluation as long as the RNG is not the main technical contribution. Meanwhile, we should avoid readers from replicating the simplification for practical development, which can cause a security incident. Therefore, such non-ideal aspects should be clearly described as limitations in the paper. II-C Research Direction Process Node. Papers that explore the cutting edge in terms of process node are exciting. However, the submission of papers in mature process nodes is still highly encouraged. Such papers make sense when architectural innovation is targeted and when no

Read the complete text