SaTC: CORE: Small: SLIQ: Securing Large-Scale Noisy-Intermediate Scale Quantum Computing

SaTC：核心：小型：SLIQ：确保大规模噪声中级量子计算的安全

• 批准号: 2129675
• 负责人: Swaroop Ghosh
• 金额: $ 50万
• 依托单位: Pennsylvania State Univ University Park
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-02-01 至 2025-01-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2129675&HistoricalAwards=false
• 关键词: SaTC; CORE; Small; SLIQ; Securing

Application-specific parametric quantum circuits designed to solve societal and science problems using powerful quantum computers contain assets and embed Intellectual Property (IP). Next generation quantum computers will face two new security issues: (i) evolution of multi-tenant computing (MTC) where multiple programs will share the hardware. The unwanted coupling between qubits can leak information and allow fault/Trojan injection; (ii) dependence on untrusted third-party compilers which can steal IP and tamper the circuit. This project will identify the vulnerabilities and threat vectors and develop a suite of defenses at the circuit and system level to secure future large scale quantum computing.The intellectual merits of this project are rooted at, (i) identification of various assets in quantum circuits, algorithms and hardware; (ii) identification of vulnerabilities in MTC environment such as, crosstalk, qubit allocation, scheduling and hardware splitting policies and untrusted compiler such as, embedding of IP, crosstalk, sensitivity of circuit quality to qubit allocation and gates; (iii) identification of security and privacy threats from MTC environment e.g., crosstalk-based fault/Trojan injection and data leakage and untrusted compiler such as, reverse engineering and tampering; (iv) development of circuit and system-level defenses such as, machine learning-based screening, obfuscation and split compilation. By addressing security, this project will remove a major roadblock towards applicability of quantum computers in security/privacy sensitive sectors. It will lead to widespread adoption of quantum computers in healthcare, energy and defense sectors and therefore, make a positive economic impact. This project will establish collaboration with industry and experts from computer science and cryptography communities to exchange ideas and will contribute directly to the Quantum Initiative Act and AI initiative Act. It will train new generation of workforce to solve problems using quantum computing. The tools and methodologies from this project can be plugged into existing commercial optimization tools. All publicly released data will be posted at PI’s academic website http://personal.psu.edu/~szg212 and the code developed in this project will be released in github (https://github.com/szg212) throughout the project and preserved without restrictions after the award ends. The conference and journal papers produced in this project will be archived in publisher’s website (e.g., ACM, IEEE). The presentations, tutorials and images will be hosted in PI’s academic website throughout the project and preserved without restrictions after the award ends. The video files of demonstrations and presentations will be uploaded in YouTube and will be preserved without restrictions after the project.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

专用的参数量子电路，旨在使用包含资产和嵌入知识产权（IP）的强大量子计算机解决社会和科学问题。下一代量子计算机将面临两个新的安全问题：（i）多租户计算（MTC）的演变，其中多个程序将共享硬件。配额之间不需要的耦合会泄漏信息并允许故障/木马注射； （ii）依赖不受信任的第三方编译器，这些编译器可以窃取IP并篡改电路。该项目将确定脆弱性和威胁向量，并在电路和系统级别开发一套防御措施，以确保未来的大规模量子计算。该项目的智力优点植根于（i）识别量子电路，算法和硬件中各种资产的识别； （ii）识别MTC环境中的漏洞，例如，串扰，配额分配，调度和硬件拆分策略以及不信任的编译器，例如IP的嵌入，串扰，串扰，电路质量对配额分配和门的敏感性； （iii）确定MTC环境的安全性和隐私威胁，例如基于串扰的故障/特洛伊木马注入以及数据泄漏以及不受信任的编译器，例如，逆向工程和篡改； （iv）开发电路和系统级防御措施，例如基于机器学习的筛选，混淆和拆分汇编。通过解决安全性，该项目将消除量子计算机在安全/隐私敏感部门中适用的主要障碍。这将导致量子计算机在医疗保健，能源和防御部门的广泛采用，因此产生了积极的经济影响。该项目将与来自计算机科学和密码社区的行业和专家建立合作，以交换思想，并将直接为《量子倡议法》和AI Initiative Act做出贡献。它将培训新一代劳动力，以使用量子计算解决问题。该项目的工具和方法可以插入现有的商业优化工具中。所有公开发布的数据将发布在PI的学术网站http://personal.psu.edu/~szg212，该项目中开发的代码将在Github（https://github.com/szg212）发布，并在整个项目中保留，并在授予结束后不限于限制。该项目中生产的会议和期刊论文将在出版商的网站（例如ACM，IEEE）中存档。演讲，教程和图像将在PI的学术网站中托管，并在奖励结束后无限制地保存。演示和演示文稿的视频文件将在YouTube上载，并将在项目之后不限于限制。该奖项反映了NSF的法定任务，并使用基金会的知识分子优点和更广泛的影响评估审查标准，通过评估被认为是珍贵的支持。

项目成果

Trainable PQC-Based QRAM for Quantum Storage

用于量子存储的可训练的基于 PQC 的 QRAM

• DOI: 10.1109/access.2023.3278600
• 发表时间: 2023
• 期刊: IEEE Access
• 影响因子: 3.9
• 作者: Phalak, Koustubh;Li, Junde;Ghosh, Swaroop
• 通讯作者: Ghosh, Swaroop

Special Session: On the Reliability of Conventional and Quantum Neural Network Hardware

特别会议：论传统和量子神经网络硬件的可靠性

• DOI: 10.1109/vts52500.2021.9794194
• 发表时间: 2022
• 期刊: VLSI Test Symposium
• 作者: Sadi, Mehdi;He, Yi;Li, Yanjing;Alam, Mahabubul;Kundu, Satwik;Ghosh, Swaroop;Bahrami, Javad;Karimi, Naghmeh
• 通讯作者: Karimi, Naghmeh

Split Compilation for Security of Quantum Circuits

量子电路安全的分割编译

• DOI: 10.1109/iccad51958.2021.9643478
• 发表时间: 2021
• 期刊: 10.1109/ICCAD51958.2021.9643478
• 作者: Saki, Abdullah Ash;Suresh, Aakarshitha;Topaloglu, Rasit Onur;Ghosh, Swaroop
• 通讯作者: Ghosh, Swaroop

Knowledge Distillation in Quantum Neural Network using Approximate Synthesis

• DOI: 10.1145/3566097.3567877
• 发表时间: 2022-07
• 期刊: 2023 28th Asia and South Pacific Design Automation Conference (ASP-DAC)
• 作者: M. Alam;Satwik Kundu;Swaroop Ghosh

SCANN: Side Channel Analysis of Spiking Neural Networks

• DOI: 10.3390/cryptography7020017
• 发表时间: 2023-03
• 期刊: Cryptogr.
• 作者: Karthikeyan Nagarajan;Rupshali Roy;R. Topaloglu;Sachhidh Kannan;Swaroop Ghosh