Cyber-Physical Systems Security.pdf

发布时间：2022-06-06 发布人：admin 分类：说明书资料大小：15.39M 资料格式：pdf 举报版权申诉

weixin_38290023-11178972-4744300845404484576.pdf-第1页.png

第1页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第2页.png

第2页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第3页.png

第3页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第4页.png

第4页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第5页.png

第5页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第6页.png

第6页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第7页.png

第7页 / 共347页

weixin_38290023-11178972-4744300845404484576.pdf-第8页.png

第8页 / 共347页

Preface

Contents

Contributors

Robust Digital Computation in the Physical World

1 Introduction

2 Limitations of Digital Models in an Analog World

2.1 Introduction

2.2 Physics of Hybrid Modeling

2.3 Definition of the Thermostat Model

2.4 Informal Analysis of the Thermostat Model

2.5 Formal Implementation

3 Modeling and Verification of Out-of-Nominal Logic

3.1 Introduction

3.2 Modeling Out-of-Nominal Safety Properties

3.3 Example Turnstile Model

3.4 Design and Out-of-Nominal Verification Via Abstraction

3.4.1 Refinement (High Level)

3.4.2 Implementation (Low Level)

4 Resilience of Computational Physics Simulation

4.1 Introduction

4.2 Methodology

4.2.1 Operations in PDE Solvers

4.2.2 Error Model and Mitigation Approach

4.3 Application to the Conjugate Gradient Solver

4.3.1 The Basic Conjugate Gradient Solver

4.3.2 Controlling Convergence of the CG Algorithm

4.3.3 Evaluation Using In Situ Interpolation

5 Conclusion

References

Constraint-Based Framework for Reasoning with DifferentialEquations

1 Introduction

2 Preliminary Notions

2.1 Numerical Constraint Satisfaction Problems

2.2 Branch-and-Contract Solving Method

2.3 Some Limitations on NCSP

2.3.1 Equality Constraints

2.3.2 Differential Constraints

3 Set-Based Constraint Satisfaction Differential Problems

3.1 Dynamical Systems

3.2 Set-Based Constraints

3.3 Set-Based Differential Constraint Satisfaction Problems

4 Solving SCSDP

4.1 Interval-Based Constraints

4.2 Interval-Based Differential Constraints

4.2.1 Outer Approximation of Differential Constraints

4.2.2 Inner Approximation of Differential Constraints

4.3 Revisiting Branch-and-Contract Solving Method

4.3.1 Contraction

4.3.2 Propagation

5 Numerical Example

6 Conclusion

References

Approximate Computing and Its Application to Hardware Security

1 Introduction

2 Approximate Circuit

2.1 Approximate Adders

2.2 Approximate Multipliers

2.3 Approximate Dividers

3 Approximate Software/Algorithm

4 Approximate Computing for Hardware Security

4.1 Security Primitives Based on Approximate Computing

4.1.1 Floating-Point Format with Embedding Security

4.1.2 Approximate Computing with Embedded Security Information

4.2 A Low-Voltage Approximate Computing Adder for Authentication

5 Future Research Directions

5.1 PUFs and SCAs

5.2 SCAs

5.3 Hardware Trojans (HTs)

5.4 Approximate Arithmetic Circuit for Logic Obfuscation

6 Conclusion

References

Mathematical Optimizations for Deep Learning

1 Introduction

2 Pruning

3 Quantization

3.1 Binary Weights

3.2 Binary Weights and Activations

4 Weight Sharing and Compression

4.1 Weight Sharing

4.2 Compression

5 Model Distillation

6 Filter Decomposition

7 Conclusion

References

A Zero-Entry Cyber Range Environment for Future Learning Ecosystems

1 Introduction

2 Limitations of Current Practice

2.1 Specific Problem Being Solved

3 Research

3.1 Research Question

3.2 Learning Science Approach

3.2.1 Simulation Experience Design Method

3.2.2 Distributed Cognition Theory

3.3 Learner Analysis

3.4 Cybersecurity Terminal Learning Objectives

4 Cyber Scorpion Design

4.1 Cyber Scorpion Learner Sketch

5 Software Environment

5.1 Cyber Scorpion Underlying Technology: Minimega

5.2 Software Development Approach

6 Learner Experience Demonstration and Lessons Learned

7 Limitations and Future Work

8 Conclusion

References

Parallel Programming in Cyber-Physical Systems

1 Introduction

1.1 Parallel Architectures

2 Parallel Programming in CPSs

2.1 Experimental Methodology

2.2 Modular Exponentiation

2.2.1 m-ary Approach

2.2.2 Slicing Approach

2.2.3 Conclusions: Modular Exponentiation

2.3 Karatsuba Multiplication

3 Conclusions

Appendix: Sequential and Basic Parallel Code for Karatsuba

References

Automatic Application of Software Countermeasures Against Physical Attacks

1 Introduction

2 Background

2.1 Side-Channel Attacks

2.2 Fault Injection Attacks

2.3 Combined Attacks

2.4 Countermeasures

2.5 Compilation of Secured Code

3 Automatic Application of Software Countermeasures

3.1 At Source Code Level

3.1.1 Side-Channel Attack Countermeasures

3.1.2 Fault Injection Countermeasures

3.1.3 Pros and Cons of Source Code Level

3.2 During Compilation

3.2.1 Side-Channel Attack Countermeasures

3.2.2 Fault Injection Countermeasures

3.2.3 Pros and Cons of Compiler Level

3.3 At Link Time/At Assembly Level

3.3.1 Side-Channel Attack Countermeasures

3.3.2 Fault Attack Countermeasures

3.3.3 Pros and Cons of Assembly Level

4 Discussion

4.1 Confrontation of Pros and Cons of the Different Levels

4.2 Future Works

5 Conclusion

References

Time-Delay Attacks in Network Systems

1 Introduction

2 Problem Setup

2.1 Network Model

2.2 Attack Model

2.3 Problem Formulation

3 Minimum Cardinality Attack Sets

3.1 Optimal Delay Attacks

3.2 Numerical Methods for Finding Optimal Attacks

4 Optimal Attack Sets and Relation with Topology

5 Conclusions

References

Attack Tree Construction and Its Application to the ConnectedVehicle

1 Introduction

1.1 Attack Trees in the Automotive Domain

1.2 Attack Tree Generation

1.3 Contributions

2 Background and Definitions

2.1 Automotive Architecture

2.2 Architectural Graph

Running Example

2.3 Graph Transformation

Example

3 Attack Graph Generation

4 Attack Tree Generation

5 Countermeasure

6 Conclusion

References

Reinforcement Learning and Trustworthy Autonomy

1 Introduction

2 Reinforcement Learning Preliminaries

2.1 Markov Decision Processes

2.2 Reinforce Method

3 Microsoft's AirSim

3.1 Overview

3.2 Python APIs

4 Reinforcement Learning in AirSim

4.1 Unreal Dynamic Environment

4.2 Python Environment Library

4.3 REINFORCE Method in AirSim

5 Increased Trustworthiness Through Visualization

5.1 t-SNE

5.2 Action Visualization

5.3 Attribution Visualization

6 Conclusion

References

Identifier Randomization: An Efficient Protection Against CAN-Bus Attacks

1 Introduction

2 State-of-the-Art CAN Protections

2.1 Controller Area Network Overview

2.2 Payload Protection

Flaws

2.3 Intrusion Detection and Prevention Systems

Flaws

2.4 Identifier Protection

Flaws

3 Solutions Based on Randomization and Their Evaluation

3.1 Principle and Formalism

3.2 Evaluation Metrics

3.2.1 Reverse-Engineering Attack

3.2.2 Replay and Injection Attacks

3.3 The IA-CAN Approach

3.3.1 Particular Case

3.3.2 Testing

3.4 Equal Intervals

3.4.1 Testing

3.5 Frequency Intervals

3.5.1 Testing

3.6 Dynamic Intervals

3.6.1 Illustrative Example

3.6.2 Testing

3.7 Arithmetic Masking

3.7.1 Testing

4 Comparison

5 Conclusion

Appendix

Entropy of Fixed Mapping

Conditional Entropy of Fixed Mapping

Entropy of Dynamic Intervals

Entropy of Arithmetic Masking

Conditional Entropy of Arithmetic

Fixed Mapping Optimality Proof

References

Public Key-Based Lightweight Swarm Authentication

1 Introduction

1.1 Related Work

1.2 Structure of This Chapter

2 Preliminaries

2.1 Fiat–Shamir Authentication

2.2 Topology-Aware Distributed Spanning Trees

2.2.1 Topology-Aware Networks

2.2.2 Mooij–Goga–Wesselink's Algorithm

3 Distributed Fiat–Shamir Authentication

3.1 The Approach

3.2 Backup Authentication

4 Security Proofs

4.1 Soundness

4.2 Zero Knowledge

4.3 Security Analysis

4.3.1 Choice of Parameters

4.3.2 Algorithmic Complexity

4.3.3 Root Causes of Authentication Failure

4.3.4 Effect of Network Noise

4.3.5 Man-in-the-Middle Attacks

5 Variants and Implementation Trade-Offs

5.1 Shorter Challenge Variant

5.2 Multiple-Secret Variant

5.3 Precomputed Alphabet Variant

5.4 Precomputed Combination Variant

6 Conclusion

References

Physical Security Versus Masking Schemes

1 Context About the Protection Problem

1.1 Nature of Computation

1.2 Combinational or Sequential?

1.3 Outline of the Article

2 Definition of t-Order Security by ISW [12]

2.1 Revisiting of ISW Definition

2.2 Ill-Formed Definition

2.3 Attack on Coron's Higher-Order Masking of Look-Up Tables [8]

2.4 Motivation for Bit-Mixing Masking Schemes

3 Analysis of the Security Issue

3.1 Hardware Case

3.2 Software Case

6502

LEON3

Analysis of the 6502 and LEON3 Codes

4 New Definition of Security Order

5 Conclusion

References

Embedded Classifiers for Energy-Constrained IoTNetwork Security

1 Sensor Nodes Empowered by SoC FPAA Devices

2 Low-Power Context-Aware FPAA Architectures

3 FPAAs as Physical Computation Devices

4 Embedded FPAA Security Concerns

4.1 Positive FPAA Security Attributes

4.2 Addressing FPAA Security Issues

5 FPAAs for Investigating IC Validation

5.1 Black Box (BB) Exam: CNS 182

5.2 Training IC Deconstruction Using FPAA BB Approach

5.3 FPAAs for Unique Functions

6 Summary and Next Directions

References

Challenges in Cyber Security: Ransomware Phenomenon

1 Spora Ransomware

2 DMA Locker Ransomware

3 WannaCry Ransomware

Applying Model-Based Situational Awareness and Augmented Reality to Next-Generation Physical Security Systems

1 Introduction

2 Model-Based Situational Awareness for Physical Security

3 A Secure Facility Meets Its Digital Twin

4 Scene: Sunday, 1600. Somewhere in the Countryside…

4.1 A Digital Twin Saves the Day

5 Toward Digital Twin

6 Conclusion

References

Çetin Kaya Koç Ed. Cyber-Physical Systems Security

Cyber-Physical Systems Security

Çetin Kaya Koç Editor Cyber-Physical Systems Security 123

Editor Çetin Kaya Koç ˙Istinye University, ˙Istanbul, Turkey Nanjing University of Aeronautics and Astronautics, Nanjing, China University of California Santa Barbara Santa Barbara, CA, USA ISBN 978-3-319-98934-1 https://doi.org/10.1007/978-3-319-98935-8 ISBN 978-3-319-98935-8 (eBook) Library of Congress Control Number: 2018963758 © Springer Nature Switzerland AG 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface Cyber-physical systems involve interactions between computer-controlled and actu- ator-enabled components, whose dynamics have thus far been modeled, designed, and analyzed separately, but now need to be investigated and taught jointly. They control much of the world’s critical infrastructure: power generation, telecommuni- cations, water supply, and industrial control systems. Because of their critical nature, known, predictable, and secure behavior of cyber-physical systems is necessary to ensure the safety of the people whom these systems serve. Yet most cyber-physical systems provide limited operational guarantee outside of nominal conditions. Understanding how diverse physical and digital systems can be safely and securely combined is not a simple task. Solving the safety and security deﬁciencies in next generation cyber-physical systems will require contributions from every branch of engineering, from mechanical and power engineering to computer science and mathematics. Partners from university research labs, governments, and industry must come together. It is necessary that we establish an engaged, multidisciplinary cyber-physical security community committed to developing uniﬁed foundations, principles, and technologies. Cyber-physical as a term explains much of the underlying theory and practice; it is the interplay of physics and computation. Our understanding of the physical world through the models of classical and quantum physics, together with our models of computation from analog to digital, helps us build a better understanding of the cyber-physical world. Insights from physics, methods of complex systems theory, and formal methods borrowed from various facets of mathematical and computational sciences will help us to build reliable, safe, and secure systems. The advantages of digital computation are relatively low power requirements and ﬂexibility; their carefully built discrete combinational states offer abstractions of Boolean logic. However such abstractions are limited, and we need to discover the conditions under which a digital abstraction of a system or subsystem can be valid. Physical properties appear continuous and constrain how cyber-physical systems models can be constructed and analyzed using formal logic. The study of the cyber- physical system also necessitates the study of the topologies of complex systems in addition to the computational and physical properties of their components. v

vi Preface A cyber-physical system is a complex set of systems and subsystems requiring communication channels among the cooperative entities and tasks, for example, a coordinated platoon of interconnected vehicles or a countrywide power system of different generating and consuming plants. Overall stability of the system will be affected by adversary attacks that tamper with the temporal characteristics, causing the delays of signals from nodes to nodes. Understanding which channels are less robust and furthermore what kind of network topologies have more resilience will help us minimize the number and the overall effect of compromised channels. We created the Cyber-Physical Security Workshop and brought together researchers, engineers, and teachers into a common forum of exchange in order to achieve several goals, the primary one being to expose researchers, educators, and students to the world of CPS research. Invited speakers from academia and government labs offered glimpses of their work on modeling, analyzing, and understanding cyber-physical systems. Sandro Bartolini, University of Siena Alexandre Chapoutot, ENSTA ParisTech Télécom SudParis Hervé Debar, Arizona State University Georgios Fainekos, Georgia Institute of Technology Jennifer Hasler, Israel Koren, University of Massachusetts Sandia National Laboratories Jackson Mayo, University of California Riverside Fabio Pasqualetti, Sandia National Laboratories Elaine Raybourn, John D. Siirola, Sandia National Laboratories University of California, Santa Barbara Sam Green, The Workshop was initiated and organized by the steering committee, com- posed of Çetin Kaya Koç (˙Istinye University, Nanjing University of Aeronautics and Astronautics, and UC Santa Barbara), Patrick Duvaut (Télécom ParisTech), David Naccache (École normale supérieure), and Jennifer Troup (Sandia National Laboratories). The primary sponsor of the Workshop was the National Science Foundation with the award number 1638470 and the title “Cyber Physical Systems Security Educa- tion Workshop” with the Principal Investigator as Çetin Kaya Koç. The sponsors included Almerys, École normale supérieure, and Sandia National Laboratories. The Cyber-Physical Security Workshop was held on July 17–19, 2017, on the campus of Télécom ParisTech, located in central Paris, in the heart of a rich urban and cultural environment. My thanks are also due to the faculty and students at Télécom ParisTech for running the Workshop. I also sincerely thank Ronan Nugent of Springer for his valuable advice and the Editorial Ofﬁce of Springer for their help in getting the book published. Paris, France Çetin Kaya Koç

Contents Robust Digital Computation in the Physical World .......................... Jackson R. Mayo, Robert C. Armstrong, Geoffrey C. Hulette, Maher Salloum, and Andrew M. Smith Constraint-Based Framework for Reasoning with Differential Equations ......................................................................... Julien Alexandre dit Sandretto, Alexandre Chapoutot, and Olivier Mullier Approximate Computing and Its Application to Hardware Security ...... Weiqiang Liu, Chongyan Gu, Gang Qu, and Máire O’Neill Mathematical Optimizations for Deep Learning ............................. Sam Green, Craig M. Vineyard, and Çetin Kaya Koç A Zero-Entry Cyber Range Environment for Future Learning Ecosystems ........................................................................ Elaine M. Raybourn, Michael Kunz, David Fritz, and Vince Urias 1 23 43 69 93 Parallel Programming in Cyber-Physical Systems ........................... 111 Sandro Bartolini and Biagio Peccerillo Automatic Application of Software Countermeasures Against Physical Attacks .................................................................. 135 Nicolas Belleville, Karine Heydemann, Damien Couroussé, Thierno Barry, Bruno Robisson, Abderrahmane Seriai, and Henri-Pierre Charles Time-Delay Attacks in Network Systems ...................................... 157 Gianluca Bianchin and Fabio Pasqualetti Attack Tree Construction and Its Application to the Connected Vehicle ............................................................................. 175 Khaled Karray, Jean-Luc Danger, Sylvain Guilley, and M. Abdelaziz Elaabid vii

viii Contents Reinforcement Learning and Trustworthy Autonomy ....................... 191 Jieliang Luo, Sam Green, Peter Feghali, George Legrady, and Çetin Kaya Koç Identiﬁer Randomization: An Efﬁcient Protection Against CAN-Bus Attacks ................................................................ 219 Khaled Karray, Jean-Luc Danger, Sylvain Guilley, and M. Abdelaziz Elaabid Public Key-Based Lightweight Swarm Authentication ...................... 255 Simon Cogliani, Bao Feng, Houda Ferradi, Rémi Géraud, Diana Maimu¸t, David Naccache, Rodrigo Portella do Canto, and Guilin Wang Physical Security Versus Masking Schemes . .................................. 269 Jean-Luc Danger, Sylvain Guilley, Annelie Heuser, Axel Legay, and Tang Ming Embedded Classiﬁers for Energy-Constrained IoT Network Security ................................................................. 285 Jennifer Hasler Challenges in Cyber Security: Ransomware Phenomenon .................. 303 Vlad-Raul Pa¸sca and Emil Simion Applying Model-Based Situational Awareness and Augmented Reality to Next-Generation Physical Security Systems ...................... 331 Elaine M. Raybourn and Ray Trechter

分享到：

赞收藏

资料库

Cyber-Physical Systems Security.pdf

相关推荐

安全技术

热门标签

最新资料