Emerging Information Security and Applications: 4th International Conference, EISA 2023, Hangzhou, China, December 6–7, 2023, Proceedings (Communications in Computer and Information Science)

✍ Scribed by Jun Shao (editor), Sokratis K. Katsikas (editor), Weizhi Meng (editor)

• Publisher: Springer
• Year: 2024
• Tongue: English
• Leaves: 195
• Category: Library

✦ Synopsis

This volume constitutes the proceedings presented at the 4th International Conference on Emerging Information Security and Applications, EISA 2023, held in Hangzhou, China, in December 2023.
The 11 full papers presented in this volume were thoroughly reviewed and selected from the 35 submissions. The topics of the book are related but not limited to cyber intelligence techniques, multimedia security, blockchain and distributed ledger technology, malware and unwanted software, vulnerability analysis and reverse engineering, usable security and privacy, intrusion detection and prevention, authentication and access control, anonymity and privacy, cryptographic protection, digital forensics, cyber physical systems security, adversarial learning, security measurement, security management and policies, hardware and physical security.

✦ Table of Contents

Preface
Organization
Contents
PtbStolen: Pre-trained Encoder Stealing Through Perturbed Samples
1 Introduction
2 Preliminaries
2.1 Encoder Pre-training
2.2 Application in Downstream Tasks
3 Threat Model
3.1 Objectives
3.2 Capability
3.3 Background Knowledge
4 Details of PtbStolen
4.1 Overview
4.2 Problem Formulation
4.3 Algorithm of PtbStolen
5 Experiments
5.1 Attack Settings
5.2 Experimental Results
5.3 Cost Analysis
6 Related Works
6.1 Contrastive Learning
6.2 Model Stealing Attack
7 Conclusion
References
Towards Efficient Universal Adversarial Attack on Audio Classification Models: A Two-Step Method
1 Introduction
2 Related Works
2.1 Input-Dependent Adversarial Attacks
2.2 Universal Adversarial Attacks
3 The Proposed Method
3.1 Preliminaries
3.2 Craft Individual Perturbations
3.3 Fine-Turn Universal Perturbations
4 Experiments
4.1 Settings
4.2 Quantitative Evaluation on Untargeted Attack
4.3 Quantitative Evaluation on Targeted Attack
4.4 Attack with Different Normalization Methods
5 Conclusions
References
Privacy-Preserving Authenticated Federated Learning Scheme for Smart Healthcare System
1 Introduction
2 Related Work
3 Preliminaries
3.1 Federated Learning
3.2 Boneh-Goh-Nissim (BGN) Cryptosystem
3.3 Schnorr Signature
4 System Architecture and Security Model
4.1 System Architecture
4.2 Security Model
4.3 System Requirements
5 Proposed Specific Scheme
5.1 System Setup
5.2 Group Training
5.3 Server Aggregation
5.4 Local Model Update
6 Security Analysis
6.1 Correctness
6.2 Authentication of the Scheme
6.3 Data Privacy
7 Performance Evaluation
7.1 Evaluation Environment
7.2 Characteristic Analysis
7.3 Accuracy and System Runtime
7.4 Cryptosystem Runtime
7.5 Computational Cost of Clients Participating and Revoking Analysis
8 Conclusion
9 Discussion and Future Work
References
A Systematic Method for Constructing ICT Supply Chain Security Requirements
1 Introduction
2 Related Work
3 ICT Supply Chain Security Environment
3.1 Supply Chain Relationship
3.2 Product Life Cycle Stages
3.3 Security Driving Factors
3.4 Security Properties
4 ICT Supply Chain Security Requirements
4.1 Establishment Process of Security Requirements
4.2 Validation Process of Security Requirements
4.3 Summary of the ICT Supply Chain Security Requirements
5 Application of ICT Supply Chain Security Requirements
6 Conclusions and Future Work
References
Pairing Compression on Some Elliptic Curves with Subgroups of Embedding Degree 6 and Its Applications to Pairing-Based Cryptography
1 Introduction
1.1 Related Work
1.2 Contributions
2 Preliminaries
3 Pairing Compression
4 Applications to Pairing-Based Cryptography
4.1 BLS Short Signatures
4.2 Identity-Based Encryption
5 Conclusion
References
Enhancing Chinese Named Entity Recognition with Disentangled Expert Knowledge
1 Introduction
2 Related Work
2.1 Named Entity Recognition
2.2 Chinese Named Entity Recognition
3 Methods
3.1 Contextual Representation Generation
3.2 Disentangled Expert Knowledge Learning
3.3 Knowledge Enhancement
3.4 NER Prediction
3.5 Model Training
4 Experiments
4.1 Dataset
4.2 Evaluation Metrics
4.3 Implementation Details
4.4 Model Comparison
4.5 Performance on Fine-Grained Category
5 Conclusion
References
Deep Neural Network Model over Encrypted Data
1 Introduction
2 Related Work
3 DNN Classification Model for Ciphertext Data
3.1 General Framework of the Model
3.2 Communication Protocols
3.3 Classification Process
3.4 Security Analysis
4 Experiments
4.1 Experimental Setup
4.2 Analysis of Experimental Results
5 Conclusion
References
Toward the Tradeoffs Between Privacy, Fairness and Utility in Federated Learning
1 Introduction
2 Related Work
2.1 Fairness of FL
2.2 Privacy of FL
2.3 Fairness and Privacy of FL
3 Preliminaries
3.1 Fairness in FL
3.2 Privacy in FL
3.3 Problem Formulation
4 Method
4.1 Fairness Predictor (Model) in Client
4.2 Privacy Protection Method in Fair FL
5 Experiments
5.1 Dataset and Experimental Settings
5.2 Experimental Hyperparameter Settings
5.3 Performance Comparison Results
5.4 Analysis of Privacy and Fairness
6 Conclusion
References
Chinese Named Entity Recognition Within the Electric Power Domain
1 Introduction
2 Related Work
3 Method
3.1 Architecture
3.2 BERT Model
3.3 BiLSTM Model
3.4 CRF Model
4 Experiments
4.1 Electric Power Domain Dataset
4.2 Performance of Electric Power Dataset
4.3 Performance Comparison of NER Models
5 Conclusions
References
Adversarial Example Attacks and Defenses in DNS Data Exfiltration
1 Introduction
2 Background and Related Work
2.1 A Deeper Look into the DNS Protocol
2.2 Data Exfiltration
3 Preliminaries
3.1 Payload Analysis
3.2 Traffic Analysis
3.3 Data Exfiltration Detection Methods
4 Problem Formulation
4.1 Data Exfiltration Detection Using Machine Learning
4.2 Optimization Approach to Adversarial Example Packet Generation of Data Exfiltration Packets
4.3 Our Findings
5 Defense Against Malicious Evasion of DNS Exfiltration Detection
5.1 Ensemble Voting Approach
5.2 Adversarial Example Training
6 Conclusion
References
CONNECTION: COvert chaNnel NEtwork attaCk Through bIt-rate mOdulatioN
1 Introduction
2 Background
2.1 Covert Channel Characteristics
2.2 Brief Overview of Enterprise Networks
3 Related Works
4 Attacker Model
5 Bit-Rate Modulation as Covert Channel
5.1 Covert Sender
5.2 Covert Receiver
6 Evaluation of the Covert Channel
6.1 Results of Adversary Emulation in the Cyber Range
7 Discussion
8 Conclusions
References
Author Index