Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption
The increasing sophistication of cyber threats, particularly in decentralized and resource-constrained environments such as the Internet of Things (IoT), demands adaptive and efficient security solutions. This study introduces SignReencryption, a unified framework that integrates signcryption, proxy...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project / Dissertation / Thesis |
| Published: |
2025
|
| Subjects: | |
| Online Access: | http://eprints.utar.edu.my/7281/1/SE_2105387_FYP_Report_%2D_TeeJunnJeh_TEE_JUNN_JEH.pdf http://eprints.utar.edu.my/7281/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1855616549126144000 |
|---|---|
| author | Tee, Junn Jeh |
| author_facet | Tee, Junn Jeh |
| author_sort | Tee, Junn Jeh |
| building | UTAR Library |
| collection | Institutional Repository |
| content_provider | Universiti Tunku Abdul Rahman |
| content_source | UTAR Institutional Repository |
| continent | Asia |
| country | Malaysia |
| description | The increasing sophistication of cyber threats, particularly in decentralized and resource-constrained environments such as the Internet of Things (IoT), demands adaptive and efficient security solutions. This study introduces SignReencryption, a unified framework that integrates signcryption, proxy re-encryption (PRE), and Transformer-based intrusion detection to deliver both cryptographic assurance and intelligent adaptability. Signcryption ensures confidentiality and authenticity in a single lightweight operation, while PRE enables scalable, fine-grained access control without exposing plaintext. A TabTransformer-based intrusion detection system complements these cryptographic mechanisms, achieving classification accuracies of 94% on CICIDS2017, 99% on CIDDS-001, and 97% on NSL-KDD, with particular strength in detecting minority attack classes traditionally overlooked by baseline models. Optuna-driven hyperparameter optimization revealed dataset-specific configurations, demonstrating the adaptability of the TabTransformer across heterogeneous traffic distributions. Experimental evaluation further shows that SignReencryption reduces ciphertext expansion by up to 50% and lowers per-message execution time by nearly half compared to conventional Sign-Then-Encrypt schemes, confirming its practicality for real-time and bandwidth-limited environments such as intelligent transportation systems. Overall, the framework advances intrusion detection by uniting cryptographic efficiency with adaptive intelligence, offering a scalable, resilient, and operationally viable defense model for modern cybersecurity challenges.
Keywords: Signcryption; Cryptography; Transformer Neural Network; Intrusion Detection System; Internet of Things
Subject Area: QA75.5-76.95 |
| format | Final Year Project / Dissertation / Thesis |
| id | my-utar-eprints.7281 |
| institution | Universiti Tunku Abdul Rahman |
| publishDate | 2025 |
| record_format | eprints |
| spelling | my-utar-eprints.72812026-01-13T10:03:57Z Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption Tee, Junn Jeh QA75 Electronic computers. Computer science QA76 Computer software The increasing sophistication of cyber threats, particularly in decentralized and resource-constrained environments such as the Internet of Things (IoT), demands adaptive and efficient security solutions. This study introduces SignReencryption, a unified framework that integrates signcryption, proxy re-encryption (PRE), and Transformer-based intrusion detection to deliver both cryptographic assurance and intelligent adaptability. Signcryption ensures confidentiality and authenticity in a single lightweight operation, while PRE enables scalable, fine-grained access control without exposing plaintext. A TabTransformer-based intrusion detection system complements these cryptographic mechanisms, achieving classification accuracies of 94% on CICIDS2017, 99% on CIDDS-001, and 97% on NSL-KDD, with particular strength in detecting minority attack classes traditionally overlooked by baseline models. Optuna-driven hyperparameter optimization revealed dataset-specific configurations, demonstrating the adaptability of the TabTransformer across heterogeneous traffic distributions. Experimental evaluation further shows that SignReencryption reduces ciphertext expansion by up to 50% and lowers per-message execution time by nearly half compared to conventional Sign-Then-Encrypt schemes, confirming its practicality for real-time and bandwidth-limited environments such as intelligent transportation systems. Overall, the framework advances intrusion detection by uniting cryptographic efficiency with adaptive intelligence, offering a scalable, resilient, and operationally viable defense model for modern cybersecurity challenges. Keywords: Signcryption; Cryptography; Transformer Neural Network; Intrusion Detection System; Internet of Things Subject Area: QA75.5-76.95 2025 Final Year Project / Dissertation / Thesis NonPeerReviewed application/pdf http://eprints.utar.edu.my/7281/1/SE_2105387_FYP_Report_%2D_TeeJunnJeh_TEE_JUNN_JEH.pdf Tee, Junn Jeh (2025) Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption. Final Year Project, UTAR. http://eprints.utar.edu.my/7281/ |
| spellingShingle | QA75 Electronic computers. Computer science QA76 Computer software Tee, Junn Jeh Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title | Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title_full | Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title_fullStr | Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title_full_unstemmed | Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title_short | Adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signReencryption |
| title_sort | adaptive cryptography: a transformer neural network-based approach for anomaly detection and secure messaging with signreencryption |
| topic | QA75 Electronic computers. Computer science QA76 Computer software |
| url | http://eprints.utar.edu.my/7281/1/SE_2105387_FYP_Report_%2D_TeeJunnJeh_TEE_JUNN_JEH.pdf http://eprints.utar.edu.my/7281/ |
| url_provider | http://eprints.utar.edu.my |