59 research outputs found
Reconfigurable Intelligent Surfaces for Enhanced Localisation Advancing Performance with KAN-Based Deep Learning Models
Get PDF, ; Shah, Syed Tariq; Ullah, Insaf; Mahboob, Tahira; Abdellatif, Ahmed Gamal; Elaziz, Mohamed Abd; Almogren, Ahmad; Shawky, Mahmoud A
Bioethanol production from renewable sources as alternative valorization of waste of starting dates in south Algeria
No full textPost-Quantum Protected Federated Learning with Explainable and Adaptive Intelligence for Smart City Transportation
No full textExisting AI-powered Intelligent Transportation Systems (ITS) have limitations in scalability, privacy, and vulnerability to cyberattacks, as well as a lack of transparency in decision-making. In this work, we present a hybrid framework based on Post-Quantum-protected Federated Learning, a lightweight CNN-Transformer model, LIME explanations, and a local model, achieving a loss of 0.02% and a validation accuracy of 98%. At the boundary, congestion is determined using CityFlowV2 traffic camera feeds, which are based on Federated Learning, a distributed training framework that does not require sharing raw data, and the architecture is privacy-respectful. Reinforcement learning trained on OpenStreetMap road networks in Los Angeles coordinates rerouting plans in a simulated environment at the global level, and SHAP provides an explanation of the decision. The Federated aggregation retained accuracy at the zone level, exceeding 97%. Furthermore, this affirms its strength. CRYSTALS-Kyber is used to encrypt V2I and V2V communications, ensuring they are resistant to attacks in the quantum era. The framework is scalable and interpretative, and offers a secure, adaptable, city-neutral blueprint of next-generation ITS
Post-Quantum Protected Federated Learning with Explainable and Adaptive Intelligence for Smart City Transportation
No full textExisting AI-powered Intelligent Transportation Systems (ITS) have limitations in scalability, privacy, and vulnerability to cyberattacks, as well as a lack of transparency in decision-making. In this work, we present a hybrid framework based on Post-Quantum-protected Federated Learning, a lightweight CNN-Transformer model, LIME explanations, and a local model, achieving a loss of 0.02% and a validation accuracy of 98%. At the boundary, congestion is determined using CityFlowV2 traffic camera feeds, which are based on Federated Learning, a distributed training framework that does not require sharing raw data, and the architecture is privacy-respectful. Reinforcement learning trained on OpenStreetMap road networks in Los Angeles coordinates rerouting plans in a simulated environment at the global level, and SHAP provides an explanation of the decision. The Federated aggregation retained accuracy at the zone level, exceeding 97%. Furthermore, this affirms its strength. CRYSTALS-Kyber is used to encrypt V2I and V2V communications, ensuring they are resistant to attacks in the quantum era
Safe and Quickest Medical Image Encryption using Logistic Map-Derived S-Boxes and Galois Field
Get PDFThe pseudo-randomness, simplicity of use, and extreme sensitivity to even the slightest change in the initial value and handling parameters make chaotic maps attractive. The use of medical imaging to diagnose illnesses has grown in significance. These photographs need strong security measures because they are exchanged over public networks. Several techniques have been proposed to decode medical images, but they are not widely used due to their speed and complexity. Given these problems, we suggest a new method for quickly and efficiently encrypting medical images to safeguard private medical information from adversary assaults while it is being sent. This method uses the Logistic Map (LM), which is the main source of inspiration for this work. A simple polynomial that is not reducible to linear components is used to construct the Substitution Box (S-Box). When the LM is put into practice, many point pairs are produced. One of the coordinate values is selected from each point. The Galois Field (GF) is then added to that value. The finite field inverse is applied if the value is not zero; otherwise, nothing is altered. Choose any value below 256 at random. Make an S-Box using these randomly selected values. Lastly, the Exclusive-NOR (XNOR) operation between the S-Box and picture matrices was used to encrypt the medical image. High-security tests are performed to verify the reliability of the proposed technique. According to performance studies, the medical picture encryption approach based on LM and S-boxes offers extremely secure encryption in a short period
Proxy Promised Signcrypion Scheme Based on Elliptic Curve Crypto System
Get PDFWith the rapid growth in internet technology anonymity, repudiation and smacking the contents of messages are required for illegal businesses such as money laundering etc. In this paper we design and analyze a proxy promised signcrypion scheme based on elliptic curve cryptosystem. In this system the sender/original signer can give the authority of signcrypion to another entity namely proxy signcrypter and he generates promised signcryptext on the place of sender. The scheme is accomplished aim to improve the previous crypto-systems, due to the elliptic curve small system parameter, small public key certificates, faster implementation, low power consumption and small hardware processor requirements. This ECC based scheme provides high security and efficiency
An Anonymous Certificateless Signcryption Scheme for Secure and Efficient Deployment of Internet of Vehicles
No full textInternet of Vehicles (IoV) is a specialized breed of Vehicular Ad-hoc Networks (VANETs) in which each entity of the system can be connected to the internet. In the provision of potentially vital services, IoV transmits a large amount of confidential data through networks, posing various security and privacy concerns. Moreover, the possibility of cyber-attacks is comparatively higher when data transmission takes place more frequently through various nodes of IoV systems. It is a serious concern for vehicle users, which can sometimes lead to life-threatening situations. The primary security issue in the provision of secure communication services for vehicles is to ensure the credibility of the transmitted message on an open wireless channel. Then, receiver anonymity is another important issue, i.e., only the sender knows the identities of the receivers. To guarantee these security requirements, in this research work, we propose an anonymous certificateless signcryption scheme for IoV on the basis of the Hyperelliptic Curve (HEC). The proposed scheme guarantees formal security analysis under the Random Oracle Model (ROM) for confidentiality, unforgeability, and receiver anonymity. The findings show that the proposed scheme promises better security and reduces the costs of computation and communication
An Efficient Online/Offline Signcryption Scheme for Internet of Things in Smart Home
Get PDFpublishedVersio
A Lightweight Multi-Message and Multi-Receiver Heterogeneous Hybrid Signcryption Scheme based on Hyper Elliptic Curve
No full textAn Improved Aggregation-Based Signcryption for Secure Drone to Ground Station Communication System
Get PDFSecure and efficient communication in drone-assisted networks is critical for maintaining the integrity
of cyber-physical systems. This paper presents a rigorous cryptographic analysis of an existing
aggregation-based signcryption scheme, revealing key vulnerabilities, including susceptibility to
forgery, impersonation attacks, and inconsistencies in key generation and verification processes. To
overcome these limitations, we propose a novel aggregation-based signcryption framework built upon
hyperelliptic curve cryptography (HECC), offering enhanced security and lightweight computation.
The proposed scheme is formally proven to satisfy fundamental cryptographic properties, including
confidentiality, authentication, integrity, unforgeability, and impersonation resistance under the
hardness assumption of the hyperelliptic curve discrete logarithm problem (HECDLP). Performance
evaluation demonstrates that our approach achieves up to 36% reduction in computational cost and
12% lower communication overhead compared to the schemes of Verma et al. [1], Aithekar et al. [2],
and Ali et al. [3]. These results confirm the practical applicability of our design for secure and efficient
drone-to-ground station communication in resource-constrained environments
- …
