Alanoud M. Almhlbdi, Norah D. Altowairqi, Areej Alshutayri, Rehab Qarout
Identifying hidden payloads in images has become increasingly critical as steganography continues to challenge traditional security measures. This paper introduces a deep learning framework for both the detection (binary classification) and fine-grained classification (multi-class) of steganographic payloads embedded using Least Significant Bit (LSB) techniques. The proposed system distinguishes between benign images and stego images containing five different payload types: HTML, JavaScript, PowerShell, URLs, and Ethereum-related data. To achieve this, we systematically evaluate various architectures, including a custom Convolutional Neural Network (CNN), hybrid CNN-GRU and CNN-LSTM models, and a Vision Transformer (ViT) at different input resolutions using 5-fold cross-validation. Our experiments reveal a critical finding: image resizing significantly degrades detection performance, as subtle LSB artifacts are often corrupted. While our custom CNN model achieved the highest mean cross-validation accuracy (0.9702), the hybrid CNN-GRU model demonstrated superior generalization on the held-out test set and external dataset, achieving a multi-class accuracy of 0.98 on the testset and 0.97 on the external. This result highlights the advantage of combining the CNN’s spatial feature extraction with the GRU’s ability to model sequential dependencies for robust payload identification on unseen data.
Open access
Advanced Steganography and Watermarking Techniques
This systematic literature review compares two leading selective-disclosure primitives for decentralised identity-BBS+ signatures and Selective-Disclosure JSON Web Tokens (SD-JWT)-to clarify their suitability for privacypreserving credentials. Following Kitchenham's protocol, 226 records from 2017-2025 were screened across IEEE, ACM, SpringerLink, ScienceDirect, IETF and W3C repositories, yielding 31 primary studies with empirical data. Quantitative synthesis shows that BBS+ derived proofs remain constant-size at roughly 140 bytes and verify in about 12 ms on consumer hardware, whereas SD-JWT presentations grow with the number of revealed claims but still verify in under 10 ms for typical twoclaim use cases. Qualitative analysis confirms BBS+ provides strong unlinkability, predicate proofs and zero-knowledge disclosure, while SD-JWT offers seamless integration with existing JOSE/OAuth infrastructures yet carries correlation risk due to stable salted digests. Standardisation progress is comparable: the BBS+ cryptosuite reached W3C Candidate Recommendation in April 2025, and SD-JWT is in late-stage IETF review. The review concludes that privacy-critical scenarios such as age-gated services favour BBS+, whereas high-throughput web applications benefit from SD-JWT; consequently, hybrid wallet support for both formats is recommended. Future research should tackle scalable revocation, post-quantum migration and multi-credential aggregation to sustain long-term trust and interoperability.
Urooj Waheed, Muhammad Ahsan Khan, Yusra Mansoor, Huma Jamshed · 5 authors
In any democratic electoral system the fundamental right of every eligible citizen is to vote in order to elect the desired representative. However minimal efforts have been made to improve the voting mechanism. Many states are still utilizing paper based balloting systems, however technological advancements have led to the introduction of electronic voting machines (EVMs) to improve security and maintain public trust in the electoral process. Irrespective of these efforts, lack of transparency, low voter turnout, and susceptibility to vote rigging persist in these E-Voting systems. Blockchain technology is emerging as a solution to address these challenges by enhancing security and trust in E-Voting, effectively mitigating its longstanding issues. The transparency, immutability, and resistance to tampering provided by blockchain enhance the reliability and credibility of the voting process. This paper presents a decentralized E-Voting system that uses a public blockchain built on Ethereum smart contracts to ensure public accountability and transparency in the E-Voting process.
This thesis investigates how blockchain technology and data science methods can jointly improve the security, transparency, and trustworthiness of electoral systems. Against a backdrop of rising concerns over election integrity in traditional voting systems, scholars and nations have begun to integrate blockchain systems. However, adoption appears to remain low. This study systematically reviews 116 peer-reviewed articles and analyzes five national case studies (Estonia, Switzerland, the United States, Russia, and Romania). It examines (1) blockchain’s potential to secure voter verification, prevent fraud, and ensure immutable vote records; (2) data-science methods (e.g. machine learning, zero-knowledge proofs, homomorphic encryption) for anomaly detection, performance optimization, and privacy preservation; and (3) the ethical, political, and societal implications of digital voting, including the digital divide and regulatory compliance. A mixed-methods approach was adopted. First, SLR was used to identify, screen, and synthesize 28 core studies. Then multiple case studies were used to analyse real-world blockchain-voting system implementations to highlight practical successes and setbacks. From this, a conceptual framework for a hybrid blockchain voting system that integrates smart contracts, layered consensus model, and an off-chain data-science layer for real-time monitoring was developed. Findings show consensus that blockchain can enhance election integrity, and data-science techniques further strengthen authentication, detect intrusions, and enable privacy-preserving analytics. However, it was also found that legal and regulatory gaps, infrastructure and literacy barriers, lack of scalability, and the need to build public trust remain huge hindrances to widespread adoption. The thesis recommended a hybrid voting architecture that combines public and private blockchains and off-chain data science monitoring, and its interface is user-friendly. Policymakers and election administrators are urged to pilot such integrated systems, refine identity-management protocols, and invest in voter education to ensure both technical robustness and broad societal acceptance, thereby paving the way toward more secure, transparent, and efficient democratic processes.
Online voting promises greater convenience and accessibility, but moving from supervised polling places to unsupervised settings magnifies the risk of coercion and vote buying. A compelling strategy is to give voters fake credentials: credentials that look and behave like real voting credentials but whose ballots are silently excluded from the tally. Despite its conceptual appeal, practical realizations and usability evidence for fake credentials have remained limited. This dissertation presents Votegral, the first end-to-end verifiable, coercion-resistant online voting system with empirical evidence towards practical usability. Votegral has two components: TRIP and VLT. TRIP is a trust-limited, in-person registration scheme that issues voters a real credential and any number of fake credentials on paper, without trusted hardware. TRIP embeds an interactive zero-knowledge proof into the physical printing process so that real credentials carry sound proof transcripts while fake credentials carry identically formatted but unsound proof transcripts -- distinguishable only by the voter during issuance and not transferable thereafter. VLT is a tallying scheme that constrains ballots to registrar-issued credentials to enable linear-time filtering of fake ballots. VLT also introduces standing votes: a voter facing extreme coercion can, at registration, delegate their voting rights to a publicly registered political party and leave the booth with only fake credentials. Tallying then credits the party's ballot by the number of such delegations and publishes publicly auditable proofs, resulting in both transparency and coercion evidence -- evidence that an aggregate number of voters felt unsafe to leave the registrar with a real credential. Our prototype tallies 1 million ballots in about 14 hours on a 128 core, 256 GB RAM machine; this puts Votegral on par with modern end-to-end verifiable systems such as Swiss Post, while significantly outperforming prior JCJ-style systems such as Civitas. TRIP's end-to-end, voter-observable registration session completes in under 20 seconds on resource-constrained hardware. In our main user study with 150 demographically diverse participants recruited in Boston, Massachusetts, 83% successfully registered and cast a ballot in our mock election. Among the 120 participants exposed to fake credentials, 96% correctly understood the purpose of fake credentials. These promising results suggest a path for practical viability of coercion-resistant, end-to-end verifiable online voting using fake credentials.
Electronic cash (e-cash) systems must provide users with anonymity while preventing criminal misuse and enabling selective tracing of malicious actors, such as double-spenders. One of the most ingenious solutions, introduced by Chaum and Brands, involves deploying observers on users’ devices to enforce legitimate payment behavior without compromising anonymity. In this paper, we propose a novel e-cash scheme based on BBS+ signatures and open-source observers running within Attested Execution Secure Processors (AESPs). Our construction employs zero-knowledge proofs to conceal private information, while the open-source observer program—installed within a tamper-proof secure element—acts as a proactive gatekeeper, preventing double-spend attempts before they succeed. The AESP-based attestation mechanism binds each output to the program’s hash, guaranteeing that the officially vetted observer is indeed installed and operating unmodified on the user’s device. This scheme provides three key properties. First, it achieves anonymity and one-more unforgeability using a technique inspired by Brickell et al.’s EPID. Second, it ensures robust double-spend prevention and traceability of double-spenders, even if the AESP assumption is compromised. Third, the scheme remains reversible: if the observer is compromised, it gracefully reverts to the basic scheme without the observer, which preserves anonymity and unforgeability. Overall, this work bridges cryptographic privacy and practical regulatory oversight, providing a transparent and extensible foundation for secure digital payments.
Lukman Adewale Ajao, Buhari Ugbede Umar, Henry Ohiani Ohize, Eustace M. Dogo · 6 authors
The prevalence of political interference during election processes remains a significant challenge to the free, fair, and credible conduct of general elections. This election event is crucial as a pillar of democratic governance in any democracy, with the potential for unfitting disturbance and chaos, such as multiple votes, illicit voting, and malicious actors that can be exploited electronically to disrupt voting or affect vote counts. However, this research proposed to develop a secure, smart, verifiable e-voting system (SSVEVS) that can offer an authenticated end-to-end tally voting system, a top-secret ballot election system, and confidence in overall election integrity. The proof of this smart and secure electronic voting system utilizes a 64-bit quad-core ARM Cortex-A76 processor, integrated with multimodal biometric (facial and fingerprint) authentication systems, programmed with a deep-learning image processing (DLIP) algorithm to optimize image detection and recognition. Also, an Ethereum blockchain technology (EBT) with a homomorphic encryption algorithm was implemented on the system to ensure that the original information record is maintained, immutable, tamper-resistant, and transparent throughout the electoral process, and stores the information in a decentralized database application. The system achieved 1,424.501 TPS for 10,000 transactions with a mining time of 7.02 seconds. The biometric (facial and fingerprint) authentication achieved a False Acceptance Rate (FAR) and False Rejection Rate (FRR) of 0.00% respectively. The True Acceptance Rate (TAR) is 100%, with a false image template of 0.04% true identification rate (TPIR).
Open access
Internet Traffic Analysis and Secure E-voting
Advanced Steganography and Watermarking Techniques
Blockchain technology has been envisioned as an emerging facilitator of auditable, transparent, and secure electronic voting (e-voting) systems to overcome issues with traditional and electronic voting systems. However, preserving data integrity, offering voter privacy, and scalability in blockchainbased e-voting systems are persistent issues. In this systematic literature review of peer-reviewed research articles from 2018 to 2025, this paper explores cryptographic schemes, architecture designs for blockchain-based e-voting systems, and solutions for scalability. By taking an PRISMA-congruent structured research methodology approach, nine core studies are reviewed to discuss Zero Knowledge Proofs and blind signature schemes for maintaining privacy conservation, blockchain immutability to maintain integrity, and layer-2 scaling solutions to bypass throughput bottlenecks. Conclusions suggest that although transparency and audita-bility are elevated with applications of blockchain technology, implementation for massive-scale elections remains in its nascent stage and requires development in privacypreservation cryptographic schemes and scalable architecture designs. As a review paper, it compiles an updated summary of the status of the landscape of blockchain-based e-voting systems and highlights existing knowledge gaps and proposes research directions for developing secure, scalable, and privacy-respecting digital elections.
Open access
3 source records
Blockchain Technology Applications and Security
Internet Traffic Analysis and Secure E-voting
Advanced Steganography and Watermarking Techniques
Jan 1, 2025·Proceedings of the 5th LACCEI International Multiconference on Entrepreneurship, Innovation and Regional Development (LEIRD 2025): "Entrepreneurship with Purpose: Social and Technological Innovation in the Age of AI"
Jeremies Enmanuel Chinchay Camargo, Massiel Fiorella Parvina Huaman, Carmen Luz Cuba Cornejo, Cesar Augusto Cabrera Garcia
Digital electoral security has become fundamental to the development of reliable, integrated and available technological systems, driven by the growing demand for transparency and protection against threats. The purpose of this study is to analyze the impact of cryptographic protocols on the security of electoral processes, evaluating their effectiveness against traditional methods. For this purpose, a systematic review of the literature was carried out, considering 50 articles extracted from the Scopus database. The analysis focused on cryptographic techniques applied to blockchain-based environments, such as homomorphic encryption, zero-knowledge proofs and smart contracts, evaluating their contribution to design more secure, auditable and reliable voting systems. The results show that these protocols contribute to prevent recurring vulnerabilities, such as vote tampering, electoral fraud, impersonation and lack of validation, in addition to strengthening auditability and operational reliabilityFinally, the study concludes that the adoption and assessment of cryptographic protocols are essential to reduce risks in electronic voting, and promote more secure, transparent and efficient electoral processes.
Federated Learning (FL) is a decentralized learning method that enables collaborative model training while preserving data privacy. This makes FL a promising solution in various applications, particularly in cross-silo settings such as healthcare, finance, and transportation. However, FL remains highly vulnerable to adversarial threats, especially backdoor attacks, where malicious clients inject poisoned data to manipulate global model behavior. Existing outlier detection techniques often struggle to effectively isolate such adversarial updates, compromising model integrity. To address this challenge, we propose Backdoor Attack Resilient Technique for Federated Learning (BART-FL), a novel lightweight defense mechanism that enhances FL security through malicious client filtering. Our method integrates Principal Component Analysis (PCA) for dimensionality reduction with cosine similarity for measuring pairwise distances between model updates andK-means clustering for detecting potentially malicious clients. To reliably identify the benign cluster, we introduce a multi-metric statistical voting mechanism based on point-level mean, median absolute deviation (MAD), and cluster-level mean. This approach strengthens model resilience against adversarial manipulations by identifying and filtering malicious updates before aggregation, thereby preserving the integrity of the global model. Experimental evaluations conducted on the LISA traffic light dataset, CIFAR-10, and CIFAR-100 demonstrate the effectiveness of BART-FL in maintaining model performance across diverse FL settings. Additionally, we perform a comparative analysis against existing backdoor defense techniques, highlighting BART-FL’s ability to improve security while ensuring computational efficiency. Our results showcase the potential of BART-FL as a scalable and adversary-resilient defense mechanism for secure training in cross-silo FL applications.
The end-to-end verifiable e-voting system Ordinos [26] is primarily characterized by its tally-hiding property, which ensures that only the actual election result, e. g., the winner of the election, is revealed while the full tally consisting of the aggregated votes stays hidden. Ordinos is an abstract model that guarantees tally-hiding, verifiability and vote privacy if the underlying cryptographic primitives satisfy certain requirements. It uses a multi-party-computation protocol over an additively homomorphic encryption scheme and guarantees active security with zero-knowledge proofs. Ordinos has already been instantiated for several election systems using the Paillier [35] encryption scheme, which can be broken by Shor’s algorithm [41]. The aim of this thesis is to instantiate Ordinos post-quantum secure using a variant of Regev’s LWE-based cryptosystem [39], which is adapted to realize an actively secure threshold encryption scheme over an arbitrary plaintext space. Then a noise analysis of the arithmetic and logical components used in the MPC-protocol of the Paillier instantiation is conducted, and the components are slightly adapted to restrict the noise growth. Additionally, valid zero-knowledge proofs are provided and a concrete instantiation achieving a security level of 128 bits is shown.
The end-to-end verifiable e-voting system Ordinos [26] is primarily characterized by its tally-hiding property, which ensures that only the actual election result, e. g., the winner of the election, is revealed while the full tally consisting of the aggregated votes stays hidden. Ordinos is an abstract model that guarantees tally-hiding, verifiability and vote privacy if the underlying cryptographic primitives satisfy certain requirements. It uses a multi-party-computation protocol over an additively homomorphic encryption scheme and guarantees active security with zero-knowledge proofs. Ordinos has already been instantiated for several election systems using the Paillier [35] encryption scheme, which can be broken by Shor’s algorithm [41]. The aim of this thesis is to instantiate Ordinos post-quantum secure using a variant of Regev’s LWE-based cryptosystem [39], which is adapted to realize an actively secure threshold encryption scheme over an arbitrary plaintext space. Then a noise analysis of the arithmetic and logical components used in the MPC-protocol of the Paillier instantiation is conducted, and the components are slightly adapted to restrict the noise growth. Additionally, valid zero-knowledge proofs are provided and a concrete instantiation achieving a security level of 128 bits is shown.
Fraud risks are on the rise with the increase in cryptocurrency transactions; the traditional detection methods become inadequate. This paper proposes a hybrid deep-learning framework for real-time fraud detection in Bitcoin transactions. Algorithms like Random Forest, support vector machine (SVM), Logistic Regression, and XGBoost are used to analyse transaction patterns and anomalies with a high level of accuracy. Different models have been tested in transaction data for Bitcoin, and the best-performing model was XGBoost with an accuracy of 96.94%. Advanced machine learning techniques enrich a system through secure data-driven insights and real-time anomaly detection, thus enhancing fraud risk detection. The obstacles faced are scalability, privacy issues, and inability to adjust models according to the evolving fraud technique. Future advancements might deal with federated learning, encryption methods, and cross-platform prevention to make the detection of fraud more secure. This paper indicates how well deep learning-based detection of fraud can scale and work efficiently to strengthen trust in digital financial systems.
Brugeres, Maxence, Languille, Victor, Kuznetsov, Petr, Zarfaoui, Hamza
We propose a decentralized asset-transfer system that enjoys full privacy: no party can learn the details of a transaction, except for its issuer and its recipient. Furthermore, the recipient is not aware of the sender’s identity. Our system does not rely on consensus or synchrony assumptions, and therefore, it is responsive, since it runs at the actual network speed. Under the hood, every transaction creates a consumable coin equipped with a non-interactive zero-knowledge proof (NIZK) that confirms that the issuer has sufficient funds without revealing any information about her identity, the recipient’s identity, or the payment amount. Moreover, we equip our system with a regulatory enforcement mechanism that can be used to regulate transfer limits or restrict specific addresses from sending or receiving funds, while preserving the system’s privacy guarantees. Finally, we report on PaxPay, our implementation of Fully Private Asset Transfer (FPAT) that uses the Gnark library for the NIZKs. In our benchmark, PaxPay exhibits better performance than earlier proposals that either ensure only partial privacy, require some kind of network synchrony or do not implement regulation features. Our system thus reconciles privacy, responsiveness, regulation enforcement and performance.
Mosbah Alown, Mehmet Sabır Kiraz, Muhammed Ali Bingöl
Electronic voting (e-voting) systems have significantly improved the traditional voting process by addressing key concerns such as security, public acceptability, and convenience. However, these systems often face unique challenges, such as ensuring voter privacy and verifiability, preventing coercion and double voting, and maintaining scalability while protecting participant confidentiality. This study critically analyses and compares various e-voting schemes and technologies, evaluating their security features, verifiability mechanisms, and potential vulnerabilities. This paper reviews Direct Recording Electronic (DRE) voting, internet voting, and blockchain-based e-voting systems. In so doing, we provide an understanding of cryptographic primitives employed in e-voting systems and how they address specific characteristics and challenges associated with each voting scheme. Furthermore, we examine the applications proposed by previous studies in the context of these voting systems, assessing their strengths, limitations, and impact on democratic procedures. The cryptographic primitives reviewed include techniques like homomorphic encryption, blind signatures, and zero-knowledge proofs, which can enhance voter privacy, verifiability, and resistance to coercion and double voting.
Edward Danso Ansong, Simon Bonsu Osei, Raphael Adjetey Adjei
The surge in identity fraud, driven by the rapid adoption of mobile money, internet banking, and e-services during the COVID-19 pandemic, underscores the need for robust cybersecurity solutions. Zero-Knowledge Proofs (ZKPs) e... | Find, read and cite all the research you need on Tech Science Press
Zubaida Rehman, Iqbal Gondal, Hai Dong, Mengmeng Ge · 6 authors
Eclipse attacks, which isolate victim nodes by monopolizing their peer connections, remain a critical threat to Ethereum’s consensus mechanism. To address this, we present a principled framework for detecting Eclipse attacks in Ethereum peer-to-peer networks, grounded in a formal adversarial model. Existing defenses are either ad-hoc or lack provable guarantees, leaving open questions about their reliability under adaptive adversaries. Our work aims to bridge this gap by formally defining eclipse attack detection as a security property. We specify soundness, completeness, and robustness theorems under bounded adversarial drift, and derive formal guarantees within false positive and false negative bounds, resilience to adversarial manipulation, and multi-node compositional reliability. We then instantiate a lightweight detection framework that maps packet-level traffic features to predictions using ensemble classifiers (Random Forest, XGBoost). The system was validated using a controlled Ethereum testbed and extended with CTGAN-generated synthetic traces to emulate networks of up to 100 nodes. Empirical evaluation shows that our framework achieves up to 96% F1-score with sub-second inference latency, well within Ethereum’s 12-second Proof-of-Stake validator time slots. These findings demonstrate that lightweight statistical features, when coupled with formal analysis, enable accurate, efficient, and scalable detection of network-level partitioning attacks. Our work establishes a deployable and theoretically grounded defense foundation for securing modern blockchain systems against eclipse adversaries.