Simeon Okechukwu Ajakwe
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Simeon Okechukwu Ajakwe has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Dong-Seong Kim
ICT Convergence Research Center, Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
24 shared publications - Jae-Min Lee
Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
14 shared publications - Ihunanya Udodiri Ajakwe
Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
6 shared publications - Victor Ikenna Kanu
Kumoh National Institute of Technology
5 shared publications - Chigozie Athanasius Nnadiekwe
Department of IT Convergence Engineering, Kumoh National Institute of Technology, Gumi, South Korea
3 shared publications - Opeyemi. O. Deji-Oloruntoba
Biohealth Convergence Unit, Food and Drug Biotechnology, Inje University, Gimhae 50834, Republic of Korea
3 shared publications
Latest Publications
- Open Access
- 17 November 2026
- Published by Elsevier in High-Confidence Computing
PureSearch: A search-centric hybrid blockchain framework with formal latency separation for healthcare and military logistics
- 17 November 2026
- Published by Elsevier in High-Confidence Computing
Open AccessPureSearch: A search-centric hybrid blockchain framework with formal latency separation for healthcare and military logistics
- Open Access
- 10 September 2026
- Published by Frontiers Media SA in Frontiers in Future Transportation
- Vol. 7, 1923890
Explainable and cyber-resilient vision transformers for joint steering and collision-avoidance perception in autonomous drone navigation: a multi-task Grad-CAM framework
Micro aerial vehicles (MAVs) flying through indoor corridors and other GNSS-denied environments cannot rely on satellite positioning, so a single monocular RGB camera is often the only sensing modality available. The navigation objective addressed in this paper is reactive, map-free corridor following: from each camera frame the onboard perception stack must choose a discrete steering direction that keeps the vehicle advancing along the corridor while avoiding walls and obstacles, rather than tracking a pre-specified waypoint, route, or global path. Prior work (DroNet, PULP-DroNet) addresses this problem with lightweight convolutional networks that map camera frames directly to a steering command and a collision flag; this paper asks whether a Vision Transformer can perform the same mapping with higher accuracy while remaining auditable. We present ViTDrone, an explainable artificial intelligence (XAI) framework that couples a Vision Transformer (ViT-Base/16) backbone with multi-task learning and gradient-weighted class activation mapping (Grad-CAM). From a single shared backbone the model simultaneously predicts a three-class steering command (left/forward/right) and a binary collision label. We build a scalable data pipeline that aggregates 309 acquisition sequences of the PULP-DroNet v3 benchmark, yielding 195,455 training and 76,390 test frames annotated with yaw-rate commands and collision risk. After five epochs of AdamW fine-tuning from ImageNet-21k weights, ViTDrone attains 76.35% yaw classification accuracy and 60.77% collision accuracy. To make the model auditable we adapt Grad-CAM to the transformer with a reshape transform over the patch tokens. The resulting saliency maps are reassuring: the network looks where a human pilot would, at floor-wall boundaries when it decides to turn and at overhead structure when it senses a collision. However, aggregate accuracy conceals substantial variation across classes. A per-class analysis, combining loss curves, confusion matrices, and a comparison against convolutional baselines, shows that the reported accuracies are driven disproportionately by the two majority classes (forward steering and no-collision), while minority classes, sharp turns and imminent collisions, are recognized far less reliably. We attribute this shortfall to two measurable causes rather than to insufficient backbone capacity: severe label imbalance (78.6% forward, 60.7% no-collision) and a persistent gap between training and validation loss that is consistent with distribution shift across acquisition sequences. Relative to prior work, which has evaluated either convolutional navigation policies without explainability or Vision Transformer explainability outside the UAV navigation setting, this paper contributes (i) to our knowledge, the first integration of transformer-native Grad-CAM explainability with a multi-task steering-and-collision policy for MAV navigation, and (ii) a systematic diagnosis showing that class imbalance and domain shift, rather than model capacity, are the dominant factors limiting both accuracy and interpretability in this setting, together with a concrete research agenda for addressing them.
- 10 September 2026
- Published by Frontiers Media SA in Frontiers in Future Transportation
- Vol. 7, 1923890
Open AccessExplainable and cyber-resilient vision transformers for joint steering and collision-avoidance perception in autonomous drone navigation: a multi-task Grad-CAM framework
Micro aerial vehicles (MAVs) flying through indoor corridors and other GNSS-denied environments cannot rely on satellite positioning, so a single monocular RGB camera is often the only sensing modality available. The navigation objective addressed in this paper is reactive, map-free corridor following: from each camera frame the onboard perception stack must choose a discrete steering direction that keeps the vehicle advancing along the corridor while avoiding walls and obstacles, rather than tracking a pre-specified waypoint, route, or global path. Prior work (DroNet, PULP-DroNet) addresses this problem with lightweight convolutional networks that map camera frames directly to a steering command and a collision flag; this paper asks whether a Vision Transformer can perform the same mapping with higher accuracy while remaining auditable. We present ViTDrone, an explainable artificial intelligence (XAI) framework that couples a Vision Transformer (ViT-Base/16) backbone with multi-task learning and gradient-weighted class activation mapping (Grad-CAM). From a single shared backbone the model simultaneously predicts a three-class steering command (left/forward/right) and a binary collision label. We build a scalable data pipeline that aggregates 309 acquisition sequences of the PULP-DroNet v3 benchmark, yielding 195,455 training and 76,390 test frames annotated with yaw-rate commands and collision risk. After five epochs of AdamW fine-tuning from ImageNet-21k weights, ViTDrone attains 76.35% yaw classification accuracy and 60.77% collision accuracy. To make the model auditable we adapt Grad-CAM to the transformer with a reshape transform over the patch tokens. The resulting saliency maps are reassuring: the network looks where a human pilot would, at floor-wall boundaries when it decides to turn and at overhead structure when it senses a collision. However, aggregate accuracy conceals substantial variation across classes. A per-class analysis, combining loss curves, confusion matrices, and a comparison against convolutional baselines, shows that the reported accuracies are driven disproportionately by the two majority classes (forward steering and no-collision), while minority classes, sharp turns and imminent collisions, are recognized far less reliably. We attribute this shortfall to two measurable causes rather than to insufficient backbone capacity: severe label imbalance (78.6% forward, 60.7% no-collision) and a persistent gap between training and validation loss that is consistent with distribution shift across acquisition sequences. Relative to prior work, which has evaluated either convolutional navigation policies without explainability or Vision Transformer explainability outside the UAV navigation setting, this paper contributes (i) to our knowledge, the first integration of transformer-native Grad-CAM explainability with a multi-task steering-and-collision policy for MAV navigation, and (ii) a systematic diagnosis showing that class imbalance and domain shift, rather than model capacity, are the dominant factors limiting both accuracy and interpretability in this setting, together with a concrete research agenda for addressing them.
- Open Access
- 29 June 2026
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 139
DRIVERDAPP: Driver’s distraction record using deep learning and blockchain
- 29 June 2026
- Published by Elsevier in Computers and Electrical Engineering
- Vol. 139
Open AccessDRIVERDAPP: Driver’s distraction record using deep learning and blockchain
- Open Access
- 28 May 2026
- Published by Elsevier in ICT Express
- Vol. 12 (5) , 1130-1151
Security-critical data exchange in digital twin ecosystems: Hybrid blockchain architectures and design trade-offs
- 28 May 2026
- Published by Elsevier in ICT Express
- Vol. 12 (5) , 1130-1151
Open AccessSecurity-critical data exchange in digital twin ecosystems: Hybrid blockchain architectures and design trade-offs
- Open Access
- 2 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. PP (20) , 1
XS-FedPRS: Explainable and Secure Federated Learning for Privacy-Preserving Genomic Analytics in Federated Healthcare Networks
In federated healthcare networks (FHNs), collaborative learning enables distributed genomic intelligence across hospital sites and clinical institutions without centralizing sensitive patient data. However, federated polygenic risk prediction (PRS) across heterogeneous healthcare networks faces three critical challenges: population-stratified rare variant distributions that diverge across clinical sites, byzantine attacks mimicking rare variant signals, and lack of interpretability for clinical decision support. We present XS-FedPRS, a framework that decouples common variants (global aggregation) from rare variants (population-specific clustering), defends against byzantine attacks through Hardy-Weinberg equilibrium testing and gradient plausibility analysis ,and provides multi-level explainability for clinical decision support. Proof-of-Authority-and-Association (PoA2) ensures auditable governance across participating sites. Evaluation across three genomic cohorts of increasing scale and realism—CINECA synthetic (483 samples), 1000 Genomes Phase 3 (2,504 samples, 5 continental populations), and msprime-simulated (50,000 samples, Out-of-Africa demography)—demonstrates that XS-FedPRS achieves 84.4% AUC on the largest cohort under population-stratified non-IID partitioning, detects gradient poisoning, model poisoning, sybil, and combined byzantine attacks with perfect precision and recall (F1=1.0), maintains privacy (0.026 MIA advantage), and incurs only 3.3% communication overhead versus FedAvg. XS-FedPRS enables trustworthy, interpretable federated genomic analytics across heterogeneous healthcare networks devices for precision medicine.
- 2 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. PP (20) , 1
Open AccessXS-FedPRS: Explainable and Secure Federated Learning for Privacy-Preserving Genomic Analytics in Federated Healthcare Networks
In federated healthcare networks (FHNs), collaborative learning enables distributed genomic intelligence across hospital sites and clinical institutions without centralizing sensitive patient data. However, federated polygenic risk prediction (PRS) across heterogeneous healthcare networks faces three critical challenges: population-stratified rare variant distributions that diverge across clinical sites, byzantine attacks mimicking rare variant signals, and lack of interpretability for clinical decision support. We present XS-FedPRS, a framework that decouples common variants (global aggregation) from rare variants (population-specific clustering), defends against byzantine attacks through Hardy-Weinberg equilibrium testing and gradient plausibility analysis ,and provides multi-level explainability for clinical decision support. Proof-of-Authority-and-Association (PoA2) ensures auditable governance across participating sites. Evaluation across three genomic cohorts of increasing scale and realism—CINECA synthetic (483 samples), 1000 Genomes Phase 3 (2,504 samples, 5 continental populations), and msprime-simulated (50,000 samples, Out-of-Africa demography)—demonstrates that XS-FedPRS achieves 84.4% AUC on the largest cohort under population-stratified non-IID partitioning, detects gradient poisoning, model poisoning, sybil, and combined byzantine attacks with perfect precision and recall (F1=1.0), maintains privacy (0.026 MIA advantage), and incurs only 3.3% communication overhead versus FedAvg. XS-FedPRS enables trustworthy, interpretable federated genomic analytics across heterogeneous healthcare networks devices for precision medicine.
- Open Access
- 28 March 2026
- Published by Elsevier in Franklin Open
- Vol. 15
Lightweight data security system for resource-constrained IIoT devices
- 28 March 2026
- Published by Elsevier in Franklin Open
- Vol. 15
Open AccessLightweight data security system for resource-constrained IIoT devices
- Open Access
- 26 February 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3615
ProtChain: a blockchain-enabled framework for secure and reproducible biomedical IoT data workflows
The rapid growth of biomedical Internet of Things (IoT) systems has intensified the challenge of ensuring secure, interoperable, and reproducible data workflows across distributed devices, laboratories, and repositories. Conventional repositories such as the Protein Data Bank (PDB) and UniProt provide standardized data formats, yet they lack scalable mechanisms for real-time synchronization, provenance verification, and trusted sharing in IoT-driven environments. This article presents ProtChain, a blockchain-enabled framework that integrates a permissioned PureChain ledger with decentralized storage (InterPlanetary File System (IPFS)) to deliver secure and high-throughput biomedical workflows. Designed with a microservices architecture, ProtChain enforces Advanced Encryption Standard (AES)-256 encryption, role-based access control, and seamless application programming interface (API) connectivity with bioinformatics repositories. A prototype case study using the HIV-1 protease (1HPV) demonstrates end-to-end ingestion, provenance verification, and retrieval of proteomic data. Performance evaluations show throughput of 485.96 requests per second under 50 concurrent workloads, with blockchain transaction rates reaching 445 transactions per second (TPS) at 10–25 transactions per block, validating scalability for IoT-based biomedical research. By combining blockchain immutability with decentralized storage and standardized APIs, ProtChain addresses pressing challenges of interoperability, data trust, and reproducibility. The results establish ProtChain as a trust backbone for next-generation biomedical IoT ecosystems, enabling secure collaboration, verifiable research pipelines, and scalable healthcare applications.
- 26 February 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3615
Open AccessProtChain: a blockchain-enabled framework for secure and reproducible biomedical IoT data workflows
The rapid growth of biomedical Internet of Things (IoT) systems has intensified the challenge of ensuring secure, interoperable, and reproducible data workflows across distributed devices, laboratories, and repositories. Conventional repositories such as the Protein Data Bank (PDB) and UniProt provide standardized data formats, yet they lack scalable mechanisms for real-time synchronization, provenance verification, and trusted sharing in IoT-driven environments. This article presents ProtChain, a blockchain-enabled framework that integrates a permissioned PureChain ledger with decentralized storage (InterPlanetary File System (IPFS)) to deliver secure and high-throughput biomedical workflows. Designed with a microservices architecture, ProtChain enforces Advanced Encryption Standard (AES)-256 encryption, role-based access control, and seamless application programming interface (API) connectivity with bioinformatics repositories. A prototype case study using the HIV-1 protease (1HPV) demonstrates end-to-end ingestion, provenance verification, and retrieval of proteomic data. Performance evaluations show throughput of 485.96 requests per second under 50 concurrent workloads, with blockchain transaction rates reaching 445 transactions per second (TPS) at 10–25 transactions per block, validating scalability for IoT-based biomedical research. By combining blockchain immutability with decentralized storage and standardized APIs, ProtChain addresses pressing challenges of interoperability, data trust, and reproducibility. The results establish ProtChain as a trust backbone for next-generation biomedical IoT ecosystems, enabling secure collaboration, verifiable research pipelines, and scalable healthcare applications.
- Open Access
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (9) , 18544-18561
Offline-Capable AI–Blockchain Architecture for Biochemical Threat Detection in Mission-Critical MANET Environments
Biochemical threats remain a serious concern in mission-critical environments, particularly those characterized by intermittent connectivity and infrastructure degradation. Traditional centralized detection systems are ill-suited for such conditions, as they depend on stable communication channels and are inherently vulnerable to cyber–physical disruptions. This work introduces a decentralized solution integrating artificial intelligence (AI) and blockchain (BC) for autonomous biochemical threat detection within tactical mobile ad hoc networks (MANETs). The framework uses a random forest (RF) classifier trained on acetylcholinesterase (AChE) sensor data to identify sarin exposure with 100% accuracy and sub-25-ms inference latency. Threat verification is secured using a lightweight proof of authority and association ( $\mathtt {PoA^{2}}$ ) BC, which provides tamper-resistant logging and distributed consensus. The architecture supports offline operations and maintains functionality under conditions of 20% packet loss and node disruption. Simulations conducted in degraded network environments confirmed the system’s robustness and scalability, establishing it as a resilient and efficient platform for secure biochemical threat detection in dynamic, resource-constrained mission-critical settings.
- 11 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (9) , 18544-18561
Open AccessOffline-Capable AI–Blockchain Architecture for Biochemical Threat Detection in Mission-Critical MANET Environments
Biochemical threats remain a serious concern in mission-critical environments, particularly those characterized by intermittent connectivity and infrastructure degradation. Traditional centralized detection systems are ill-suited for such conditions, as they depend on stable communication channels and are inherently vulnerable to cyber–physical disruptions. This work introduces a decentralized solution integrating artificial intelligence (AI) and blockchain (BC) for autonomous biochemical threat detection within tactical mobile ad hoc networks (MANETs). The framework uses a random forest (RF) classifier trained on acetylcholinesterase (AChE) sensor data to identify sarin exposure with 100% accuracy and sub-25-ms inference latency. Threat verification is secured using a lightweight proof of authority and association ( $\mathtt {PoA^{2}}$ ) BC, which provides tamper-resistant logging and distributed consensus. The architecture supports offline operations and maintains functionality under conditions of 20% packet loss and node disruption. Simulations conducted in degraded network environments confirmed the system’s robustness and scalability, establishing it as a resilient and efficient platform for secure biochemical threat detection in dynamic, resource-constrained mission-critical settings.
- Open Access
Symmetric–Asymmetric Security Synergy: A Quantum-Resilient Hybrid Blockchain Framework for Incognito IoT Data Sharing
Secure and auditable data sharing in large-scale Internet of Things (IoT) environments remains a significant challenge due to weak trust coordination, limited scalability, and susceptibility to emerging quantum attacks. This study introduces a hybrid blockchain-based framework that integrates post-quantum cryptography with intelligent anomaly detection to ensure end-to-end data integrity and resilience. The proposed system utilizes Hyperledger Fabric for permissioned device lifecycle management and Ethereum for public auditability of encrypted telemetry, thereby providing both private control and transparent verification. Device identities are established using quantum-entropy-seeded credentials and safeguarded with lattice-based encryption to withstand quantum adversaries. A convolutional long short-term memory (CNN–LSTM) model continuously monitors device behavior, facilitating real-time trust scoring and autonomous revocation via smart contract triggers. Experimental results demonstrate 97.4% anomaly detection accuracy and a 0.968 F1-score, supporting up to 1000 transactions per second with cross-chain latency below 6 s. These findings indicate that the proposed architecture delivers scalable, quantum-resilient, and computationally efficient data sharing suitable for mission-critical IoT deployments.
Open AccessSymmetric–Asymmetric Security Synergy: A Quantum-Resilient Hybrid Blockchain Framework for Incognito IoT Data Sharing
Secure and auditable data sharing in large-scale Internet of Things (IoT) environments remains a significant challenge due to weak trust coordination, limited scalability, and susceptibility to emerging quantum attacks. This study introduces a hybrid blockchain-based framework that integrates post-quantum cryptography with intelligent anomaly detection to ensure end-to-end data integrity and resilience. The proposed system utilizes Hyperledger Fabric for permissioned device lifecycle management and Ethereum for public auditability of encrypted telemetry, thereby providing both private control and transparent verification. Device identities are established using quantum-entropy-seeded credentials and safeguarded with lattice-based encryption to withstand quantum adversaries. A convolutional long short-term memory (CNN–LSTM) model continuously monitors device behavior, facilitating real-time trust scoring and autonomous revocation via smart contract triggers. Experimental results demonstrate 97.4% anomaly detection accuracy and a 0.968 F1-score, supporting up to 1000 transactions per second with cross-chain latency below 6 s. These findings indicate that the proposed architecture delivers scalable, quantum-resilient, and computationally efficient data sharing suitable for mission-critical IoT deployments.