DODaniel Onwuchekwa

Daniel Onwuchekwa

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Daniel Onwuchekwa has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Hierarchical Scheduler with Adaptive Time-Budget Reallocation for Time-Triggered Edge-Fog-Cloud Architectures

    by Omar Hekal, Josepaul Paulachan, Daniel Onwuchekwa, Roman Obermaisser

    The lack of determinism restricts the integration of safety-critical applications into Edge–Fog–Cloud (EFC) architectures. Existing EFC schedulers are typically designed for dynamic, best-effort operation based on unmanaged resource allocation and elastic virtualization. This paradigm introduces unbounded queueing, resource contention, and timing jitter, making standard schedulers unsuitable for hard-deadline workloads. Moreover, most approaches focus on computational placement, while communication is abstracted or treated as a secondary cost term. As a result, bounded-latency routing and deterministic task execution are rarely co-optimized under a unified timing model. This paper addresses these gaps by utilizing a managed Time-Triggered Edge–Fog–Cloud (TTEFC) architecture that supports safety-critical workloads, orchestrates IEEE Time-Sensitive Networking (TSN) for local intra-domain communication, and uses IETF Deterministic Networking (DetNet) for routed inter-domain paths. On this infrastructure, a hierarchical genetic algorithm (HGA) is proposed to jointly schedule partition-to-execution-location allocation, partition execution order, inter-partition route selection, and negotiated per-partition time budgets that act as temporal boundaries for parallel partition-level optimizers. An adaptive slack reallocation operator redistributes unused temporal slack from over-satisfied partitions to budget-violating partitions, improving feasibility convergence. Experiments on synthetic DAG workloads with 100–500 tasks compare the proposed HGA against HEFT and round-robin baselines. These baselines are included as scoped external references to contextualize the end-to-end scheduling performance of the proposed method. Ablation results show that slack reallocation improves partition-budget feasibility, reaches feasible budget assignments earlier, and produces tighter budget–makespan alignment than feedback-free and static-budget variants. An automotive-characteristic DAG case study further evaluates the method on an application-oriented workload under the same timing and communication assumptions.

  • Open Access

    Leveraging Multi-Cycle Information to Improve Event Prediction in Time-Triggered Systems

    by Dongchen Li, Daniel Onwuchekwa, Roman Obermaisser

    In adaptive time-triggered systems, dynamic schedule switching in response to environmental and internal state changes is essential. However, existing approaches relying on offline-generated Multiple Scheduling Graphs (MSGs) often suffer from high memory overhead, whereas online schedule generation incurs adaptation latency. To address both limitations, proactive speculative scheduling with event prediction has emerged as a promising solution. Existing event prediction models rely on single-cycle input, which may lack sufficient temporal context. In this work, we argue that multi-cycle inputs capture richer temporal patterns, particularly periodic relationships, that can significantly improve prediction performance. We propose the Slack Time and Event Prediction Model (SEPM), a novel attention-based model designed to effectively learn from long multi-cycle inputs. Through extensive experiments, we show that SEPM substantially outperforms three baseline models in terms of prediction accuracy and stability. Moreover, we identify that the number of input cycles is a critical hyperparameter. Although exhaustive search yields optimal results, it is computationally expensive. To this end, we introduce a lightweight FFT-based slack time analysis method, which efficiently estimates the effective number of input cycles closely matching those obtained from experiments. On the test dataset, SEPM achieves a median event coverage of 80.02%, significantly outperforming its single-cycle variant at 74.81%, demonstrating the substantial benefit of multi-cycle inputs. These findings validate the advantage of multi-cycle inputs and attention mechanisms in event sequence prediction. This work establishes a practical foundation for building efficient and robust predictive components in adaptive time-triggered systems.

  • Open Access

    A Time-Triggered Edge–Fog–Cloud Architecture With Hierarchical Genetic Optimization for Safety-Critical Applications

    by Josepaul Paulachan, Omar Hekal, Daniel Onwuchekwa, Roman Obermaisser

    The growth of safety-critical Internet of Things (IoT) applications, such as autonomous driving and industrial automation, demands predictable, real-time performance from distributed Edge-Fog-Cloud systems. Existing scheduling algorithms for this continuum are fundamentally unsuitable for safety-critical domains. These algorithms primarily use best-effort communication networks. This results in non-deterministic latencies and an inability to provide temporal guarantees. This paper introduces a new solution that bridges the gap between task scheduling and network determinism: a Time-Triggered Edge-Fog-Cloud (TTEFC) architecture that is based on Time-Sensitive Networking (TSN) and Deterministic Networking (DetNet) standards, ensuring bounded end-to-end communication delays; and a Hierarchical Nested Genetic Algorithm (HNGA) that effectively maps the scheduling problem onto this three-tier architecture. The HNGA consists of two genetic algorithms that use a feedback mechanism. At the top level, it is called the global genetic algorithm (GGA), which utilises graph partitioning techniques to partition the task graph into latency-aware subgraphs and assign them across the continuum. Meanwhile, nested local genetic algorithms (LGA) receive these subgraphs and perform fine-grained, time-triggered scheduling within each tier. The LGA provides the update back to the GGA, which uses the information to optimise the subgraphs such that deadlines are met and also generates the global schedule for the safety-critical applications. Experimental validation shows that this integrated approach significantly outperforms state-of-the-art scheduling methods. The proposed method was compared with the Min-Min, Particle Swarm Optimisation (PSO), and Heterogeneous Earliest Finish Time (HEFT) algorithms. The results show that the proposed method significantly reduces the makespan and consistently meets deadlines when compared to the above algorithms. Thus, the main contribution of this work is a robust and verifiable framework that provides a comprehensive solution for deploying safety-critical, real-time applications across the Edge-Fog-Cloud continuum.


  • AD-TTS: A Multi-Task Simulation Framework and Dataset for Event Prediction in Time-Triggered Autonomous Driving

    by Dongchen Li, Daniel Onwuchekwa, Roman Obermaisser

    Time-triggered systems are widely deployed in safety-critical applications due to their deterministic timing behavior. To enhance adaptability under dynamic workloads and environmental changes, adaptive time-triggered systems have received increasing attention. Existing approaches either consume excessive memory or introduce adaptation delays. Proactive speculative scheduling, driven by event prediction, has emerged as a promising alternative. Its core component is an event prediction model. Despite the availability of various time-series and event log datasets in other domains, no suitable dataset exists for event prediction in time-triggered systems. This paper introduces Autonomous Driving Time-Triggered System Dataset (AD-TTS), a dataset tailored for learning-based event prediction in multi-task time-triggered systems. We extend an autonomous driving simulation from CARLA into a multi-task time-triggered system that consists of 21 tasks and 32 event types. Events are logged at each scheduling cycle along with metadata such as timestamps, task identifiers, CPU utilization, memory usage, and event attributes. AD-TTS provides a foundation for future research on event prediction models and adaptive scheduling techniques in time-triggered systems.

  • Open Access

    Scheduling the Exchange of Context Information for Time-Triggered Adaptive Systems

    by Daniel Onwuchekwa, Omar Hekal, Roman Obermaisser

    This paper presents a novel metascheduling algorithm to enhance communication efficiency in off-chip time-triggered multi-processor system-on-chip (MPSoC) platforms, particularly for safety-critical applications in aerospace and automotive domains. Time-triggered communication standards such as time-sensitive networking (TSN) and TTEthernet effectively enable deterministic and reliable communication across distributed systems, including MPSoC-based platforms connected via Ethernet. However, their dependence on static resource allocation limits adaptability under dynamic operating conditions. To address this challenge, we propose an offline metascheduling framework that generates multiple precomputed schedules corresponding to different context events. The proposed algorithm introduces a selective communication strategy that synchronizes context information exchange with key decision points, thereby minimizing unnecessary communication while maintaining global consistency and system determinism. By leveraging knowledge of context event patterns, our method facilitates coordinated schedule transitions and significantly reduces communication overhead. Experimental results show that our approach outperforms conventional scheduling techniques, achieving a communication overhead reduction ranging from 9.89 to 32.98 times compared to a two-time-unit periodic sampling strategy. This work provides a practical and certifiable solution for introducing adaptability into Ethernet-based time-triggered MPSoC systems without compromising the predictability essential for safety certification.


  • Adaptive Time-Triggered Network-on-Chip Architecture: Enhancing Safety

    by Rakotojaona Nambinina, Veit Wiese, Pascal Muoka, Daniel Onwuchekwa, Roman Obermaisser

    Real-time computing systems are designed to meet strict timing constraints and respond to events or inputs within specific deadlines. These systems are commonly used in safety-critical applications such as spacecraft, medical devices, industrial control, and automotive systems. Engineers rely on various scheduling techniques to ensure that timing constraints are met. One such technique is static allocation in time-triggered systems. This technique offers valuable advantages in terms of system safety and dependability. It enables efficient resource usage in Network-on-Chip (NoC) architectures by minimizing message congestion and contention. However, its static nature can present limitations in terms of fault tolerance. This paper focuses on developing and evaluating fault tolerance techniques tailored for NoC-based multi-core architectures to enhance safety. It introduces adaptation techniques that tolerate permanent faults within the time-triggered NoC architecture. Additionally, the paper outlines the incorporation of seamless redundancy mechanisms at the network interface (NI) level, specifically designed to tolerate transient faults that may occur within NoC components, such as routers and links. These mechanisms play a crucial role in safeguarding critical data, further enhancing the reliability and safety of real-time systems in safety-critical applications.

  • Fault-Tolerant Lightweight High Level Architecture

    by Daniel Onwuchekwa, Krishi Savla, Devika Joshi, Roman Obermaisser, Tobias Pieper

    This research presents a fault-tolerant distributed Run-Time Infrastructure (RTI) using the High-Level Architecture (HLA) standard, emphasizing its importance for safety-critical use cases such as automotive systems. In this context, HLA serves as an enabler for hardware-in-the-loop testing, particularly vehicle-in-the-loop testing. Ensuring system reliability, managing real-time responses, facilitating seamless integration and interoperability of various components, effectively handling potential component failures, and maintaining the scalability of complex systems is paramount. Furthermore, the ability to comply with stringent safety regulations also underlines the necessity for safety mechanisms. These safety measures are crucial in protecting users and ensuring dependable system performance, particularly in safety-critical automotive applications where user safety is paramount. The study develops a fault-tolerant distributed RTI, validates it with an automotive use case, and provides a comparative assessment of centralized and distributed HLA implementations. The findings offer insight into large-scale distributed simulations' efficiency, dependability, and safety in various safety-critical applications.

  • Towards Melanoma Detection Using Radar and Image Data

    by Fatima Mammadova, Daniel Onwuchekwa, Roman Obermaisser

    Melanoma is a skin cancer type and has the most dangerous consequences. Melanoma spreads to other organs very fast if not detected on time. Several non-invasive techniques exist which are applied in melanoma detection. An example is dermoscopy, which is an optical method and has the advantage of being less costly and easy to use. However, professional expertise is required to detect cancer in the early stage. Artificial Intelligence (AI) has been utilized in skin cancer detection by developing algorithms that can analyse images of skin lesions and identify the characteristics associated with various types of skin cancer, including melanoma. Nevertheless, information about the depth of the melanoma is not provided by the popular technique of using 2D images in training neural networks. The missing depth information is crucial to detecting melanoma and reaching decisions to execute biopsy when necessary. Radar sensors have shown the potential to provide this depth information due to its penetrating capability, allowing them to be applied in the detection of melanoma. The application of AI techniques using 2D images to detect melanoma, and the use of radar, has been investigated independently in recent literature. However, the combined technique still remains to be investigated. We propose integrating radar and image data to improve melanoma classification in this work. Based on the unavailability of radar data, the proposed technique is applied to the skin with nevi and birthmarks, clear skin, and body parts like inner palms, lower arms, and upper arms. The data from both sources are fused by applying an early fusion technique and later utilised for AI classification. Despite the small sample size, the fusion positively impacted classification compared to using only image data. The AI classification was performed on the first two cases, where the overall accuracy increased by 36% for both. Radar signals were also tested on wet and dry skin and have shown distinguishing results.