EUVIMEDEUVIMED entdecken
Crossref

Heterogeneous Hardware–Software Co‐Design for Multi‐MCU IoT Edge Computing: Architecture, Runtime Support, and Task Scheduling

Yuliang Lu, Min Liu, Songrong Qian

Concurrency and Computation: Practice and Experience · 2026-07 · 38 (14)

DOI: 10.1002/cpe.70861 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Concurrency and Computation: Practice and Experience, online erschienen am 14.07.2026. Verfasst von Yuliang Lu und 2 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Battery‐powered internet of things (IoT) edge terminals are increasingly required to perform local preprocessing, event analysis, and lightweight decision making under strict constraints on energy consumption, latency, memory capacity, and hardware cost. However, conventional single‐MCU nodes and gateway‐centric edge solutions are often unable to balance low‐power operation with computation‐intensive workloads. To address this problem, this paper proposes a hardware–software co‐designed heterogeneous multi‐MCU edge computing platform for resource‐constrained IoT sensing applications. The proposed architecture organizes heterogeneous processing resources into a scheduling node, high‐performance MCU nodes, and low‐power MCU nodes, thereby enabling task execution to be matched with different computational and energy characteristics. A lightweight runtime mechanism is developed to support task dispatching, dynamic program loading, node activation, state maintenance, and result collection without relying on a heavy operating system. Furthermore, a practical node capability profiling procedure is introduced to measure execution capability, active‐power increment, and loading overhead under real operating conditions. Based on these profiled parameters, a latency–energy scheduling model and a discrete particle swarm optimization‐based task allocation method are developed. Experimental results on representative monitoring workloads show that the proposed scheduler achieves a better latency–energy trade‐off than static and extreme mapping strategies while satisfying hardware feasibility constraints.

Zitieren

APA

Lu, Y., Liu, M., & Qian, S. (2026). Heterogeneous Hardware–Software Co‐Design for Multi‐MCU IoT Edge Computing: Architecture, Runtime Support, and Task Scheduling. Concurrency and Computation: Practice and Experience, 38(14). https://doi.org/10.1002/cpe.70861

Vancouver

Lu Y, Liu M, Qian S. Heterogeneous Hardware–Software Co‐Design for Multi‐MCU IoT Edge Computing: Architecture, Runtime Support, and Task Scheduling. Concurrency and Computation: Practice and Experience. 2026;38(14). doi:10.1002/cpe.70861

BibTeX
@article{Lu2026,
  author = {Lu, Yuliang and Liu, Min and Qian, Songrong},
  title = {Heterogeneous Hardware–Software Co‐Design for Multi‐MCU IoT Edge Computing: Architecture, Runtime Support, and Task Scheduling},
  journal = {Concurrency and Computation: Practice and Experience},
  year = {2026},
  volume = {38},
  number = {14},
  doi = {10.1002/cpe.70861}
}
Kostenlos starten →