Self-Healing Neuromorphic Architecture for Resilient Consumer Internet of Things
IEEE Consumer Electronics Magazine · 2026-11 · 15 (6) · 16-25
DOI: 10.1109/mce.2026.3692960 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Consumer Electronics Magazine, 2026-11. Verfasst von Yellapragada Charan Krishna und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 19 Literaturangaben.
Abstract (Originalsprache)
The advent of collaborative intelligence swarms puts severe pressure on consumer hardware systems needing chips that remain functional despite wear and tear without manual interference. As the workload is shared among Internet of Things endpoints, the on-chip communication network infrastructure needs to handle fluctuating demands and physical faults without external input. This article introduces plastic, hierarchical, event-driven neuromorphic–network-on-chip (PHEN-NoC), a neuromorphic architecture designed explicitly as a robust core for future consumer edge devices. Reinforcement learning in conjunction with biological structural plasticity enables the platform to optimize and restore itself on chip. Simulation results on a set of spiking benchmarks indicate that the architecture is capable of achieving autonomous multilink fault tolerance within 5,000 steps while maintaining highly reliable data transmission in accordance with hardware buffer capacity limits for mission-critical scenarios. Pruning unused connections yields an estimated 15.2% reduction in static power. Hardware validation on a Xilinx Zynq UltraScale+ field-programmable gate array shows synthesis feasibility for the cognitive routing engine, using 19.63% of the total lookup table capacity, with 1.352-W power consumption. PHEN-NoC represents a decisive step toward self-organizing silicon sustaining continuous intelligent operation across the consumer digital ecosystem.
Zitieren
APA
Krishna, Y. C., & Reddy, B. N. K. (2026). Self-Healing Neuromorphic Architecture for Resilient Consumer Internet of Things. IEEE Consumer Electronics Magazine, 15(6), 16-25. https://doi.org/10.1109/mce.2026.3692960
Vancouver
Krishna YC, Reddy BNK. Self-Healing Neuromorphic Architecture for Resilient Consumer Internet of Things. IEEE Consumer Electronics Magazine. 2026;15(6):16-25. doi:10.1109/mce.2026.3692960
BibTeX
@article{Krishna2026,
author = {Krishna, Yellapragada Charan and Reddy, B. Naresh Kumar},
title = {Self-Healing Neuromorphic Architecture for Resilient Consumer Internet of Things},
journal = {IEEE Consumer Electronics Magazine},
year = {2026},
volume = {15},
number = {6},
pages = {16--25},
doi = {10.1109/mce.2026.3692960}
}
EUVIMED