×
Home Current Archive Editorial board
Instructions for papers
For Authors Aim & Scope Contact
Original scientific article

NEUROMORPHIC-INSPIRED HYBRID COGNITIVE MODEL FOR SELF-OPTIMIZING RESOURCE MANAGEMENT IN 6G EDGE NETWORKS

By
Reji K Kollinal Orcid logo ,
Reji K Kollinal

Assistant Professor, BPC College , Piravom , India

Mariya T Cheeran Orcid logo
Mariya T Cheeran

Senior Manager, Reply India , Kochi , India

Abstract

Introduction: The 6G world requires connected intelligence, but there is a crucial paradox between the standards of Large Language Model (LLM) and edge constraints. The premium devices have up to 6-12GB of DRAM, whereas the typical 175B models need 350GB of storage, which is 30 times that of the premium version.  Literature Survey: It has been proposed that the bandwidth can be reduced by 90 % with Semantic Communication (Scom) and Edge Semantic Cognitive Intelligence (ESCI). Besides, neuromorphic-based Spiking Neural Networks (SNNs)-model quantization (INT4/INT8) are also known to be necessary to achieve order-of-magnitude energy efficiency (J/token) on resource-constrained hardware. Methodology: This paper proposes a Hybrid Cognitive Model utilizing a three-tier Cloud-Edge-Device hierarchy. The model integrates event-driven neuromorphic principles with self-optimizing resource management, utilizing paged KV-cache and resource-aware agents for dynamic task offloading. Results: Quantitative evidence is used to show that the hybrid strategy helps to address the 30x resource gap by attaining a 10-100x energy-per-token efficiency due to event-driven neuromorphic sparsity. Statistical analysis makes it evident that semantic filtering substantially reduces communication overhead and maintains reasoning faithfulness by 90 %, and, effectively, it keeps the thermal conditions of devices stable in the case of prolonged 6G edge communications. This model can be used to make sustainable and multi-step thinking on the edge. The hybrid solution achieves the 6G vision of pervasive intelligence by bridging the hardware-software gap via cross-layer co-design.

References

1.
Yang H, Lam KY, Xiao L, Xiong Z, Hu H, Niyato D, Vincent Poor H. Lead federated neuromorphic learning for wireless edge artificial intelligence. Nature communications. 2022 Jul 25;13(1):4269.
2.
Yang B, Cao X, Xiong K, Yuen C, Guan YL, Leng S, Qian L, Han Z. Edge intelligence for autonomous driving in 6G wireless system: Design challenges and solutions. IEEE Wireless Communications. 2021 May 14;28(2):40-7.
3.
Wei P, Guo K, Li Y, Wang J, Feng W, Jin S, et al. Reinforcement Learning-Empowered Mobile Edge Computing for 6G Edge Intelligence. IEEE Access. 2022;10:65156–92.
4.
Letaief KB, Shi Y, Lu J, Lu J. Edge artificial intelligence for 6G: Vision, enabling technologies, and applications. IEEE journal on selected areas in communications. 2021 Nov 8;40(1):5-36.
5.
Shang X, Huang Y, Liu Z, Yang Y. NVM-enhanced machine learning inference in 6G edge computing. IEEE Transactions on Network Science and Engineering. 2021 Sep 3;11(6):5615-26.

Citation

This is an open access article distributed under the  Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 

Article metrics

Google scholar: See link

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.