A hierarchical neuromorphic multi agent framework for energy aware and secure 6G resource optimization using Neuro6G...

Area of Science:

• * Artificial Intelligence and Machine Learning
• * Wireless Communication Systems
• * Computer Engineering and Architecture

Background:

• * Sixth-Generation (6G) wireless networks demand intelligent, energy-efficient resource management in distributed systems.
• * Neuromorphic computing offers event-driven, low-power processing suitable for complex network tasks.
• * Existing 6G resource management faces challenges in energy sustainability, latency, and security.

Purpose of the Study:

• * To introduce Neuro6G-Agent, a hierarchical neuromorphic agentic intelligence framework.
• * To enable energy-conscious cognitive collaboration across cloud-edge-end 6G deployments.
• * To address energy sustainability, latency, and security in distributed 6G resource management.

Main Methods:

• * Integration of Energy-Aware Spiking Neural Networks (EA-SNNs) with multi-agent reinforcement learning.
• * Implementation of a three-tier architecture (cloud, edge, end devices) with dedicated neuromorphic agents.
• * Utilized adaptive threshold EA-SNNs, distributed trust computation, and hierarchical resource optimization.

Main Results:

• * Demonstrated a 34.7% reduction in energy consumption.
• * Achieved a 28.3% decrease in end-to-end latency.
• * Reached 95.6% security threat detection accuracy, validated across benchmark datasets.

Conclusions:

• * Neuro6G-Agent effectively addresses key challenges in 6G resource management.
• * Neuromorphic intelligence is viable for complex optimization in next-generation wireless architectures.
• * The framework shows significant improvements in energy efficiency, latency, and security.