Abstract
This paper introduces the Adaptive Hierarchical Multi-Objective Resource Optimizer (AH-MORO), a ground-breaking framework for subcarrier allocation in Smart Grid Neighborhood Area Networks (NANs), addressing critical limitations of existing methods in dynamic, high-density environments. Traditional approaches suffer from static resource allocation, inefficient interference management, and poor scalability, leading to suboptimal throughput, latency, and energy consumption. AH-MORO innovates through three core mechanisms: (1) a hierarchical multi-objective optimization model that dynamically balances throughput maximization, latency minimization, and energy efficiency using adaptive weight parameters (λ₁, λ₂, λ₃), (2) a dual-layered interference mitigation system combining constraint-based subcarrier assignment and adaptive power control to suppress co-channel interference, and (3) a metaheuristic solver (Genetic Algorithm-Deep Reinforcement Learning hybrid) enabling real-time, low-complexity optimization under fluctuating traffic loads. Rigorous simulations demonstrate AH-MORO's superiority over state-of-the-art methods, achieving 37.5% higher throughput, 34.2% lower latency, 24% reduced energy consumption, and 33.3% improved interference reduction in dense urban NANs (1,000 + devices). The framework uniquely guarantees QoS via fairness constraints, ensuring minimum throughput ([Formula: see text]) for all users while adhering to strict latency ([Formula: see text]and energy ([Formula: see text]) bounds. These results validate AH-MORO as the first holistic solution for real-time, energy-efficient, and interference-resilient Smart Grid communications, setting a new benchmark for adaptive resource management in next-generation NANs.