Hybrid Machine Learning based Optimization with Beamforming of 6G Mobile Radio Network for LEO Satellite base station in Urban Areas

The rapid evolution of sixth-generation (6G) mobile communication networks requires advanced optimization and intelligent deep learning models to meet the demands of ultra-reliable, low-latency, and energy-efficient wireless systems. Low Earth Orbit (LEO) satellites has a vital role in enabling non-terrestrial net works (NTN), particularly in dense urban environment areas. This paper presents a comparative analysis of hybrid optimization approaches integrating Artificial Bee Colony (ABC), Grey Wolf Optimizer (GWO), Ant Lion Optimizer (ALO), Whale Optimization Algorithm (WOA), Salp Swarm Algorithm (SSA), and Par ticle Swarm Optimization (PSO) with Extreme Learning Machine (ELM) for beamforming in 6G mobile radio networks. Integrated with beamforming for 6G NTN. The models are analysis using pathloss (PL),signal-to-interference plus-noise ratio (SINR), coverage area, capacity, spectral efficiency (SE), energy efficiency (EE), bandwidth efficiency (BE), bit error rate (BER), and propaga tion delay. ABC-ELM has performed with result mean values of the PL (187.46 dB), SINR (22.62 dB), coverage area (68.03%), capacity (10.82 Gbps), SE (10.82 bits/Hz), EE (44.10%), BE (44.10%), BER (1.45−8 bps ) and propagation delay (48.65%) in urban areas. As a result, demonstrating of hybrid ELM-based opti mization integrated with beamforming is a promising approach for enabling 6G LEO satellite networks.