MULTIOBJECTIVE 5G AND B5G OPTIMAL NETWORK PLANNING USING METAHEURISTIC WITH MMWAVE FOR BASE STATION DEPLOYMENT

Abstract

Fifth-generation and Beyond 5G(B5G) wireless systems increasingly rely on the millimeter-wave (mmWave) spectrum, whose blockage sensitivity and severe propagation loss make base stations deployment a complex multi-objective problem. This paper compares two well-known metaheuristic algorithms, Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), for deploying in two-tier heterogeneous networks over 5 km × 5 km dense urban area, under two mmWave scenarios differing only in Urban Micro frequencies: 28 GHz or 38 GHz. Both algorithms pursue four objectives: maximizing spectral efficiency (SE), energy efficiency (EE), and the coverage, while minimizing total number of active base stations, Urban Macro(UMa) and Urban Micro(UMi) combined. Applied under an identical propagation, the link-budget, and the constraint conditions, results show PSO reduces active base stations more aggressively, up to 40.7%, achieving higher EE (up to 42.5%) and SE (up to 27.5%), while GA retains more base stations to achieve higher coverage, up to 94.6%. In scenario 1 (28GHz), PSO drops to 83 active stations with 91.3% coverage, versus GA's 90 stations at 94.6%. In scenario 2 (38GHz), PSO retains 90 stations at 90.2% coverage, versus GA's 97 at 93.1%. This EE versus coverage tradeoff recurs identically across both bands. Our proposed model results are confirmed with existing studies. Neither algorithm dominates unconditionally, GA suits coverage critical, PSO suits efficiency constrained deployments ones. Future work will model inter-cell interference under hybrid PSO-GA approach.