Empirical Analysis of 4G LTE Network Capacity Dimensioning Models for High-Density Urban Environments in Lagos

Accurate capacity dimensioning is fundamental to preventing congestion and maintaining quality of service as mobile data demand continues to grow at double-digit annual rates in Nigerian metropolitan areas. This study empirically analyzes the accuracy of established 4G LTE capacity dimensioning models when applied to high-density urban environments in Lagos State, identifying the gap between theoretical predictions and observed network behavior. An empirical measurement study was conducted across six high-traffic zones in Lagos Island and Victoria Island, collecting throughput, resource block utilization, and subscriber density data from MTN and Airtel network management systems over twelve weeks under a research data sharing agreement. The Shannon-Hartley theorem-based theoretical capacity estimates and the standard 3GPP RAN capacity planning model were applied to the measured environment parameters and compared against actual observed capacity performance. Results demonstrate that both models overestimate practical network capacity by between 23% and 41% under real-world conditions, primarily because they do not adequately account for the overhead from handover signaling, MIMO spatial multiplexing efficiency reductions in dense scattering environments, and the impact of non-uniform traffic spatial distribution. A correction factor model incorporating spatial traffic density distribution is derived from the empirical dataset. The study fills a research gap by providing locally calibrated capacity estimation parameters for Lagos urban environments, offering network planners improved accuracy compared to vendor-provided generic models. Keywords: LTE capacity, dimensioning model, Lagos, empirical analysis, network planning.