Abstract:
The mobile network generation has undergone an unprecedented change that attempts to meet
the large demand for data-driven by large connected devices and systems. To meet the massive
demand of large capacity, current mobile networks should employ technologies that utilize a
broad spectrum, offer higher cell density and high spectral efficiency. Small cell deployment
has been a critical upgrade in Fourth Generation (4G) mobile networks. As higher spectrum
bands require denser network deployments to support higher traffic volumes per unit area, the
need for small cells will has become more critical in Fifth Generation (5G) networks, which is
expected to satisfy the requirements for future heterogeneous usage situations, such as huge
expansion in connectivity and traffic volume. Together with small cells, network densification
is expected to play a significant role in the fifth generation and beyond networks.
In this thesis, ultra-dense networks (UDNs) 5G network planning and multiobjective
optimization have been done to analyze the practical deployment for standalone ultra-dense 5G
small-cell network in Bahir Dar city. Propagation modeling is performed using a deterministic
dominant path loss model that incorporates the city's buildings and terrain maps. The
multiobjective optimization algorithm that gives more emphesises mainly on the constraints
of coverage (the number of base satations) and capacity has been solved using Genetic
algorithm. The optimal network and propagation parameters that maximizes the throughput is
selected based on the proposed multiobjective optimization algorithm.
To realize and compare the performance of ultradense network topologies in a realistic urban
case study area, an ultra-dense network multiobjective optimization approach is developed in
this study. The outdoor base station planning considers small cell, streetlamp posts, utility pole,
buildings and new candidate locations. The obtained Pareto optimal networks shows the
aggregate capacity of 1.728 Gbps for steet lamp post sites consideration, while considering
streetlamp posts, buildings and new candidate loctitions the aggregate capacity is increased to
4.304 Gbps. Finaly, by considering the hybrid of all scenarios (i.e. streetlamp posts, utility pole,
buildings and new candidate locations) the aggregate capacity reaches 7.686 Gbps. According
to planning outcomes for small cell distribution, locations other than main streets require a
greater need for outdoor small cell deployment.
Keywords: 5G Ultra-dense, Multiobjective optimization, Network Planning and
Optimization, GA, Small cells