1.0 Introduction
Cellular network optimization has traditionally been associated with a process that aims to adjust the network air interference to market-specific traffic and propagation conditions (Lee et al, 1993). The tuning operations in this process focus on a small set of hardware parameters such as cell-site locations and antenna configurations.
The cells themselves are grouped more densely around traffic nucleation points to provide capacity, and the antennas are pointed in compliance with the local terrain and clutter to reduce signal shadows and interference.
Occasionally, software parameters such as handoff thresfolds and cell-power budgets are also adjusted while hardware parameters are easily set during network installation, they are hard to change afterward.
As a result, the optimization with respect to hardware parameters occurs during network planning and development, and it is only repeated in areas where performance problems or infrastructure upgrade are required. Since it is perform as a singular event, the optimization process is fundamentally based upon time-average worst-case traffic and propagation conditions originally, these optimizations were performed through a manual, it erative process relying on network planning tools and drive testing.
Bell labs subsequently introduced the concepts of predictive optimization in the ocelots optimization tool, which computes optimum network parameters directly according to well-defined performance metrics (Drabeck et al,2005). Its introduction has translated into faster network rellouts, improved network performances, and higher capacity.
Notwithstanding its widespread use, the static optimization approach is increasingly approaching its limits. The growth of wireless customer base and the introduction of various new data services mandate the consideration of new objectives such as throughput, delay, latency, and quality of service (Qos). Indeed, the migration to IP multimedia subsystem (IMS) will promote the continuous development of the new services with different resource requirement Qos demand, and traffic characteristics.
Furthermore, data services introduce demand fluctuations that are intrinsically larger than they are for voice services. The multidimensional nature of demand, its temporal dependency, and its increased dynamic range render optimization strategies based on the peak (albelt composite) loading progressively less effective at efficiently allocating and managing network resource.
Additionally, the demand for increasing data rates and the falling costs for networks hardware will drive network architectures towards micro-cellular structures. This development will create frequent infrastructure upgrades with the demand for fast, autonomous, and inexpensive cell integration.
1.1 Aims and Objective
This work objective is to identify the dynamic optimization in future cellular network.
As a consequently we see a growing need for additional dynamic optimization mechanisms with the following capabilities:
- State and Time-dependent control parameters to help the network adapt the coverage and capacity tradeoff for multiple services in response to spatio -temporal demand variation.
- Coordinated (and potentially additional autonomous) load-balancing mechanism that can address demand and traffic fluctuation by optimally “smoothing out” uncorrelated demand peaks between neighboring cells and even between differing wireless technologies; and
- Active measures to address rare but undesirable events, such as reducing dropped and blocked calls.
1.2 Scope of Study
The work covers the dynamics optimization of future cellular networks in Nigeria, the development of conceptual approaches, algorithms, modeling, simulation, and real-time measurements that provide the foundation for future dynamic optimization techniques.
…
