A multi-cell multi-objective self-optimisation methodology based on genetic algorithms for wireless cellular networks
Rights accessRestricted access - publisher's policy
Self-organising networks (SON) are seen as one of the hottest topics in telecommunication network research and development, eagerly awaited by network operators to achieve a reduction in operational expenditures. However, there are still many challenges and dif¿culties when moving from the SON concept to practical implementation. In this context, this paper ¿rst provides a general formulation of the automated optimisation problem and a detailed description of the main challenges and dif¿culties ahead. Then, a generic multi-cell multi-objective self-optimisation methodology based on genetic algorithms is proposed. The proposed framework is formulated in detail for a joint coverage and overlap optimisation problem in a multi-cell scenario. A case study using real measurements of a Universal Mobile Telecommunications System network deployed in a medium-size European city is presented to illustrate the proposed methodology. In the presented case study, the pilot power, antenna tilt and antenna azimuth of the different cells are optimised according to certain cell coverage and cell overlap targets. Results reveal that the genetic-based approach is able to provide optimised solutions that ef¿ciently achieve the desired targets accounting for inter-cell coupling effects.
CitationSanchez, J., Sallent, J., Perez, J., Agusti, R. A multi-cell multi-objective self-optimisation methodology based on genetic algorithms for wireless cellular networks. "International journal of network management", Juliol 2013, vol. 23, núm. 4, p. 287-307.