Autor(es): Torres, Pedro ; Marques, Paulo ; Marques, Hugo ; Dionísio, Rogério Pais ; Alves, Tiago Ferreira ; Pereira, Luis Miguel Cardoso ; Ribeiro, Jorge Miguel Afonso
Data: 2017
Identificador Persistente: http://hdl.handle.net/10400.11/6076
Origem: Repositório Científico do Instituto Politécnico de Castelo Branco
Assunto(s): LTE; SON; Machine Learning; Forecasting
This paper presents a methodology for forecasting the average downlink throughput for an LTE cell by using real measurement data collected by multiple LTE probes. The approach uses data analytics techniques, namely forecasting algorithms to anticipate cell congestion events which can then be used by Self-Organizing Network (SON) strategies for triggering network re-configurations, such as shifting coverage and capacity to areas where they are most needed, before subscribers have been impacted by dropped calls or reduced data speeds. The presented implementation results show the prediction of network behaviour is possible with a high level of accuracy, effectively allowing SON strategies to be enforced in time.
