Cell outage detection based on improved BP neural network in LTE system

With the development and evolution of the LTE system, the operators are experiencing huge challenges on operations and service maintenance, which sets off a wave of SON (Self-organizing Network) research. As one of the crucial component of SON, the cell outage detection is an effective way to automatically detect outage cells resulted from hardware or software problems. Our work in this paper aims to introduce a cell outage detection mechanism to timely and accurately detect outage cells. After classifying the cell into four states, namely the healthy, degraded, damaged and outage state, we present a cell outage detection mechanism based on BP network. In order to enhance the training speed of traditional BP, the Differential Evolution (DE) algorithm is adopted as the training algorithm. We have simulated the proposed mechanism in the matlab environment and get a better performance of high detection accuracy by comparing it with the standard BP algorithm.