Big data-driven anomaly detection in cellular networks

With rising capacity demand in mobile networks, the infrastructure is also becoming increasingly denser and more complex. This results in collection of larger amount of raw data (big data) that is generated at different levels of network architecture and is typically underutilized. To unleash full value of the complex data sets, innovative machine learning algorithms need to be utilized for the comprehensive analysis of the big data in order to extract valuable insights which can be used for improving the overall performance of the network. In addition, a major challenge for cellular network operators is to cope up with increasing number of complete (or partial) cell outages and to simultaneously reduce operational expenditure (OPEX). This paper contributes towards the aforementioned problems by exploiting big data generated from the core network of 4G LTE-A known as call detail record (CDR) to detect anomalous behavior of the network. We present a semi-supervised statistical-based anomaly detection technique to identify in almost real-time: First, unusually low user activity region depicting sleeping cell, which is a special case of cell outage and is particularly challenging to detect as it does not trigger any alarm and remains invisible until a number of complaints are received from the subscribers; and second, unusually high user traffic area corresponding to a situation where special action such as additional resource allocation, fault avoidance solution etc. may be needed. Achieved results demonstrate that the proposed method can be used for timely and reliable anomaly detection in current and future cellular networks.