Solving signaling storms in LTE networks: a software-defined cellular architecture

The LTE network infrastructure is composed by monolithic devices that carry out a convoluted set of tasks in a vendor-speci c manner. Therefore, LTE networks are largely in exible, and consequently unable to adapt to a constantly increasing number of mobile subscribers and the changeable usage pattern of the Internet service. In fact, current LTE networks are a ected by signaling storms, which come from the inability to reduce the number of signals exchanged among the infrastructural elements of the network when the number of subscribers' requests grows. In this work, we propose a software-de ned cellular architecture, whose logical entities can be mapped to an arbitrary number of physical devices, allowing di erent implementations depending on the speci c use case. In particular, we show that the proposed model actually mitigates the impact of signaling storms, as it can be tailored to reduce signi cantly the number of signals owing in the network during the occurrences of the most frequent network events.