Page 251 - Big data - Concept and application for telecommunications

Big data - Concept and application for telecommunications 5

to understand the requirements of utilizing big-data analytics to provide user services with personalized QoE
and to enable highly efficient resource utilization in the future network.

Big data in the future network need to be extensively analysed in order to retrieve relevant and valid
information. Big data provide unprecedented opportunities for MNOs to understand the behaviour and
requirements of mobile users, which in turn allow for intelligent real-time decision-making in a wide range
of applications. By analysing these data, the future network can provide and support different smart services.
However, the nature of big data presents vast challenges in relation to data mining, mobile sensing and
knowledge discovery. New technologies are required to handle big data in a highly scalable, cost-effective
and fault-tolerant fashion.
In order to enhance operational efficiency in network infrastructure under varying environments, MNOs are
encouraged to adjust network traffic requirements and improve resource allocation efficiency using
intelligence and analytics based on big data.

The collection of big data can be achieved from user equipment (UE), the RAN, the core network and Internet
service providers (ISPs). The events that occur at UEs are collected either through user applications or via
control signalling. At the RAN eNB, the cell-level data (including the signalling exchanged over the air) and
instantaneous measurement reports are collected by DPI technology. Meanwhile, MNOs possess huge
amounts of data obtained by DPI technology relating to user bearers or services in the core network. When
the cell size becomes smaller in the HetNet, the number of nodes B increases. As this trend continues,
network data may explode and impose a great burden on data collection.

Figure 7-1 – Big-data-driven networking in the future network (including 5G)

Furthermore, big data storage infrastructure needs to have scalable capacity as well as scalable performance.
Thus, storage management needs to be simple and efficient so that storing and sorting big data can be
achieved easily. After data are collected and stored, another big challenge for MNOs is how to process such
huge volumes of data. The collected data are multi-source, heterogeneous, real-time and voluminous. For
this reason, data analytics and knowledge extraction techniques are required to process the data and convert
it into actionable knowledge. Consequently, this knowledge can be used to design adaptive schemes for
network optimization.

Network and infrastructure 243