Page 243 - Big data - Concept and application for telecommunications

Big data - Concept and application for telecommunications 5

Evaluating closeness (both geographic and social) among users in the same base station requires collection
of the following:
1) geographic data: the location and mobility pattern of users;
2) social data: the social characters of users, including contact information and device information;

3) network data: wireless channel states, including channel fading and interference.
6.8.2 Big-data-driven resource allocation based on mobile user moving behaviour

According to mobile user moving behaviour, resources are allocated to each base station with a fixed
configuration, resource utilization must be underestimated. On one hand, it is difficult for users in a hotspot
to obtain good QoE during peak hours. On the other, resources may be greatly wasted in idle times at low
traffic locations.
Current and historical data can be utilized by data analytics to predict traffic for high-density areas in the
networks. Using the respiratory effect of the base station may help to accurately serve the right place at the
right time, promoting recourse utilization.
To adjust the recourse utilization of the base station, it is necessary to collect the following:
1) network data: the wireless channel states, including channel fading and interference;
2) user data: the location and access pattern of users that can be collected from the interface between
service general packet radio service (GPRS) support node (SGSN) and gateway GPRS support node
(GGSN).

6.9 Big-data-driven network planning and design
In most traditional deployment cases, the sites of cells in mobile network are not optimized due to insufficient
statistical data. By tracking mobile devices, their detailed activities can be recorded to provide real-time
where, when and what information about mobile users in the network. A feasible solution is to make use of
both network and anonymous user data including dynamic position information, and other various service
features. Consequently, massive volume, velocity and variety of data need to be processed by advanced
analytics techniques, which can transform the data into actionable knowledge. In order to understand traffic
trends well, it is imperative to analyse the data in relation to corresponding content and events.
Given the actionable knowledge inferred from big datasets, the MNOs can make wise decisions about where
and how to deploy cells in the networks. This also allows them to predict traffic trends and prepare plans for
future investment.

Network design can be more efficient and reasonable with comprehensive data study of network status,
external environment and various elementary issues.

Analysing customer behaviour in their use of network services is crucial to understand traffic. Thus,
concurrent weather and temperature data and social event information are taken into consideration to
better understand traffic.

6.9.1 Big-data-driven mobile network planning and construction based on customer experience
When mobile customers use the services provided by MNOs, big data is generated both in the user plane and
signal plane. By analysing them comprehensively, actual user experience in their use of services can be
determined. According to the experience of users within the same base station, a judgement can be made
about whether the base station is poor. A reasonable and intelligent plan to increase quantity or expand
capacity of the base station can then be made to improve mobile network quality efficiently. Figure 6-8 shows
the architecture of big-data-driven mobile network planning and construction.

Network and infrastructure 235