Abstract:
5G is expected to accommodate exceptional services beyond current cellular systems.
The three main use cases envisioned for 5G networks are enhanced Mobile Broadband,
massive Machine Type Communication and Ultra Reliable and Low Latency Communi-
cation. The enormous competition and the increasing number of UEs in the telecom-
munication market have garnered attention towards the necessity of optimized and cost-
e cient network. Tracking the position of UEs while minimizing the signalling cost is a
key challenge that needs to be addressed for the upcoming 5G networks. Tracking area
(TA) is a term used for the assemblage of cells in long term evolution (LTE) network.
Planned distribution of TAs, depending on the mobility pattern of UEs, is a technique
used to optimize the signalling overhead, and plays a vital role in utilizing the resources
e ciently. Compared to the earlier generations of cellular networks, 5G networks will
have more
exible conguration of TAs as it will utilize the concept of TA list, given in
3GPP release 12. One of the objectives of 5G mobile network is to cope up with an ever
increasing number of UEs. This thesis discusses TA recon guration for 5G network using
game theory, and proposes an optimal solution for reducing the signalling overhead of
the network.
In order to satisfy the 5G requirements, heterogeneous radio access technologies (RATs)
have been proposed. To use the RAT e ciently, massive connectivity of devices has been
proposed in 3GPPP Release 15. For the earlier phase, non-standalone architecture has
been proposed in which existing 4G LTE infrastructure will be used. To have a full
5G network coverage stand alone architecture has been proposed in 3GPP release 15 for
which low-cost and low-power small base-stations (microcells) will be deployed, and these
power supplied active nodes will form the clusters as per proposed algorithm. Increasing
the number of devices will give rise to signalling overhead issue consisting of TA update
(TAU) and paging overhead. This thesis proposes a method to reduce the signalling cost
using clustering algorithm and the uni cation of both the layers (4G and 5G).
