Showing Page:
Ph.D. Next Generation Networks
The Smart Network slice mobility in NGN of Mobile System by deploying Fuzzy logic and Artificial
Neural Network
Aqeel Ahmed
PhD candidate
The new industries sectors having customized services that has been un-served or inadequately served by current mobile
network operators. To overcome these issue Network slicing offer various benefits by deploying the advanced Mobile network
like 5G even though these technologies still cannot handle the use cases that targeted it in term of mobility. These mobility uses
cases are e.g. low latency communication services, shared by group of moving devices e.g. autonomous vehicles that share
sensor data further more it demand ultra low latency by leveraging the Multi Access Edge Computing (MEC) concept ,Software
defined Network, Network slicing, Virtual resources live migration techniques etc.In this article, we define various mobility
pattern with various priorities based on QoS with fuzziness and different methods of grouping users and different trigger for
network slice mobility with smart fuzzy control. Furthermore we used artificial neural network for the selection of network slice
with accuracy and customized computation and storage features of Edge computing for low latency communication within
scale and also can utilized resources from neighbor Edge and cloud as per demand. Furthermore we evaluated the mobility of
services and network slice based on simultaneous migrations of multiple containers.
(at least 10 lines)
Keywords: (1 to 2 lines)
SDN,software defined Network,NFV network function virtualization ,QoE Quality of Experience
When high data rate is expected on mobile system the 5
Generation network is only solution that not only provide the data
rate for mobile user but also provide the platform for wider communication ecosystem for the internet of Things and machine
type communication applications [1].The vertical industries e.g. automotive ,e health, public safety, and smart grids require
services with high performance, scalability and availability which restructured the 5
generation mobile network compared to
its predecessor mobile technologies.
In restructuring, From physical infrastructure, logical network functions are decoupled in order to achieved the basis
of deployment of self contained, programmable and customize networks [2].The technologies already available to implement
the separation of logical network functions from the infrastructure leading to softwarization[3]. These known technology are
software defined network SDN) and Network Function Virtualization(NFV).Where software defined Network provide the
programmability of the connectivity between the network functions while NFV offer the means to instantiate and manage the
virtualization network function which need to create multiple logical or virtual networks. By using these two technology we can
build the concept of network slicing[4] where we can share the same underlying infrastructure among self contained or isolated
network and hence providing the slicing concept.
All standardized organization like SDO(Standard development organization)NGMN(Next generation mobile network Alliance
3GPP (third generation partnership project and ITU-T(International Telecommunication Union-Telecommunication standard
sector) and all academic community treating the network slicing as active research field now a days. The network slice is
already defined as end to end (E2E) logical networking running on the top of a common underlying network either physical or
virtual, fully isolated with independent control and management and flexibly programmable to meet service level agreements
(SLA) of specific services [100].Where network slice consists of computing and storage resources associated with virtual
networks. In these virtual network, multiple virtual sub network segments that can span across multiple technologies as well as
administrative domains. When mobile user connect to multiple slices simultaneously depending on the type of service they are
using[5].Furthermore ,for latency sensitive services,MEC[6]even with minimal delay but provide powerful service delivery.MEC
servers as well as network access points can be sliced to serve multiple slices with different service characteristics in MEC
environments. It is obvious that when mobile user move from one domain to another their ongoing communication and service
session on network slice may result in drop of QoS issue if it is not completely disconnect. Which is most often when their
ongoing network services have to be served by instantiating a totally new network slice of the same type at destination. In
simple word Backhaul, new radio resources, computing and storage are the basic characteristics that effect in creating new slice
at destination. This leads to a network slice that has to support moving its computation., following its users mobility patterns.
Showing Page: