Delay Based Network Slicing And Its Resilience Analysis

Over the last decade in the telecommunication area a significant change and great advancement.The telecommunications industry has made rapid progress now not only typical communication along with it connected the society.The fifth-generation(5G)mobile communications system provides a far higher level of performance than the previous generations of mobile communications systems.However,due to the proliferation of numerous connected devices,5G is struggling to support user demand.To tackle this challenge,the network slicing is considered as a promising approach in the 5G and even beyond.Network slicing is a technique that can provide operators with an efficient way to segment a single network according to different service specifications and implement a logical network over a similar physical infrastructure for each type of service.It is a revolutionary idea for customizing specific logical networks for certain services from a common general infrastructure.To optimized the end-to-end delay of network,the placement of virtual network functions(VNFs)in the slice is crucial due to different effects on VNFs.This is case sensitive and failure of this due to unwanted bugs and variation of traffic demands significantly hamper the overall performance of the network.To minimize this effect,here need to enable the resilience of the system.In this dissertation,the latency optimization problem is formulated and then the network slicing method is adopted to meet the demand of the UE and minimize latency.The numerical results depict that the proposed approach outperforms and achieves low latency compared to the other benchmark schemes.The main contributions of this work are summarized as follows:1.Investigate the fundamental knowledge and basic tools of complex networks and adapt this concept for the optimization of the End-to-End(E2E)delay in network slice deployment,and mention the efficiency improvement and delay reduction when using VNF instead of dedicated hardware.2.Propose mathematical models to reduce end-to-end delay on behalf of proper placement of VNFs in the slice.Actually,for the placement(link&node)of the network slice proposed the shortest path(heuristic approach)based algorithm.In first the slice separated into multiple chain functions.This shortest-path(heuristic approach)based algorithm enables the network slice to convert in multiple chains.This algorithm has less complexity and it allows us to find the optimal way of placement to ensure the minimum delay.The mathematical formulation demonstration of this proposed algorithm is describing to deploy the network slice upon request.After that,simulate this proposed method and analyze the performance of this algorithm on behalf of the results(E2E delay,execution time,and acceptance ratio).This algorithm enables us to make numerous network slice chains and finally the results.3.Study the resilience behavior of the network system when any failure or attack on any physical server(complex network topologies)where the specific VNFs can be placed as slice and this hamper the performance of the system.In the simulation,the result carried out by the performance of the normal state of the network and performance while the attack occurs in various nodes in the physical network.