Network Slicing for 5G and Beyond Networks

Choong Seon Hong

Department of Computer Science and EngineeringKyung Hee University

2

Chapter 1: 5G Networks• Background and Motivation• 5G Deliverables• 5G Networks: Industrial efforts • Challenges to realize 5G Networks

2/85

Background

• Traffic growth due to:• Tsunami of heterogeneous connections:

• Smartphones

• Connected vehicles

• Wearables devices

• IoT sensors

• And so on…

• Novel bandwidth hungry applications:

• Real time HD streaming

• Online Gaming

• Ultra-reliable and low-latency

communication

• Virtual reality services

• Enhanced mobile broadband

• And so on…

Source: Cisco Visual Networking Index (VNI), Feb. 2017.

3/85

Bottleneck and New Paradigms in 5G

• Network capacity is a

bottleneck due to:• Radio Access Networks (RANs)

• Mostly wireless and highly dynamic

• New paradigms to support:• Small cell (SC) deployment

• Device-to-device(D2D)

• Network virtualization

• LTE-unlicenced

• And so on…

.Source: http://www.eurescom.eu/news-and- events/eurescommessage/eurescom-message-1-2014/3gpp-system-standards-heading-into-the-5g-era.html.

4/85

5G Deliverables

5G Deliverables: • Higher data rates

• Reduced end-to-end latency

• Higher energy efficiency

• Better network coverage

• Enhanced security

• Ultra reliability

• and so on…

Source: “5G Use Cases and Requirements,” a white paper from Nokia.

5/85

• KT & SK-Telecom, Korea:

• Successfully collaborated with Samsung Electronics to develop a 5G end-to-end network that includes:

• 5G virtualized core

• Virtualized RAN

• Distributed Unit (baseband unit and radio unit)

• Test device - based on the 3GPP 5G New Radio (5G NR)

5G Networks: Industrial efforts 6/85

• Nokia, Finland:

• Actively focusing for providing 5G services such as:

• 5G mobility service supporting enhanced mobile broadband (eMBB)

• 5G mobility service supporting ultra-reliable and ultra-low latency communications (URLLC)

• Huawei, China:

• Huawei is actively working to enhance the antenna capabilities for 5G networks

• It has released its new FDD antenna and FDD/TDD converged antenna platforms

5G Networks: Industrial efforts 7/85

• Ericsson, Sweden:

• Ericsson’s is actively participating in the development of 5G networks

• It has recently developed the Ericsson’s 5G radio test-bed that comprises of Massive MIMO, multi-user MIMO, and beamforming technologies

• Ericsson has introduced a new radio product, AIR 3246, for Massive Multiple Input Multiple Output (Massive MIMO).

• This launch will enable operators – especially in metropolitan areas – to bring 5G to subscribers using today’s mid-band spectrum and boost capacity in their LTE networks.

5G Networks: Industrial efforts 8/85

5G cellular networks were assumed to be the key enabler and infrastructure provider in the ICT industry, by offering three types of services:

• Enhanced mobile broadband (eMBB)

• Ultra-reliable low latency service (URLLC)

• Massive machine-type communications (mMTC)

5G Promises

http://www.itu.int/dms_pubrec/itu-r/rec/m/R-REC-M.2083-0-201509-I!!PDF-E.pdf

Enhanced

Mobile Broadband

Massive Machine type

CommunicationsUltra-reliable and low

latency communications

Smart City

Gigabytes in a second

3D video, UHD screens

Work and Play in the cloud

Augmented reality

Industry automation

Self driving car

Mission critical application

Smart Home

Voice

Future IMT

9/859/85

• The existing mobile network architecture was designed to meet requirements for voice and conventional mobile broadband (MBB) services

• To meet the requirements of novel bandwidth hungry services, there is a need to deploy smarter 5G networks

• Note that, all novel services have very diverse requirements, thus having traditional RAN and core solutions for every service cannot guarantee the QoS

Challenges to realize 5G Networks (1)

How to fulfil the diverse 5G networks requirements?

10/85

• Generally, to realize diverse 5G use cases there is need to resolve the given challenges:

• Scalability and Reliability

• Interoperability

• Sustainability

• Network Slicing

• Security

• Integration of AI in 5G

Challenges to realize 5G Networks (2) 11/85

Chapter 2: Network Slicing: The Concept• Network Slicing• Key Enablers• Network Slicing: Industrial Efforts

12/85

Network Slicing (1) 13/85

• Network Slicing enablers: How to do it ?• Software-defined networking (SDN)

• Network Functions Virtualization (NFV)

• A single physical network will be sliced into multiple virtual networks:• Different service types running across each virtual network, i.e., URLLC, EMBB, etc.

• Support different radio access networks (RANs), i.e., LTE, Wi-Fi, etc.

• It is envisaged that network slicing will be used to partition the core network and radio access networks

Network Slicing (2) 14/85

• Network Slicing Principles

• Slice Isolation

• Elasticity

• End-to-End Customization

• Network Slicing Key Enablers

• Software defined networking

• Network function virtualization

Network Slicing (3) 15/85

Network slicing enablers: Software defined network (SDN)

API to the data plane

(e.g., OpenFlow , ONOS)

Decentralized control plane

(which is closely tied to data

planes)

API to SDN application programmer

(who can now program the network as a system and not as a collection of

individual boxes)

Logically-centralized DP-

decoupled control

SDN Controller

At the highest level, the SDN movement is an effort to build networks you can

program at a higher level of abstraction— just as you can program a computer.

16/85

SDN enables programmability

Vertically integrated

Closed, proprietary

Slow innovation

Horizontal

Open interfaces

Rapid innovation

Specialized

Control

Plane

Specialized

Hardware

Specialized

Features

AppAppAppAppAppAppAppAppAppAppApp

Control

Plane

Control

Plane

Control

Plane or or

Open Interface

Merchant

Switching Chips

Open Interface

SDN enables programmability which is important for network slicing

17/85

• A network architecture concept that uses the technologies of IT virtualization to virtualize entire classes of network node functions that may connect, or chain together, to create communication services

• NFV is envisioned to play a crucial role in network slicing as it will be responsible to build isolated slices based on user service requirements

Network slicing enablers: Network function virtualization (NFV)

WindowsWindows

Physical Hardware

Virtualization

Virtual

Compute

WindowsWindowsVirtual

Storage

WindowsWindowsVirtual

Network

Mobility

Management

Entity (MME)

Serving Gateway

(S-GW)

Policy and

Charging Rules

Function (PCRF)

Man

agem

ent an

d O

rchestratio

n

18/85

• Due to massive success of NFV and SDN in wired domain, a number of studies are being conducted to adopt them both in the core and radio access networks (RANs) for future cellular networks such as:

• CORD (Central Office Re-architected as a Datacenter) [1]

• Radisys M-CORD [2]

• Wireless network virtualization (WNV) is a novel concept for virtualizing the RANs of future cellular networks

• WNV has a very broad scope ranging from spectrum sharing, infrastructure virtualization, to air interface virtualization

Wireless Network Virtualization

[1] https://opencord.org/[2] http://www.radisys.com/radisys-m-cord-open-platform-emerging-5g-applications

19/85

• Network Slicing certainly is one of the most discussed technologies these days. Network operators like KT, SK Telecom, China Mobile, DT, KDDI and NTT, and also vendors like Ericsson, Nokia and Huawei are all recognizing it as an ideal network architecture for the coming 5G era.

• Ericsson has been working on network slicing with NTT DOCOMO since 2014. In 2016 the two announced a successful proof of concept of dynamic network slicing technology for 5G core networks.

• They created a slice management function and network slices based on requirements such as latency, security or capacity.

Network Slicing: Industrial Efforts 20/85

Chapter 3: Resource Management for Network Slicing

• Motivation and Introduction• Network Slicing Resources• Use Case: Virtual Reality• Summary

21/85

Network Slicing Key Resources

• Radio resources

• Access network resources

• Core network resources

• Caching

• On-device caching

• Edge caching

• Core network caching

• Edge Computing Servers

• Cloudlets

• Fog servers

• Multi-access edge computing servers

22/85

Virtual Reality: A Use Case

• Virtual Reality Applications

• Smart Health-Care

• Smart Industries

• Smart Gaming

• Virtual Reality Challenges

• High Computational Power for Processing complex Algorithms

• Strict-Latency Constraints

23/85

Virtual Reality: System Model (1) 24/85

Virtual Reality: System Model (2)

Where H = Horizontal Pixels

V = Vertical Pixels

F = Frames per second

L = Video Length

• Video Size Computation

25/85

Virtual Reality: Problem Formulation

• Total transmission and processing cost minimization problem. Uplink delay

Downlink delay Processing cost

Guarantees that all users are served by network.

Maximum MEC server capacity constraint

Budge cost must not be greater than processing cost

Total time constraint variable

Total cost constraint variable

Video frames size variable

Latency limit constraint

26/85

ADMM-Based Solution (1)

• Re-write the objective function.

27/85

ADMM-Based Solution (2)

• Modified Problem.

28/85

ADMM-Based Solution (3)

• For ADMM-based solution new variable z is introduced.

29/85

ADMM-Based Solution (4)

• ADMM-Based Task Offloading Algorithm.

30/85

Performance Evaluation (1)

Average utility vs. number of MEC servers

31/85

Performance Evaluation (2)

Average Processing cost vs. number of MEC servers

32/85

• An overview of resource management for network slicing has been presented in this chapter.

• Numerous key resources for network slicing are discussed.

• A use case of virtual reality is along with its ADMM-based solution is presented.

Summary 33/85

Chapter 4: Network Slicing: Radio Resource Allocation

• Radio Resource Allocation with Single InP• Radio Resource Allocation with Multiple InP• Summary

34/85

Radio Resource Allocation with Single InP

MVNO 1 MVNO 2 MVNO V

UE 1 UE 5UE 2UE 3 UE 4

UE k

Infrastructure Provider (InP)

2

FrequencyRad

io E

lem

ent

1

FrequencyRad

io E

lem

ent

C

FrequencyRad

io E

lem

ent

Mobile virtual network operators (MVNOs)

UE NUE n

UE l

User Equipment (UEs)

35/85

Radio Resource Allocation with Single InP

• Infrastructure Provider Network Utility.

• MVNO Utility.

Price of virtual resource

Binary variable

Throughput

36/85

Problem Formulation

Virtual resource limit

Minimum rate requirement for MVNO users

Binary indicator variable

Pricing constraint

37/85

Matching-Based Solution (1)• MVNO Users Preference Profile

• MVNO Preference Profile

38/85

Matching-Based Solution (2) 39/85

Simulation Setup

Simulation Parameters Simulation Scenario

40/85

Simulation Results (1)

Number of users vs. required iterations

41/85

Simulation Results (2)

Average throughput vs. number of users

42/85

Radio Resource Allocation with Multiple InP 43/85

Problem Formulation (1)

• The goal is to maximize the overall performance of area consisting of Ues, MVNOs, and InPs.

• The UE problem is given by:

UE is served by a maximum of one MVNO

44/85

Problem Formulation (2)

• The MVNO problem is given by:

45/85

Problem Formulation (3)

• The InP problem is given by:

46/85

Hierarchical Matching-Based Solution (1) 47/85

Hierarchical Matching-Based Solution (2) 48/85

Simulation Setting

Simulation parameters

49/85

Simulation Results (1)

Average Sum-rate of hierarchical matching, general sharing, and fixed sharing schemes

50/85

Simulation Results (2)

Average Sum-rate vs. network size for varying InP-BS density

51/85

Simulation Results (3)

Average Sum-rate vs. network size for varying InP-BS Bandwidth

52/85

Simulation Results (4)

Average iterations vs. network size for varying InP-BS Bandwidth

53/85

Simulation Results (5)

Slice size for bandwidth 5MHz

54/85

Simulation Results (6)

Slice size for bandwidth 10MHz

55/85

Simulation Results (7)

Average size for bandwidth 10MHz

56/85

Summary• In this chapter, we have discussed two proposals pertaining to radio

resource allocation for network slicing.

• First one consider a single InP scenario, whereas the second one consider multiple InP scenario.

• Two different solutions based on matching theory are proposed to offer effective resource management for network slicing.

57/85

Chapter 5: Network Slicing: Radio Resource Allocation Using Non-orthogonal Multiple Access• Introduction• System Model• Problem Formulation• Solution Approach• Simulation Results

58/85

• The unprecedented growth in data traffic and the tsunami of mobile devices in the existing networks demands to consider spectrum efficiency and massive connectivity in 5G networks.

Introduction-1

The greatest limitation of the existing employed access scheme, i.e., orthogonal multiple access (OMA) scheme (i.e., the OFDMA scheme) is that the number of users that can simultaneously occupy the spectrum resources is restricted by the number of available spectrum resources, i.e., subchannels, resource blocks (RBs).

• Non-orthogonal multiple access (NOMA) has been viewed as a key enabler for catering the inconveniences of OFDMA scheme in 5G networks [1, 2].

• NOMA takes advantage of the resource gain differences through which it allows multiple users to be scheduled on a single spectrum resource.

[1] Saito, Y., Kishiyama, Y., Benjebbour, A., Nakamura, T., Li, A., & Higuchi, K. (2013). Nonorthogonal multiple access (NOMA) for cellular future radio access. In 2013 IEEE 77th Vehicular Technology Conference (VTC Spring) (pp. 1–5). Piscataway: IEEE.

[2] Song, L., Li, Y., Ding, Z., & Poor, H. V. (2017). Resource management in non-orthogonal multiple access networks for 5G and beyond. IEEE Network, 31(4), 8–14.

59/85

• NOMA technique can accommodate higher number of users in the network when compared to traditional OMA schemes.

• However, users that are scheduled over the same spectrum resource create inter-user interference over the spectrum resource.

• This problem can be solved by using the successive interference cancellation (SIC) technique at the receiver.

Introduction-2 60/85

System Model

System model of WNV in which users of different MVNOs coexist on a group of resource blocks owned by an InPs using NOMA

macro base station(MBS)

61/85

Problem Formulation

(Total power)

(Slice isolation requirement in NOMA clustering)

(Assignment of a user to at most a single cluster)

(At least two users are grouped into NOMA cluster)

(Spectrum resource)

(RB allocation, User clustering)

Joint problem of resource block allocation, power assignment problem, and user clustering for weighted sum-rate maximization

62/85

• Matching Game for User Clustering (NOMA Clustering)• The users are divided into three categories according to

channel gains, i.e., strong users, normal users, and weak users.

• Resource Allocation• There is no polynomial algorithm to solve problem because of

its integer programming nature.

• The problem is simplified through relaxing discrete variables δk into real numbers in the interval [1,Ω].

• Power Assignment• This problem apparently seems non-convex due to the non-

concave nature of the rate function.

• Thus, a successive convex approximation (SCA) is adopted to compute the optimal power assignment.

Solution Approach 63/85

Simulation Results-1

The number of iteration of matching algorithm with two schemes: JUCRAN-2 andJUCRAN-3

The number of NOMA cluster with two schemes: JUCRAN-2 and JUCRAN-3

JUCRAN-3: three different users cases (such as weak users, normal users, and strong users)JUCRAN-2 : two users classes.

64/85

Simulation Results-2

Network throughput JUCRAN vs OFDMA Network throughput JUCRAN vs OFDMA

65/85

Simulation Results-3

Network throughput vs maximum transmit power with 50 users

Energy efficiency vs maximum transmit power with 50 users

66/85

Chapter 6: Network Slicing: Cache and Backhaul Resource Allocation

• Introduction• System Model• Problem Formulation• Solution Approach• Simulation Results

67/85

• Traditionally in cellular networks radio resources were only considered as a performance bottleneck.

• A number of solutions were devised to only cater the radio resource allocation challenge.

• The proliferation of end users and novel applications have also imposed limitations on other network resources such as backhaul and cache spaces.

Introduction 68/85

System Model

Virtualized cellular network hierarchical model

69/85

Problem Formulation

(Guaranteeing the required minimum rate)

(Aggregated data rate of users)

(Total transmit power)

(Restricts the allocation of a slice to at most user)

(Isolation of the slices)

(Isolation of the slices)

(Isolation of the slices)

70/85

Matching-Based Low-Complexity Algorithm

Solution Approach-1 71/85

Matching-Based Low-Complexity Algorithm

Solution Approach-2 72/85

Simulation Results

Cost convergence

PCBD: Proximal block coordinate descentJP-ADMM: Jacobi-proximal alternating direction method of multipliers

73/85

Simulation Results

Slicing allocation of four tenantsA RAN network slicing model with three tenants

74/85

Chapter 7: Network Slicing: Dynamic Isolation Provisioning and Energy Efficiency

• System Model• Problem Formulation• Solution Approach• Simulation Results

75/85

System Model

System model: system model with InP owning the physical infrastructure and providing virtualized resources to multiple MVNOs users

76/85

Problem Formulation 77/85

Solution based on Lyapunov Online Algorithm

Solution Approach 78/85

Stackelberg Game

Solution Approach 79/85

Solution Approach

Decomposition Approach

80/85

Simulation Results

Average revenueAverage energy efficiency (bit/Joule/Hz)

Average total network throughput (bps/Hz)Average throughput (bps/Hz/user)

81/85

Simulation Results

Energy consumptionThroughput

82/85

Chapter 8: Concluding Remarks• Open Issues• Conclusion

83/85

• Dynamic Slice Allocation

• A practical system would have users arriving and leaving a system with different demands at different time slots.

• Mobility Aware Network slicing

• The current approaches for network slicing are not designed to handle mobility in the network.

• Handling and orchestrating the radio access and core network will be very challenging in case of mobility.

• Require migration of services from one point to other points in the network.

Open Issues 84/85

• This lecture is mainly focus to understand a full view of the resource management problem in 5G networks.

• We learned

• The requirements and enabling technologies of 5G networks.

• A detailed overview of network slicing that can be adapted to fulfill the 5G deliverables.

• The recent research works’ motivation, issues, challenges, and solutions.

• Some open issues for future research and their potentials.

Conclusion 85/85