Network Slicing
The Future of Virtualised 5G Infrastructure
Network slicing is the process of creating multiple virtual networks on a single
shared physical infrastructure. Each "slice" is logically isolated and can be
customised to meet the specific requirements of a particular application, service, or
customer.
Conceptual Flow
Physical Infrastructure ➔ Virtual Slices ➔ Optimised
Services
Why It Matters
Cost Efficiency: Operators can run diverse services on the same set of hardware
equipment, significantly reducing Capital Expenditure (CAPEX).
Flexibility: Custom networks can be spun up or down dynamically to serve specific
temporal needs or geographic regions.
Speed: New services can be deployed in minutes through software configuration
rather than months of hardware installation.
Page 1
3 Main Slice Types
The 5G standard defines three primary categories of network slices, each designed
for a specific performance profile.
eMBB URLLC mMTC
Enhanced Mobile Ultra-Reliable Low Massive Machine Type
Broadband Latency
eMBB (Enhanced Mobile Broadband): High speed and massive capacity. This slice
is used for data-hungry applications like 4K/8K streaming, high-definition video
conferencing, and Augmented/Virtual Reality.
URLLC (Ultra-Reliable Low Latency Communications): Designed for mission-
critical tasks where delay cannot be tolerated. Applications include autonomous
vehicles, remote surgery, and industrial automation.
mMTC (Massive Machine Type Communications): Engineered to support a vast
number of devices (up to 1 million per sq km) sending small amounts of data.
Perfect for Smart City sensors and utility meters.
Page 2
How It Works - Technology Stack
Network slicing relies on a combination of advanced virtualisation technologies
working together to decouple software from hardware.
Key Technologies
1. NFV (Network Functions Virtualisation): Decouples network functions (like
firewalls or routers) from proprietary hardware, running them as software on
standard servers.
2. SDN (Software Defined Networking): Provides the "brain" or controller that
manages the flow of data across the virtual slices through software
commands.
3. Cloud & Edge Computing: Allows data processing to happen closer to the
user (at the Edge) for lower latency or in centralized data centers for scale.
Page 3
Real-World Examples
Consider a modern Smart Hospital. Instead of a one-size-fits-all Wi-Fi or cellular
connection, the hospital can use three distinct slices:
Hospital Use Case
mMTC Patient Monitoring: Connects thousands of wearable heart-rate
monitors and bed sensors. Data is small but constant.
eMBB Medical Records: Allows doctors to download huge 3D MRI scans
and high-res imaging instantly at the bedside.
URLLC Remote Surgery: Connects a specialist in another city to a surgical
robot with zero delay and a "never-fail" connection.
End of Document - Network Slicing Overview
Page 4