Network Slicing is identified as one of the top eight practical applications of Artificial Intelligence (AI) in the telecommunications industry, serving as a critical technology for managing the complexities of 5G and 6G infrastructures. It involves creating multiple virtual networks on a single shared physical infrastructure, allowing operators to cater to diverse requirements without building separate physical networks.
The role and impact of network slicing within several key contexts is as follows:
1. AI-Driven Management and Orchestration
AI is essential for the end-to-end orchestration of network slices across multi-domain environments that combine 5G, edge computing, and cloud computing.
- Intelligent Slice Management: AI addresses the specific requirements of individual slices, such as differing security clearances, bandwidth, and speeds.
- SLA Management: It supports the management of Service Level Agreements (SLAs) for each slice, ensuring that the performance standards promised to specific customers or services are consistently met.
- Dynamic Resource Allocation: AI enables capacity planning and traffic forecasting, allowing operators to automate network adjustments to reduce latency and improve the user experience.
2. Support for 5G Service Categories
Network slicing is particularly vital for the three main categories of 5G services, where AI facilitates the distinct performance needs of each:
- Enhanced Mobile Broadband (eMBB): High-bandwidth applications.
- Massive Machine Type Communications (mMTC): Large-scale IoT deployments.
- Ultra-Reliable Low Latency Communications (uRLLC): Mission-critical services requiring near-instantaneous response times.
3. Customisation and Security
AI-powered platforms, such as those developed by the startup Trento Systems, allow operators to create virtual networks that are highly customised based on time, location, device, and service type. This level of customisation provides security and low-latency benefits that traditional internet services cannot offer, making it a cornerstone for Software-Defined Networks (SDNs).
4. Strategic and Market Context
The broader context of network slicing is tied to the “explosion” of endpoints as 5G and IoT technologies become prevalent.
- Efficiency: By using AI to forecast future demand, operators can proactively adjust slices, which optimises Quality of Service (QoS) and resource usage.
- Growth Driver: The rollout of 5G is creating significant new opportunities for AI applications in traffic management and resource allocation through slicing.
- Autonomous Operations: The industry is moving toward autonomous solutions where networks can self-configure and optimise through slicing with minimal human intervention.
Analogy for Network Slicing Think of a traditional network as a massive single-lane motorway where every vehicle—from a slow-moving tractor (a simple IoT sensor) to a high-speed emergency ambulance (a surgery-assisting robot)—must share the same space. Network Slicing is like using AI to instantly transform that motorway into a multi-lane highway where one lane is reserved exclusively for high-speed emergency vehicles, another is for heavy freight, and another for regular commuters. Each lane has its own speed limit and security guards, ensuring that a traffic jam in the “commuter lane” never slows down the “ambulance lane.”
Craig Miles.
Founder & Director at Yesway Communications | Wireless Technology, Training & Two-Way Radio Solutions | Advancing Inclusive & Global Education Through Innovation