AI-Driven Dynamic Network Slice Deployment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network slice deployment in communication networks is manual, static, and time-consuming, failing to provide real-time resource allocation for dynamic events, leading to inefficient resource usage and potential service degradation.
Innovation Solution
An automated dynamic network slice deployment system using artificial intelligence, which dynamically instantiates and deploys network resources without human intervention, allowing for fast (tens of seconds to minutes) and exact dimensioning of resources based on event detection through video and network anomaly analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual network slice deployment is used, then resource configuration can be carefully planned, but deployment time becomes very long (weeks or months)
Solution Approach 1:
The patent replaces manual mechanical operations with automated systems including AI/ML models for event detection, automatic slice template selection, and orchestration platforms that execute deployment without human intervention, reducing deployment time from weeks/months to minutes while maintaining configuration accuracy through algorithmic resource dimensioning
Solution Approach 2:
The system performs preliminary actions by pre-defining slice templates with resource configurations for different event types, and using AI models to detect events and automatically select appropriate templates before actual deployment is needed, enabling rapid response without sacrificing configuration precision
2Stability of the object's composition
If static resource allocation is used for network peaks and valleys, then network stability is maintained, but real-time resource dimensioning for dynamic events is not possible
Solution Approach 1:
The patent implements dynamic resource allocation by using AI/ML models to detect events in real-time, automatically selecting slice templates based on current event conditions, and orchestrating resource deployment dynamically, allowing the network to adapt to changing conditions while maintaining stability through controlled automation
Solution Approach 2:
The system incorporates feedback mechanisms where AI models continuously monitor network conditions and events, automatically adjust slice template selection and resource allocation based on detected patterns, and refine resource dimensioning decisions through learned insights, enabling both stability and real-time adaptability
3Productivity
If automated event detection is implemented, then deployment speed increases to tens of seconds or minutes, but system complexity increases
Solution Approach 1:
The patent segments the automated deployment system into distinct functional modules: AI/ML event detection components, slice template selection logic, resource dimensioning algorithms, and orchestration execution layers, allowing each component to be independently developed, tested, and maintained while working together to achieve rapid deployment
Solution Approach 2:
The system introduces intermediary components including pre-defined slice templates that act as intermediaries between event detection and resource allocation, and orchestration platforms that mediate between automated decisions and actual network deployment, simplifying the overall system architecture while enabling rapid automated response
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The method includes receiving first video data and network performance information from at least one first agent node associated with at least one first camera, determining event detection information based on the first video data and the network performance information, determining a slice configuration for at least one network slice based upon the first video data, network performance information and the event detection information, and controlling an operation of the communication network by instantiating the at least one network slice based on the slice configuration information.