ML-Based xHaul Transport Domain Orchestration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining which xHaul transport domain to orchestrate, manage, and control in Next Generation mobile networks to meet specific service requirements of customers is a complex task, as each domain has unique characteristics and varying demands.
Innovation Solution
The use of machine learning and optimization techniques to select a particular xHaul transport domain for orchestration and management, based on Service Requirements Profiles, Service Slice Infrastructure Design Profiles, and xHaul Characterization Profiles, ensures that transport performance meets customer needs by optimizing the configuration of fronthaul, midhaul, and backhaul domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple xHaul transport domains are managed with different characteristics and varying demands, then service requirements can be met more accurately, but system complexity increases
Solution Approach 1:
The patent segments the transport network into multiple xHaul domains (fronthaul, midhaul, backhaul) with distinct characteristics and management approaches. Each domain is treated as a separate entity with its own optimization criteria, allowing tailored management that meets specific service requirements while maintaining overall system coordination through the orchestration framework.
2Productivity
If machine learning and optimization techniques are used to select transport domains, then resource allocation efficiency improves, but computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-processing service requirement profiles, infrastructure design profiles, and xHaul characterization profiles before actual transport domain selection occurs. Machine learning models are trained in advance on historical data, and optimization algorithms prepare candidate solutions beforehand, reducing the computational burden and processing time during dynamic resource allocation decisions.
3Adaptability or versatility
If dynamic resource allocation is implemented to meet changing service demands, then network flexibility improves, but control complexity increases
Solution Approach 1:
The patent implements feedback mechanisms where the orchestration system continuously monitors service requirements, network performance, and resource utilization. Based on this feedback, the machine learning models dynamically adjust transport domain selections and resource allocations. This closed-loop control enables flexible adaptation to changing demands while managing complexity through automated decision-making based on real-time information.
Data Source
AI summary
A network device obtains service requirements associated with a customer identifier, obtains a first profile describing an infrastructure design of multiple transport domains associated with at least one network slice of a network, and obtains a second profile describing performance characteristics of the multiple transport domains of the at least one network slice. The network device receives training data associated with performance measurements of the multiple transport domains of the at least one network slice, and updates a machine learning model based on the training data. The network device selects at least one of the multiple transport domains for orchestration using the updated machine learning model, the service requirements, the first profile, and the second profile.