MEC Cluster Selection via Latency and Bandwidth Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Multi-Access Edge Computing (MEC) systems lack the capability to effectively analyze various criteria beyond geographic proximity to determine which MEC clusters can satisfy Service Level Agreements (SLAs) for application services, leading to inefficiencies in latency and resource utilization.
Innovation Solution
An orchestration system that collects and analyzes transport network parameter data, including latency, bandwidth, and capability information from MEC clusters, to dynamically select the most suitable MEC cluster for an application session based on predicted parameters and SLA requirements, implementing artificial intelligence for end-to-end latency calculation and load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MEC resources are designated based primarily on geographic proximity, then the system is simple to operate, but it cannot effectively satisfy Service Level Agreements for latency and performance requirements
Solution Approach 1:
The patent changes the selection parameters from simple geographic proximity to multiple parameters including end-to-end latency, bandwidth, and capability information. The orchestration system collects transport network parameter data and uses these changed parameters to select MEC clusters that satisfy SLAs while maintaining manageable complexity through automated AI-based analysis.
2Reliability
If multiple transport networks are used for MEC services, then service reliability is improved, but network selection complexity increases
Solution Approach 1:
The orchestration system implements feedback by continuously collecting transport network parameter data from multiple networks, analyzing performance metrics, and using this feedback to dynamically select the most suitable MEC cluster. This feedback mechanism enables reliable service across multiple networks while managing selection complexity through automated performance-based decision-making.
Solution Approach 2:
The orchestration system acts as an intermediary between UE devices and multiple MEC clusters across different transport networks. It collects capability information and transport network parameters, performs AI-based analysis, and makes selection decisions, thereby enabling multi-network service reliability while abstracting the complexity from the selection process.
3Loss of time
If AI-based analysis of transport network parameters is implemented, then latency optimization is improved, but computational overhead increases
Solution Approach 1:
The orchestration system serves as an intermediary that performs AI-based analysis of transport network parameters and capability information. By centralizing this computational overhead in the orchestration system rather than distributing it across all devices, the patent optimizes end-to-end latency through intelligent selection while managing computational resources efficiently.
Data Source
AI summary
A device, method, and system provide for collecting service parameters associated with application sessions associated with a plurality of MEC clusters; obtaining capability information associated with each of the plurality of MEC clusters; generating service profiles for a set of transport networks based on the service parameters and the capability information, wherein each transport network includes at least one of the plurality of MEC clusters; receiving, from a user equipment (UE) device, a request for a MEC service having a minimum service requirement; and selecting, based on the service profiles, a MEC cluster from the plurality of MEC clusters to provide the MEC service, wherein the selected MEC cluster is included in a first transport network that meets the minimum service requirement.