ITS Network Node Load Prediction for Proactive MEC Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MEC-based V2X services face challenges in accurately and efficiently managing dynamic resource allocation due to varying traffic densities and object movements, leading to potential safety risks and inefficiencies in computing resource management.
Innovation Solution
A method for predicting processing loads by sharing object status monitoring metrics between network nodes, allowing for proactive adjustment of resources based on location, direction, speed, and detection priority of moving objects, and implementing auto-scaling policies to dynamically allocate computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reactive resource allocation is used, then system complexity is reduced, but resource availability and response time deteriorate due to several minutes scaling delay
Solution Approach 1:
The patent implements predictive resource allocation by analyzing historical traffic data and object movement patterns to forecast future resource requirements. This preliminary action allows the system to provision resources before actual demand occurs, eliminating the several-minute scaling delay inherent in reactive systems while maintaining reliable resource availability.
Solution Approach 2:
The system continuously monitors actual resource utilization and compares it with predicted values, using this feedback to refine future predictions and adjust resource allocation dynamically. This closed-loop feedback mechanism ensures that the predictive system adapts to changing traffic patterns while maintaining optimal resource availability without excessive provisioning.
2Reliability
If proactive predictive resource allocation is implemented, then resource availability improves, but system complexity increases due to machine learning algorithms and continuous monitoring
Solution Approach 1:
The patent divides the complex predictive resource allocation system into distinct functional modules: data collection modules that gather traffic and object information, prediction modules that apply machine learning algorithms to forecast resource needs, and resource provisioning modules that execute allocation decisions. This segmentation manages system complexity by making each component independent and maintainable while achieving reliable resource availability through their coordinated operation.
Solution Approach 2:
The system introduces intermediary components including buffer storage for historical data, prediction intermediaries that translate raw data into resource requirements, and coordination layers that manage communication between modules. These intermediaries simplify the overall system architecture by decoupling complex prediction algorithms from resource management operations, reducing operational complexity while maintaining high resource availability.
3Reliability
If resources are allocated based on peak demand, then service reliability improves, but resource utilization efficiency deteriorates due to idle resources during low traffic periods
Solution Approach 1:
The patent implements dynamic resource allocation that continuously adjusts resource provisioning based on real-time traffic conditions and predictive analytics. Rather than static peak-demand allocation, the system dynamically scales resources up during high-traffic periods and down during low-traffic periods, maintaining service reliability while optimizing resource utilization efficiency through adaptive, condition-based provisioning.
Solution Approach 2:
The system changes key allocation parameters including resource provisioning levels, scaling thresholds, and prediction time horizons based on observed traffic patterns and service requirements. By dynamically adjusting these parameters rather than fixing them at peak values, the system maintains adequate service availability while improving resource utilization efficiency through context-appropriate parameter selection.
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
A method by which a first network node transmits information to a second network node in an intelligent transportation system (ITS), according to various embodiments, may comprise: processing information on first objects moving in a first area monitored by the first network node; on the basis of the processing result, obtaining first prediction information on a processing load expected to be caused by the first objects in a second area at least partially contacting the first area; and transmitting the first prediction information to the second network node monitoring the second area.