Predicted User Behavior Signaling for XR Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G NR systems face challenges in efficiently managing resource allocation and energy consumption for extended reality (XR) applications, which require high-capacity and low-latency communications with varying user demands.
Innovation Solution
The proposed solution involves a method where user equipment (UE) facilitates the reception of user characteristic reporting configurations and requests from a radio access network node. The UE determines user action indications based on predicted user actions during XR sessions and transmits these indications to the radio access network node, allowing for optimized resource scheduling and energy management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the radio access network node continuously monitors and allocates resources for XR applications, then the quality of service for XR applications is improved, but the energy consumption of the network increases
Solution Approach 1:
The system performs preliminary actions by predicting future user actions and characteristics before they actually occur. The UE determines predicted user actions based on current user characteristics and reports these predictions to the network node, allowing the network to proactively allocate resources in advance rather than continuously monitoring and reacting to actual user behavior in real-time
Solution Approach 2:
The user equipment performs self-service by autonomously determining its own user characteristics, predicting its future actions, and reporting these predictions to the network. This reduces the need for continuous network-side monitoring and control, allowing the UE to manage its own resource requirements intelligently based on its usage patterns
2Ease of operation
If the network allocates more resources to accommodate varying user demands of XR applications, then the user experience is improved, but the resource utilization efficiency decreases
Solution Approach 1:
The system implements dynamic resource allocation based on predicted user behavior. Instead of static or continuously adjusted resource allocation, the network node receives predicted user characteristics and actions at specific intervals, allowing it to dynamically optimize resource allocation to match actual user needs without over-provisioning. Resources are allocated based on predicted demand rather than maximum potential demand
Solution Approach 2:
The system establishes a feedback loop where the UE reports predicted user characteristics and actions to the network node, which then uses this information to optimize resource allocation. This feedback mechanism allows the network to adapt resource allocation to actual user behavior patterns while avoiding the inefficiency of continuous monitoring and adjustment
Data Source
AI summary
An extended reality processing unit may receive from a radio access network node a user characteristic reporting configuration comprising a user characteristic indication indicative of a user characteristic, corresponding to use of an extended reality appliance during an extended reality session, that can potentially be predicted by the processing unit. The node may request, via a configuration message, reporting to the node a prediction of a user characteristic associated with the session. The request message may comprise a reporting criterion. Based on traffic corresponding to the session, the processing unit may predict values indicative of one or more user characteristics associated with the session. Upon determining that a predicted value satisfies a reporting criterion, the processing unit may transmit to the node the predicted value. The node may adjust allocation of resources usable for delivery of traffic associated with the session based on the reported predicted value.