RAN Compute QoS Architecture for Low Latency Augmented Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems, particularly in 4G and 5G networks, face challenges in providing quality of service (QoS) for augmented computing in radio access networks (RAN), especially with the increasing complexity and demand for low latency, which is not adequately addressed by existing solutions.
Innovation Solution
The introduction of a RAN Compute QoS architecture and modeling system that defines compute plane functions, including compute client service function at the UE side and compute service function at the RAN network side, to support QoS modeling and flow mapping, ensuring dynamic workload migration and execution with desired QoS characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing wireless communication systems are used, then network coverage and basic connectivity are maintained, but quality of service for augmented computing with low latency requirements is not adequately provided
Solution Approach 1:
The patent segments the QoS management into distinct functional components: compute plane functions for workload management, data plane functions for data transmission, and control plane functions for signaling. This segmentation allows specialized handling of low-latency compute traffic separate from general data traffic, improving reliability for augmented computing applications.
Solution Approach 2:
The patent introduces intermediary elements including QoS flow mapping mechanisms that act as mediators between application layer requirements and network layer transmission, and edge computing nodes that serve as intermediaries between user equipment and core network, enabling low-latency processing while maintaining overall network connectivity.
2Adaptability or versatility
If network complexity increases to support diverse applications, then service capability is enhanced, but system complexity and management difficulty increase
Solution Approach 1:
The patent implements universal QoS flow mapping mechanisms that can handle multiple application types (augmented computing, video streaming, AI training) through a common framework. The compute plane functions and QoS parameters serve multiple purposes across different services, reducing the need for separate specialized systems for each application.
Solution Approach 2:
The patent uses parameter-based QoS management where service capabilities are adjusted by changing QoS parameters (priority levels, latency budgets, resource allocation) rather than modifying system architecture. This allows flexible adaptation to diverse applications while maintaining a stable underlying system structure.
3Productivity
If compute tasks are offloaded to RAN, then processing capability is enhanced, but QoS management complexity increases
Solution Approach 1:
The patent extracts QoS management functions related to compute tasks from the core network and places them in the RAN compute plane. This separation allows RAN to autonomously manage QoS for offloaded compute tasks without requiring complex end-to-end coordination, simplifying overall QoS management while enhancing processing capability.
Data Source
AI summary
This disclosure describes systems, methods, and devices related to RAN compute QoS modeling. A device may decode a compute task request message received from a user equipment (UE), the compute task request message comprising an indication of a compute task to be offloaded to the RAN and data of the compute task. The device may establish a RAN compute service function (SF) based on support initiated by a service orchestration and chaining function (SOCF). The device may establish a RAN compute bearer based on RAN compute QoS flow with the UE, wherein the RAN compute QoS flow spans between the UE, the RAN, and the RAN compute SF.