5G Deterministic Scheduling for Low-Jitter TSN QoS Flows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing method of simulating a 5G system as a bridge in a Time-Sensitive Networking (TSN) network results in large delay variations and inefficient resource allocation due to best-effort data transmission, leading to unpredictable Quality of Service (QoS) for TSN flows.
Innovation Solution
Implementing a deterministic transmission method where network elements in the 5G system, such as the radio access network and UPF, generate and exchange capability information to dynamically balance network resources, using factors like residence time, signal-to-noise ratio, and bandwidth to reduce delay variations and ensure deterministic service transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If best effort transmission is used in RAN and UPF, then device complexity is reduced, but delay variation increases and transmission reliability deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-calculating and pre-distributing scheduling policies to network elements before data transmission begins. The centralized network configuration element computes deterministic scheduling policies based on network topology and service requirements, then distributes them to RAN and UPF in advance. This allows network elements to execute predetermined scheduling actions without real-time complex calculations, reducing delay variation while maintaining transmission reliability.
Solution Approach 2:
The patent introduces a centralized network configuration element as an intermediary that mediates between service requirements and network resource allocation. This intermediary computes deterministic scheduling policies and distributes them to distributed network elements (RAN, UPF), enabling coordinated deterministic transmission without requiring complex peer-to-peer negotiation between network elements. The intermediary consolidates the complexity of deterministic scheduling while simplifying individual node operations.
2Reliability
If deterministic scheduling policy is implemented, then delay variation is reduced, but network element complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the deterministic scheduling function into two independent parts: policy computation (centralized network configuration element) and policy execution (distributed RAN and UPF). This segmentation allows the complex computational task of determining scheduling policies to be performed centrally, while individual network elements only need to execute simple predetermined instructions. Each network element maintains low complexity by relying on the centralized element for policy generation.
Solution Approach 2:
The centralized network configuration element serves as an intermediary that handles the complexity of deterministic scheduling computations. It receives service requirements, calculates appropriate scheduling policies, and distributes them to network elements. This intermediary absorbs the computational complexity, allowing individual network elements to maintain simple execution logic while achieving deterministic transmission guarantees.
3Adaptability or versatility
If network elements operate independently, then system flexibility is improved, but overall link situation awareness deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where network elements (RAN, UPF) report their local transmission status, queue lengths, and resource availability to the centralized network configuration element. This feedback loop enables the centralized element to maintain global awareness of the network state while individual elements retain operational autonomy. The centralized element uses this feedback to compute and update scheduling policies that optimize overall network performance while respecting local constraints.
Solution Approach 2:
The centralized network configuration element acts as an intermediary that aggregates information from distributed network elements and distributes coordinated scheduling decisions back to them. This intermediary consolidates local information into global network state awareness, then translates global optimization goals into local execution instructions. Network elements maintain flexibility by executing locally-adapted policies while the intermediary ensures overall link situation awareness through centralized information aggregation and coordination.
Data Source
AI summary
This application discloses a deterministic transmission method, a communication apparatus, and a storage medium, and relates to the communication field. In the method, a radio access network device may generate deterministic transmission capability information of a first switching node, and send the deterministic transmission capability information to a first network element. The first switching node includes the radio access network device and a terminal. Then, the radio access network device may receive a first scheduling policy that is delivered by the first network element based on the deterministic transmission capability information of the first switching node, so that deterministic transmission can be performed according to the first scheduling policy. This reduces a delay variation of a QoS flow carrying a TSN flow in a 5GS, and can avoid a need to select a QoS flow whose delay indicator is excessively high for the TSN flow.