AI-Guided PDCP Traffic Splitting Across RLC Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face inefficiencies and latencies due to the inefficient splitting of protocol data units (PDUs) across radio link control (RLC) entities, particularly when one RLC entity experiences degradation, leading to unnecessary retransmissions and reordering delays.
Innovation Solution
Implementing a learning model, such as artificial intelligence (AI) or machine learning (ML), to dynamically adjust threshold values for PDU splitting and routing across RLC entities, enabling efficient monitoring of congestion and resource utilization, thereby mitigating retransmissions and delays.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PDUs are split across multiple RLC entities, then data transmission reliability is improved, but system complexity increases due to routing decisions
Solution Approach 1:
The PDCP entity autonomously monitors uplink data volume and performs routing decisions to multiple RLC entities without requiring complex external control mechanisms. The system serves itself by internally managing the split bearer routing based on real-time data volume conditions, thereby improving reliability while keeping the control architecture relatively simple.
Solution Approach 2:
The patent implements dynamic routing where the PDCP entity can adaptively switch between single RLC entity routing and multi-RLC entity splitting based on uplink data volume thresholds. This dynamic behavior allows the system to optimize for reliability when needed (high data volume) while maintaining simplicity when not needed (low data volume), resolving the contradiction between reliability and complexity.
2Productivity
If AI/ML models are used to dynamically adjust routing parameters, then resource utilization is improved, but processing complexity increases
Solution Approach 1:
The patent introduces an AI/ML model as an intermediary component that processes uplink data volume information and outputs optimized routing parameters. This intermediary handles the complex processing tasks separately from the core PDCP routing logic, allowing the main system to benefit from intelligent resource utilization while the complexity is isolated to the AI/ML module, which can be trained and updated independently.
3Productivity
If multiple RLC entities are used for PDU routing, then data transmission efficiency is improved, but reordering delays increase
Solution Approach 1:
The patent applies preliminary ordering at the PDCP entity where PDUs are pre-numbered and tracked before being routed to different RLC entities. This preliminary action establishes the correct sequence information in advance, allowing the receiving end to reorder packets efficiently without waiting for all packets to arrive, thereby reducing reordering delays while maintaining high transmission efficiency through parallel RLC entity usage.
Data Source
AI summary
Methods, systems, and devices for wireless communications are described. A first wireless device may select a value to use as a data volume threshold value for determinations of routing or splitting of data traffic across radio link control (RLC) entities. The selection by the first wireless device of the data volume threshold for data splitting by a protocol data convergence protocol (PDCP) layer may be based on an output of a learning model. In some examples, the first wireless device may be configured to override a data volume threshold value indicated by a network entity and transmit data over a secondary RLC entity regardless of data volume, or the first wireless device may be configured to disable a primary RLC path for a duration, among other embodiments.