RLC Polling Control Using Predictive Models to Cut Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges with UE memory crunch and excessive latency due to restrictive RLC polling rules, which limit the ability to transmit polling requests, leading to inefficient resource allocation and network performance issues.
Innovation Solution
Implementing an artificial intelligence/machine learning model for dynamically or semi-statically adjusting RLC polling procedures based on a predictive model, allowing UEs to transmit polling requests in accordance with a configured polling strategy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If restrictive RLC polling rules are used, then device complexity is reduced, but latency increases and memory efficiency deteriorates
Solution Approach 1:
The patent applies dynamics by transitioning from static, restrictive polling rules to dynamic, model-based polling decisions. The machine learning model continuously adapts polling behavior based on real-time network conditions, UE state, and traffic patterns, allowing the system to optimize latency without increasing complexity in a fixed manner.
Solution Approach 2:
The patent replaces the mechanical, rule-based polling mechanism with an intelligent system using machine learning models. Instead of following predetermined restrictive rules, the system uses predictive models to intelligently determine when and how to poll, substituting rigid mechanical control with adaptive intelligent control.
2Device complexity
If restrictive RLC polling rules are used, then device complexity is reduced, but memory efficiency deteriorates
Solution Approach 1:
The system dynamically adjusts polling behavior based on current memory conditions and network state, allowing optimal memory utilization. The machine learning model learns from historical data to predict when polling will be beneficial, preventing both premature polling (wasting memory) and delayed polling (causing memory crunch).
Solution Approach 2:
The patent replaces rigid mechanical polling rules with an intelligent system that adapts to memory conditions. The machine learning model substitutes fixed rule-based control with adaptive decision-making that optimizes memory efficiency without requiring complex manual configuration.
3Productivity
If dynamic polling adjustment based on predictive model is implemented, then network performance is improved, but device complexity increases
Solution Approach 1:
The patent introduces machine learning models as intermediaries between the network conditions and polling decisions. These models act as mediators that process complex inputs (network state, traffic patterns, UE conditions) and translate them into simple polling actions, hiding the complexity from the core RLC procedure while maintaining high network performance.
Solution Approach 2:
The system employs self-service through machine learning models that automatically learn and adapt to optimal polling strategies without manual intervention. The models continuously improve their performance by learning from historical data, enabling the system to maintain high productivity while the complexity is managed autonomously by the intelligent algorithms.
Data Source
AI summary
Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive a polling configuration for dynamically or semi-statically adjusting a radio link control (RLC) polling procedure in accordance with a predictive model. The UE may transmit a polling request to a network node in accordance with the polling configuration. Numerous other aspects are described.