AI-Based DRX Wake-Up Control for Downlink Traffic Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile communication systems face challenges in efficiently managing Discontinuous Reception (DRX) cycles due to inefficiencies in predicting downlink data traffic, leading to suboptimal power consumption and latency.
Innovation Solution
Implementing an AI/ML model in network nodes to infer the occurrence or non-occurrence of downlink data traffic within DRX cycles, allowing for dynamic DRX indications to be transmitted to user equipment, enabling precise wake-up and sleep timings based on inference results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If conventional DRX configuration is used, then power consumption is reduced, but latency increases due to inability to predict downlink data traffic
Solution Approach 1:
The patent applies preliminary action by using AI/ML models to predict future downlink data traffic patterns before they occur. The network node performs inference about upcoming traffic bursts and proactively adjusts DRX configurations accordingly, allowing the user equipment to wake up in advance or prepare for incoming data, thereby reducing latency while maintaining power efficiency.
Solution Approach 2:
The patent implements feedback mechanisms where the network node continuously monitors actual traffic patterns and compares them with AI/ML predictions. Based on this feedback, the system dynamically adjusts DRX parameters and refinement parameters to optimize the balance between power consumption and latency performance over time.
2Measurement precision
If AI/ML model is deployed for traffic prediction, then DRX cycle management accuracy is improved, but device complexity increases
Solution Approach 1:
The patent uses an intermediary approach by introducing AI/ML models as a mediator between the network node and the DRX mechanism. Rather than directly implementing complex prediction algorithms in the user equipment, the system employs network-side AI/ML services that process traffic data and return refined DRX indications, thereby achieving high accuracy while keeping device complexity manageable.
Solution Approach 2:
The patent applies segmentation by dividing the DRX management function into separate components: traffic data collection, AI/ML inference, and DRX parameter adjustment. This modular approach allows each component to be optimized independently and simplifies the overall system architecture while maintaining high management accuracy.
3Use of energy by moving object
If dynamic DRX indications are transmitted based on AI inference, then power consumption is optimized, but loss of information may occur due to prediction errors
Solution Approach 1:
The patent applies beforehand cushioning by incorporating refinement parameters that provide a buffer against prediction errors. When AI/ML inference indicates no data traffic, the system can still maintain a refined DRX configuration that allows for quick wake-up if data arrives, thus cushioning against the potential loss of information while still achieving power savings from reduced wake-up frequency.
Solution Approach 2:
The patent utilizes parameter changes by dynamically adjusting multiple DRX parameters (on-duration timer, inactivity timer, sleep timer) based on AI inference results. The system can modify these parameters to create a more conservative DRX configuration when prediction confidence is low, thereby reducing information loss risk while maintaining power consumption optimization.
Data Source
AI summary
The present disclosure relates to a communication control method in a mobile communication system. The communication control method includes inferring, by a base station, occurrence or non-occurrence of downlink data traffic in a next DRX on-duration, by using an AI/ML model. The communication control method also includes transmitting, to a user equipment by the base station, a first dynamic DRX indication indicating one of wake-up in the next DRX on-duration or sleep in the next DRX on-duration based on an inference result for occurrence or non-occurrence of the data traffic.