Real-Time Media Flow Throughput Prediction for Wireless UEs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless communication networks face challenges in predicting future network conditions for real-time media streaming, leading to suboptimal transmission rates and error correction in mobile environments, where rapid changes in link quality and user movements result in delayed and inaccurate feedback for adapting video bitrate and FEC redundancy.
Innovation Solution
A method and system that collect radio network statistics from multiple UEs in a cell to predict future throughput and packet loss for a specific UE, providing this information to the sender for selecting suitable transmission rates and error correction, using machine-learning algorithms to build intelligence from historical data for accurate and timely predictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If receiver-based feedback methods are used to adapt transmission rates and error correction, then the system can respond to actual network conditions, but the prediction is delayed and inaccurate due to rapid changes in link quality and user movements
Solution Approach 1:
The system performs preliminary actions by collecting radio network statistics and training machine learning models in advance, before real-time media streaming begins. Historical data from multiple UEs is gathered and processed to build prediction models that can quickly estimate future network conditions without waiting for delayed feedback loops.
Solution Approach 2:
The system prepares prediction models and statistical data in advance to cushion against the rapid changes in link quality. By having pre-trained models and historical statistics ready, the system can quickly adapt to network condition changes without the delay of real-time feedback, effectively cushioning the impact of rapid fluctuations.
2Measurement precision
If machine-learning algorithms are used to predict future network conditions, then prediction accuracy is improved, but computational complexity and data processing requirements increase
Solution Approach 1:
The system introduces an intermediary layer consisting of machine learning models that mediate between raw radio network statistics and transmission parameter decisions. These models process historical data and generate predictions, acting as an intelligent intermediary that translates complex statistical patterns into actionable insights for rate adaptation and error correction.
Solution Approach 2:
The system creates simplified copies or representations of complex network behavior through trained machine learning models. Once trained on historical data, these models serve as lightweight copies that can quickly predict future conditions without requiring complex real-time analysis of all underlying statistical variables.
3Reliability
If radio network statistics from multiple UEs are collected and processed, then prediction reliability is improved, but the amount of data to be processed and stored increases
Solution Approach 1:
The system extracts only the most relevant features and statistics from the collected radio network data, rather than processing all raw data. The machine learning models are trained to identify and utilize key predictive features from historical statistics, extracting essential information while discarding redundant data to maintain prediction reliability with reduced data volume.
Solution Approach 2:
The system applies local quality by focusing on cell-specific and UE-specific characteristics that are most relevant for prediction. Rather than uniformly processing all data from all UEs, the system identifies and processes locally relevant statistics for each cell and UE, improving prediction reliability for specific contexts while reducing overall data processing requirements.
Data Source
AI summary
A method performed by a user equipment (UE) wirelessly connected to a wireless communication network for facilitating transmission of a real-time media flow from a sender towards the UE is described. The method includes sending a request to the network, requesting a packet loss prediction and a throughput prediction for transmission of a real-time media flow from the network towards the UE. The method also includes receiving, from the network, information of a predicted UE throughput and a predicted UE packet loss predicted by the network in response to the sent request. In some embodiments, the method also includes sending the received information to a sender of the real-time media flow to the UE.