Initial Bitrate Prediction for Stable Real-Time Communication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing real-time communication systems often suffer from suboptimal initial bitrate selection due to varying user and network conditions, leading to communication degradation such as delays and freezing, as they typically rely on fixed, predetermined values that do not accurately reflect the available bandwidth.
Innovation Solution
An initial bitrate determiner uses an initial bitrate predictor model trained on historical communication data to generate a probability distribution, allowing for dynamic selection of an optimal bitrate based on geographical location, network type, and previous communication statistics, minimizing the risk of bandwidth overshooting through a rate-distortion curve.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed predetermined bitrate is used for real-time communication, then the system is simple to implement, but the communication quality degrades when the bitrate does not match the available bandwidth
Solution Approach 1:
The patent implements dynamic bitrate selection by training a machine learning model on historical communication data to predict optimal initial bitrates based on various features. Instead of using a fixed predetermined bitrate, the system adapts the bitrate dynamically by selecting from multiple candidate bitrates based on the trained model's predictions, thereby resolving the contradiction between system simplicity and communication reliability.
Solution Approach 2:
The patent changes the bitrate parameter from a fixed value to a dynamically selected value from a set of candidate bitrates. The system uses a trained machine learning model to predict the optimal bitrate based on input features, and selects the candidate bitrate closest to the predicted value. This parameter change enables the system to adapt to varying network conditions while maintaining reasonable implementation complexity.
2Productivity
If a high bitrate is used to improve communication quality, then more bits are communicated per unit time, but delays and freezing occur when the bitrate exceeds the available bandwidth
Solution Approach 1:
The patent applies preliminary action by training a machine learning model on historical communication data before actual real-time communication occurs. The model learns from past experiences what bitrates work well under various conditions. When establishing a new communication session, the system uses this pre-trained model to predict and select an appropriate initial bitrate, preventing bandwidth overshooting and associated delays before they occur.
Solution Approach 2:
The system incorporates feedback by using historical communication data that includes actual communication outcomes. The machine learning model is trained on this feedback data, which contains information about what bitrates resulted in good versus poor communication quality. This feedback mechanism enables the system to learn from past performance and make better bitrate selections in the future, balancing quality and delay avoidance.
3Reliability
If a low bitrate is used to avoid bandwidth overshooting, then delays are minimized, but communication quality deteriorates
Solution Approach 1:
The patent changes the bitrate parameter from a conservative fixed low value to a dynamically optimized value. The system maintains a set of candidate bitrates ranging from lower to higher values, and uses a trained machine learning model to select the most appropriate bitrate based on current conditions. This allows the system to achieve communication stability when needed while capturing quality improvements when bandwidth is available.
Solution Approach 2:
The patent implements dynamic bitrate selection that adapts to varying network conditions. Rather than consistently using a low bitrate to ensure stability, the system dynamically selects from multiple candidate bitrates based on predictions from a trained machine learning model. This dynamic approach enables the system to achieve high communication quality when conditions permit while maintaining stability when bandwidth is constrained.
Data Source
AI summary
A method for determining an initial bitrate for a communication includes receiving a communication request to establish a digital communication between a first user device and a second user device associated with a plurality of features including a geographical identifier identifying a geographical location associated with the first user device, a first network type connection associated with the first user device, a second network type connection associated with the second user device, and an average bitrate for a previous digital communication of the first user device. The method includes determining, using an initial bitrate predictor model configured to receive the plurality of features as feature inputs, an initial bitrate for the digital communication between the first user device and the second user device, and establishing the digital communication between the first user device and the second user device at the determined initial bitrate.