Link Performance Prediction for Wireless Network Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Emerging wireless network technologies, such as 5G, face challenges in maintaining consistent performance due to dynamic link quality conditions, especially with mobile user equipment, which existing solutions like DASH and HLS may not adequately address for new services requiring real-time and mission-critical applications.
Innovation Solution
The implementation of a Link Performance Prediction (LPP) technology using machine learning techniques to predict future network behaviors like bandwidth, latency, and coverage holes, allowing applications and infrastructure to make operational decisions to optimize performance and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If reactive solutions like DASH and HLS are used to address network condition variations, then quality of service is sufficient for traditional traffic types, but new traffic types and services requiring real-time performance cannot be adequately supported
Solution Approach 1:
The patent implements link performance prediction technology that proactively forecasts future network conditions before they actually occur. By predicting bandwidth, latency, and link quality in advance, the system can pre-adjust transmission parameters and buffer content, enabling real-time applications to maintain consistent performance even when network conditions deteriorate. This preliminary action transforms reactive QoS adjustment into proactive performance management.
2Reliability
If additional infrastructure is built to address network condition variations, then quality of service improves for traditional traffic, but deployment costs and complexity increase
Solution Approach 1:
The patent replaces physical infrastructure expansion with intelligent software-based link performance prediction and adaptive transmission control. Instead of deploying additional base stations or network elements to handle quality variations, the system uses machine learning models to predict link behavior and dynamically adjusts transmission parameters (bitrate, buffer size, segment duration). This substitution of mechanical infrastructure with intelligent control algorithms reduces deployment complexity while maintaining QoS consistency.
3Productivity
If reactive solutions are used to handle dynamic link quality, then existing infrastructure can be maintained, but new services requiring real-time performance and predictability cannot be supported
Solution Approach 1:
The patent implements a closed-loop feedback system where link performance predictions are continuously generated based on historical and real-time network data, then used to adjust transmission parameters, and the results feed back into refining the prediction models. This feedback mechanism transforms the system from purely reactive to predictively adaptive, maintaining high resource utilization while providing the predictability required for new real-time services through continuous learning and adjustment.
Data Source
AI summary
Various systems and methods for determining and communicating Link Performance Predictions (LPPs), such as in connection with management of radio communication links, are discussed herein. The LPPs are predictions of future network behaviors/metrics (e.g., bandwidth, latency, capacity, coverage holes, etc.). The LPPs are communicated to applications and/or network infrastructure, which allows the applications/infrastructure to make operational decisions for improved signaling/link resource utilization. In embodiments, the link performance analysis is divided into multiple layers that determine their own link performance metrics, which are then fused together to make an LPP. Each layer runs different algorithms, and provides respective results to an LPP layer/engine that fuses the results together to obtain the LPP. Other embodiments are described and/or claimed.