Predictive Streaming Adaptation for Mobile Private Network Disruptions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless networks, including LTE and 5G, experience unpredictable disruptions due to interference patterns, leading to degraded stream quality for moving devices, especially in high-demand events, with current solutions failing to predict and mitigate these issues effectively.
Innovation Solution
A system and method using machine learning algorithms to analyze network, device, and environmental data to predict network performance metrics for moving devices, allowing for proactive switching to maintain high-quality streaming by predicting potential disruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If broadcasters rely on public LTE, 4G or 5G networks for streaming, then capacity, bandwidth, speed, security, reliability and quality of service are provided, but unexpected disruptions occur due to interference patterns and network strain
Solution Approach 1:
The system performs preliminary actions by predicting future network performance metrics before disruptions occur. The machine learning model analyzes current network conditions, device information, and environmental data to forecast future stream quality and bandwidth availability, enabling broadcasters to proactively switch to alternative devices or locations before disruptions affect the stream.
Solution Approach 2:
The system implements continuous feedback by monitoring current network performance, comparing it against predicted metrics, and using this information to make real-time switching decisions. The machine learning model continuously updates its predictions based on incoming data streams, allowing the system to adapt to changing network conditions and maintain stream quality.
2Duration of action of moving object
If the system switches to another streaming device after disruption occurs, then service continuity is maintained, but stream quality is degraded with artifacts and interruptions
Solution Approach 1:
The system switches to alternative streaming devices before disruptions occur by predicting future network performance. The machine learning model forecasts which devices will experience poor network conditions and proactively switches to backup devices that are predicted to maintain high stream quality, preventing artifacts and interruptions in the broadcast.
3Reliability
If machine learning algorithms are used to predict network performance, then disruptions can be anticipated and mitigated, but system complexity increases
Solution Approach 1:
The machine learning model serves multiple functions: it predicts future device locations, estimates network performance metrics, identifies optimal switching opportunities, and continuously learns from new data. This multi-functionality reduces the need for separate prediction systems and integrates disruption mitigation into a single unified platform.
Solution Approach 2:
The machine learning model continuously trains itself using incoming network data, device information, and environmental data. The system automatically updates its predictions and adapts to changing conditions without requiring manual reconfiguration, reducing operational complexity while maintaining high prediction accuracy.
4Reliability
If the system monitors and predicts network metrics in real-time, then proactive switching is enabled, but data processing requirements and computational load increase
Solution Approach 1:
The system focuses computational resources on predicting only the most critical parameters needed for switching decisions, such as future device location and network performance metrics. Rather than analyzing all possible network parameters, the machine learning model concentrates on the subset of data that most directly impacts stream quality and switching timing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems (200) and methods (400) are described for predicting network performance of a private network (130) serving a device moving through the private network. Systems include a data interface (210) configured to receive and parse information, a location calculator (220) configured to determine and transmit calculated location information based on received information, and a prediction component (230) configured to determine one or more network performance metrics from data streams received from the data interface and the location calculator. Methods (400) include receiving and parsing core network information, device information in the private network, and environmental data. The prediction component (230) determines a predicted network performance metric related to the device moving through the private network (160) based, at least in part, on the information received from the data interface (210) and location calculator (220).