Deep Fusion Reasoning Engine for Wireless QoE Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Quality of Service (QoS) implementations in wireless networks do not effectively consider all content delivered features and user-provided information, making it challenging to optimize network traffic for a satisfactory Quality of Experience (QoE) for end-users, as they are not always aligned with user perceptions.
Innovation Solution
A deep fusion reasoning engine (DFRE) is introduced to provide dynamic and explainable QoE metrics by using multimodal data from networks and user perceptions, projecting raw measurements into conceptual spaces and applying symbolic reasoning to generate explainable conclusions, optimizing QoE while efficiently using networking resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If QoS implementations are used to optimize network traffic, then network throughput and latency are improved, but the alignment with user-perceived Quality of Experience deteriorates because QoS does not consider all content delivered features and user-provided information
Solution Approach 1:
The patent merges QoS metrics (network throughput, latency) with QoE metrics (user perception, content features) into a unified QoE optimization framework. The deep fusion reasoning engine combines multiple data sources including network measurements, content delivery features, and user feedback to create a comprehensive view that aligns technical performance with user experience.
Solution Approach 2:
The patent introduces a deep fusion reasoning engine as an intermediary layer between traditional QoS implementations and user experience outcomes. This engine processes and fuses multiple input sources (network metrics, content features, user feedback) to generate optimized traffic routing decisions that consider both technical performance and user perception.
2Measurement precision
If multiple network metrics and factors are monitored to improve QoE accuracy, then QoE assessment precision is improved, but the complexity of identifying and explaining QoE metrics to network administrators increases
Solution Approach 1:
The deep fusion reasoning engine serves as an intermediary that handles the complexity of processing multiple network metrics and factors. It fuses diverse inputs (network measurements, content features, user feedback) and translates them into actionable, easily interpretable QoE assessments and traffic optimization recommendations for network administrators.
Solution Approach 2:
The patent replaces manual analysis of complex network metrics with an automated deep fusion reasoning engine that uses machine learning and symbolic reasoning. This substitution automates the process of identifying, analyzing, and explaining QoE metrics, reducing the burden on network administrators while maintaining high assessment precision.
3Productivity
If traditional QoS metrics are used to optimize network traffic, then network resource utilization is improved, but the satisfaction of end-user experience requirements deteriorates because user perceptions are not always aligned with QoS metrics
Solution Approach 1:
The patent implements a feedback mechanism where user feedback and perceptions are continuously collected and fed into the deep fusion reasoning engine. This feedback loop allows the system to adjust traffic optimization decisions based on actual user experience outcomes, ensuring that network resource utilization strategies align with user satisfaction requirements.
Solution Approach 2:
The patent transforms the static QoS metric optimization into a dynamic system that adapts to changing user perceptions and experience requirements. The deep fusion reasoning engine continuously processes new data (network metrics, content features, user feedback) and dynamically adjusts traffic routing decisions to maintain alignment between resource utilization and user satisfaction.
Data Source
AI summary
In one embodiment, a network quality assessment service that monitors a network obtains multimodal data indicative of a plurality of measurements from the network and subjective perceptions of the network by users of the network. The network quality assessment service uses the obtained multimodal data as input to one or more neural network-based models. The network quality assessment service maps, using a conceptual space, outputs of the one or more neural network-based models to symbols. The network quality assessment service applies a symbolic reasoning engine to the symbols, to generate a conclusion regarding the monitored network. The network quality assessment service provides an indication of the conclusion to a user interface.