Buffer-Aware Streaming Quality Estimation From Packet Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technologies fail to accurately and real-time estimate streaming quality from packet data, leading to unsatisfactory network control for streaming reproduction.
Innovation Solution
A streaming quality estimation device that collects packet data, calculates packet statistical values, and uses a neural network-based estimation model to estimate streaming quality by inputting packet statistical values exceeding the buffer time, thereby eliminating deviations between packet arrival and data use times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If packet statistical values are used to estimate streaming quality, then network control for streaming reproduction is improved, but the estimation accuracy and real-time performance are insufficient
Solution Approach 1:
The patent applies preliminary action by collecting packet data and calculating packet statistical values in advance before they are needed for quality estimation. The system continuously accumulates packet information and pre-processes it into statistical metrics, enabling accurate real-time estimation when quality assessment is required.
Solution Approach 2:
The patent implements feedback by using the estimated streaming quality information to adjust network control parameters. The system continuously monitors packet statistics, estimates quality metrics, and feeds this information back to control mechanisms that can modify streaming parameters to maintain optimal quality.
2Productivity
If packet data is collected and processed for quality estimation, then network control capability is enhanced, but processing time and computational complexity increase
Solution Approach 1:
The patent applies segmentation by dividing the packet data processing into distinct segments: packet collection, packet statistical value calculation, and quality estimation. This modular approach allows each component to be optimized independently and enables parallel processing of different tasks, reducing overall processing time.
Solution Approach 2:
The patent applies partial action by selectively processing only the necessary packet statistics relevant to quality estimation rather than analyzing every packet detail. The system calculates only the essential statistical values needed for quality metrics, avoiding unnecessary computational overhead while maintaining estimation accuracy.
3Ease of manufacture
If conventional quality estimation methods are used, then implementation is simpler, but the estimation results do not reflect actual user experience accurately
Solution Approach 1:
The patent applies parameter changes by transforming raw packet data into derived statistical parameters that better reflect user experience. Instead of using simple packet counts or basic metrics, the system calculates comprehensive statistical values that capture temporal patterns, buffer dynamics, and quality variations, providing more accurate QoE estimation.
Data Source
AI summary
A non-transitory computer-readable recording medium stores therein a program that causes a computer to execute a process including collecting packets related to streaming data delivered from a transmission source to a transmission destination, calculating a packet statistical value of a predetermined time period based on the collected packets, and creating, based on the calculated packet statistical value, an estimation model for estimating streaming quality using, as an input, a time series of packet statistical values corresponding to a time length exceeding a buffer time of the streaming data at the transmission destination.