How Networks Keep Video Streams Smooth During Busy Times

A method for internet providers to automatically balance network resources so that multiple users streaming video simultaneously experience the best possible overall quality.

What it covers

This patent describes a system for managing network traffic when many people are streaming video at once. It monitors each individual video stream by looking at data like buffer levels—how much video is pre-loaded—and instances of freezing or skipping. It then calculates a quality score for every user and uses an objective function to decide how to distribute limited network bandwidth. By adjusting scheduling parameters for each stream, the network aims to maximize the total quality across all users, preventing one person's high-definition stream from causing everyone else's video to buffer.

What it doesn't cover

• —Does not cover methods that rely solely on static bandwidth allocation without considering real-time video quality metrics.
• —Does not cover client-side video compression algorithms or codecs like H.264 or HEVC.
• —Does not cover hardware-specific network routing protocols that do not implement a degradation control algorithm.
• —Does not cover systems that prioritize traffic based only on user subscription tier rather than actual video quality metrics.

The clever bit

Instead of just giving everyone equal bandwidth, the system uses an objective function to mathematically maximize the sum of all users' quality scores, allowing the network to 'sacrifice' a tiny bit of quality on one stream to prevent a total freeze on another.

Why it matters

As streaming services like Netflix and YouTube became the primary source of internet traffic, network congestion became a major issue. This patent provides a mathematical framework for ISPs and content delivery networks to manage 'Quality of Experience' (QoE) dynamically. It helps prevent the 'tragedy of the commons' where individual high-demand streams degrade the experience for the entire network neighborhood.

Real-world examples

• 1.Adaptive bitrate streaming management in ISP core networks
• 2.Content Delivery Network (CDN) traffic shaping
• 3.Mobile network video optimization for 4G/5G base stations