Dynamic MCU Selection for Video Conferencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wide Area Network (WAN) bandwidth utilization is inefficient due to unnecessary high video source stream routing caused by static video bridge IDs being permanently mapped to a specific Multipoint Control Unit (MCU), leading to increased latency, packet loss, and jitter in video conferencing.
Innovation Solution
Select an MCU for a video meeting based on historical data of past meetings that correlate in time, such as date, day of the week, or month, to minimize WAN traffic by predicting the region with the greatest number of participants, without considering the actual invitees, thus optimizing network resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a personal static bridge ID is permanently mapped to one MCU resource, then the meeting setup is simplified and consistent, but the WAN bandwidth utilization becomes inefficient and video latency increases
Solution Approach 1:
The patent implements dynamic MCU selection by replacing the static permanent mapping of bridge IDs to MCUs with a dynamic selection process. The system now selects MCUs based on real-time factors including participant locations, historical meeting data, and current network conditions. This dynamic approach allows the bridge ID to be flexibly assigned to different MCUs for different meetings, optimizing WAN bandwidth utilization while maintaining ease of meeting setup through automated selection.
Solution Approach 2:
The system changes the selection parameters for MCU assignment from a fixed static mapping to multiple dynamic parameters including participant geographic locations, historical meeting patterns, and current network conditions. By considering these multiple parameters, the system can select the optimal MCU for each specific meeting scenario, reducing WAN traffic while preserving the simplicity of bridge ID-based meeting initiation.
2Stability of the object's composition
If video source streams are routed through a distant MCU, then the meeting can be hosted consistently, but network latency and packet loss increase
Solution Approach 1:
The patent applies local quality by selecting MCUs based on their geographic proximity to participant locations. Instead of using a single centralized MCU for all meetings, the system chooses local MCUs that are geographically close to the majority of participants. This reduces the distance video streams must travel across the WAN, thereby reducing latency and packet loss while maintaining consistent meeting hosting through automated local selection.
Solution Approach 2:
The system performs preliminary action by analyzing participant locations and historical meeting data before selecting an MCU. This advance selection based on predicted participant locations allows the system to pre-position the meeting at an optimally located MCU, preventing latency issues before they occur rather than reacting to them during the meeting.
3Adaptability or versatility
If numerous personal static bridge IDs are deployed across multiple regions, then user accessibility is improved, but heavy traffic is generated across WANs connecting those regions
Solution Approach 1:
The patent applies segmentation by distributing meeting hosting across multiple regional MCUs instead of concentrating all traffic through a single centralized MCU. Each region can host meetings locally using its own MCU resources, segmenting the overall network traffic into smaller regional flows. This reduces the volume of traffic crossing WAN connections between regions while maintaining user accessibility through localized meeting points.
Solution Approach 2:
The system uses historical meeting data and participant location information as intermediaries to make intelligent MCU selection decisions. These intermediaries enable the system to predict which regional MCU will be most appropriate for a given meeting based on where participants are likely to join from, thereby routing traffic locally rather than across WANs and reducing overall network traffic volume.
Data Source
AI summary
To select an appropriate multipoint control unit (MCU) for hosting an upcoming video meeting, video meeting data of previous meetings is stored, including time and location data associated with meetings conducted on a bridge ID and attended by participants located in different regions. The schedule time of an upcoming video meeting using the bridge ID is compared with the time data of previous video meetings on the bridge ID to identify related video meetings among the previous video meetings, the related video meetings being time correlated with the scheduled time of the upcoming video meeting. One of a number of MCUs in the network is selected to be a host MCU to host the upcoming video meeting based on the location data of the related video meetings, in which the host MCU is selected as a function of its proximity to locations of participants of the related video meetings.