Machine-Learning VoIP Call Routing by Recipient Location
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack the ability to dynamically route incoming Voice over Internet Protocol (VoIP) calls to a single connected device within a building based on the location of the intended recipient, leading to intrusive broadcasting and interference with ongoing operations.
Innovation Solution
A smart call routing system uses machine learning to identify and locate the intended recipient within a building by analyzing information from connected devices, such as cameras and smart speakers, and routes the call to the nearest device capable of handling the call.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If incoming calls are broadcasted to all connected devices in a home, then the likelihood of reaching the intended recipient is increased, but the intrusiveness and interference with ongoing operations increases
Solution Approach 1:
The system segments the call routing decision by dividing the building into multiple zones and determining which specific zone the recipient is in, rather than broadcasting to all devices uniformly. This allows targeted routing to devices in the recipient's specific location while leaving other devices unaffected.
Solution Approach 2:
The system applies local quality by making the call routing behavior dependent on the specific location of the recipient within the building. Different devices in different zones receive calls selectively based on where the recipient is currently located, rather than applying a uniform broadcasting approach across all devices.
2Object-affected harmful factors
If calls are routed to a single targeted device, then intrusiveness is reduced, but the ability to reach the recipient reliably decreases without accurate location information
Solution Approach 1:
The system performs preliminary action by continuously tracking and determining the recipient's location within the building before the actual call arrives. This advance knowledge of location enables the system to route incoming calls to the correct device immediately, ensuring both reliability and reduced intrusiveness.
Solution Approach 2:
The system uses feedback from location tracking data to dynamically adjust call routing decisions. By continuously monitoring which zone the recipient is in and using this feedback information, the system can reliably route calls to the appropriate device while avoiding unnecessary broadcasting to other devices.
3Measurement precision
If the system continuously tracks recipient location across multiple devices, then call routing accuracy is improved, but the system complexity and computational resources increase
Solution Approach 1:
The system reduces complexity by segmenting the tracking problem into discrete zones within the building rather than continuously tracking position throughout the entire space. Each device reports when it detects the recipient in its associated zone, creating a simplified zone-based location system that maintains precision while reducing computational burden.
Data Source
AI summary
A system and method for routing incoming calls to one of many connected devices on a premises (including within a building) based on determining the location of an intended recipient of the call uses machine learning. Images and/or audio from connected devices can be analyzed with neural networks and/or natural language processing to identify an intended recipient in a particular area of the premises. In addition, the system and method include monitoring device activity to use as a proxy for the location of the intended recipient. Also, recent device activity can be used to determine where an intended recipient may be on the premises.