Client-Side Network Probe Coordination for Real-Time Load Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining and managing network loading and capacity in wireless networks are non-real-time, costly, and dependent on operator and vendor variations, making it difficult to collect and analyze channel and loading conditions in a secure and scalable manner for content providers and subscribers.
Innovation Solution
A data communication system that uses client devices to send probe packets to a control server, optimizing the timing and sizing of probes to determine system loading, with a control server coordinating client devices to gather network measurement information, including delay, packet loss, and throughput, and using pre-trained neural network algorithms to estimate future network load and capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If network side base stations collect and report channel and load information, then network state information is available, but the system becomes costly, non-real-time, and dependent on operator and vendor variations
Solution Approach 1:
Instead of having the network collect and report measurements from base stations, the patent inverts the approach by having client devices perform measurements and report back to the network. This inversion eliminates the need for complex network-side measurement infrastructure while achieving real-time, accurate network state information collection across multiple operators and vendors
Solution Approach 2:
The patent enables client devices to autonomously perform network measurements and self-report to the control server without requiring network-side intervention or complex configuration. This self-service approach reduces system complexity and cost while maintaining measurement accuracy and real-time capability
2Measurement precision
If network side base stations collect channel and load information, then network state is known, but real-time availability and scalability are limited
Solution Approach 1:
The patent inverts the traditional measurement architecture by placing measurement functionality in client devices rather than network base stations. This allows measurements to be performed continuously in the background by clients without disrupting network operations, enabling real-time availability and scalability across multiple operators
Solution Approach 2:
The patent enables continuous measurement and reporting by client devices that operate independently of network scheduling. Clients continuously perform measurements and report to the control server, ensuring real-time availability of network state information without interruption or dependency on network-side actions
3Productivity
If multiple client devices send probe packets to determine system loading, then real-time and accurate inference is achieved, but device complexity and coordination requirements increase
Solution Approach 1:
The patent segments the measurement and inference functionality by having each client device independently perform local measurements and report to the control server. This segmentation reduces the complexity at individual client devices while enabling accurate inference through aggregation of results from multiple devices
Solution Approach 2:
The patent implements a feedback mechanism where control servers receive measurement reports from multiple client devices and use this information to infer network loading and capacity. This feedback loop enables real-time, accurate inference while the control server manages the coordination complexity centrally
Data Source
AI summary
A data communication system configured to determine and coordinate loading in a mobile network with plurality of client devices. The system includes a control server coupled the client devices. The control server has a probe control function that transmits a downlink (DL) probe with an updated client probing profile to each client device. The client device has a client probing control engine that gathers network measurement information for transmission to the control server and generates uplink (UL) probes containing the network measurement information including, radio access network (RAN) metrics, DL probe measurements for transmission to the control server based on the client probing profile. The control server receives the UL probes from the client devices and determines the system load based on the measurement information, the control server updates client scheduling information based on the system load and transmits the client scheduling information to the plurality of client devices.


