Central Cloud Intelligence for IoT eUICC Subscription Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Business customers face challenges in accurately loading subscriptions onto fleets of cellular IoT devices, leading to potential service outages and increased costs due to incorrect subscription downloads.
Innovation Solution
A method is implemented where a central cloud intelligence collects teaching data from deployed IoT devices and in-lab devices, associates this data with candidate subscriptions, and determines the most adapted subscription based on the device's radio environment, triggering its download.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If exhaustive information collection from operators about roaming partners and RAT is performed, then subscription selection accuracy is improved, but system complexity and maintenance burden increase
Solution Approach 1:
The system automatically collects teaching data from deployed IoT devices about their network registration experiences and PLMN selection outcomes. This self-service approach eliminates the need for manual information collection from operators, as the devices themselves provide the data needed to train the cloud intelligence model, reducing maintenance complexity while improving selection accuracy
Solution Approach 2:
The system implements a feedback loop where deployment results (successful or failed network registrations) are collected from field devices and used to retrain the central cloud intelligence model. This continuous feedback mechanism allows the system to learn from actual deployment conditions and improve its subscription selection accuracy over time without requiring manual updates from operators
2Ease of manufacture
If static database based on operator contacts and contracts is used, then initial setup is simplified, but adaptability to changing network conditions deteriorates
Solution Approach 1:
The system transitions from a static database to a dynamic learning model that continuously adapts to changing network conditions. The central cloud intelligence is retrained periodically with new teaching data from deployed devices, allowing it to adapt to network changes, new RATs, and evolving operator policies automatically, maintaining high adaptability without complicating initial setup
3Productivity
If subscription download is performed without comprehensive teaching data collection, then deployment speed is improved, but subscription correctness deteriorates
Solution Approach 1:
The system collects teaching data from a representative sample of deployed devices before performing subscription downloads for the entire fleet. This preliminary action allows the central cloud intelligence to be trained on real-world deployment conditions, ensuring subscription correctness is improved before mass deployment, while still maintaining high deployment speed through efficient batch processing
Data Source
AI summary
The present invention relates to a method to select appropriate subscription to be loaded in the eUICC of a device from an IoT fleet among candidate subscriptions, said method comprising, for a central cloud intelligence, the steps of: collecting teaching data related to MNO subscriptions and their associated favorite and forbidden PLMNs from on-field deployed devices and/or from in-lab devices hosting sample, associating, in a database, those teaching data to candidate subscriptions, said method further comprising the steps of: receiving information relative to the radio environment of a device in need of a subscription, processing received information relative to radio environment in regard to teaching data and candidate subscriptions available in the database, determining among the candidate subscriptions an adapted subscription depending on this processing, triggering the download of the determined subscription to the device in need of a subscription.