Mobile Network Device Detection Using IMEI IMSI Hash Signatures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current device detection systems in mobile networks are inefficient due to the assumption that device and subscriber identities are unique, failing to account for multiple devices with cloned SIMs and frequent SIM swaps, leading to inefficient database indexing and difficulty in detecting new device combinations.
Innovation Solution
A method using a hash algorithm to generate a signature from device and subscriber identifiers, such as IMEI and IMSI, which is stored with a timestamp, allowing for efficient detection of new device combinations by archiving or deleting old signatures and using a simple key-value database structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional database paradigms storing device and subscriber identities as 2-tuples are used, then the system can maintain device-subscriber relationships, but the database becomes highly inefficient due to n:m relationships and high re-use of key values
Solution Approach 1:
The patent transforms the database key structure from storing raw device and subscriber identities (2-tuples) to storing cryptographic hashes of these identities. This parameter transformation changes the key characteristics from reusable, collision-prone values to unique, fixed-length hash values, resolving the efficiency problem while maintaining detection accuracy
Solution Approach 2:
The patent creates a cryptographic copy (hash) of the device and subscriber identity combination instead of storing the original identities directly. This hash copy serves as a unique identifier that preserves the ability to detect device changes while eliminating the inefficiencies of storing and querying actual identity values in n:m relationships
2Measurement precision
If the system stores and queries actual device and subscriber identities, then it can detect device changes, but it generates hundreds or thousands of requests per second requiring onerous hardware requirements
Solution Approach 1:
The patent applies cryptographic hashing to transform device and subscriber identities into fixed-length hash values. This parameter transformation reduces the computational complexity of storage and query operations, enabling the system to handle hundreds or thousands of requests per second with minimal hardware requirements while maintaining precise device change detection
Solution Approach 2:
The patent segments the device detection function into two parts: the cryptographic hashing operation that creates unique identifiers, and the simple database lookup operation that detects changes. This segmentation allows each component to be optimized independently, reducing overall system complexity and hardware requirements
3Adaptability or versatility
If the database stores MSISDN/IMSI and IMEI tuples for device detection, then it can track device-subscriber relationships, but it suffers from inefficient indexing due to cloned SIMs and IMEI re-use
Solution Approach 1:
The patent transforms the database indexing problem by applying cryptographic hashing to the device and subscriber identity combination. This transformation creates a unique hash value for each device-subscriber pairing, eliminating indexing inefficiencies caused by cloned SIMs and IMEI re-use while simplifying the database structure
Solution Approach 2:
The patent introduces cryptographic hashing as an intermediary layer between the raw device/subscriber identities and the database storage. This intermediary transformation ensures that even when original identities are reused or cloned, the hash values remain unique, simplifying database indexing and lookup operations
Data Source
Figure 1~4
AI summary
An application (1, 10, 20) in a mobile network, receives an EIR trigger (MAP CHECK_IMEI). The application generates a signature derived from a device identifier (IMEI) and a subscriber identifier (e.g. IMSI), the signature not having a semantic meaning as it is pseudo-randomly generated. The application (1, 10, 20) checks if the signature exists in a signature store (2, 11, 21); and if not, determines that the device/subscriber combination has not been detected in the network within a certain preceding time window, and stores the signature in the store, with an associated timestamp. It also transmits a notification to an external system such as a device management system. If the signature is already in the database, the application determines that the device/subscriber combination already exists in the network and updates a time stamp associated with the signature in the store to indicate the last time the signature was processed.