Cellular Network User Mobility Model Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods lack an effective approach to create a user mobility model for a large number of users in cellular networks, such as GSM networks, which is essential for optimal resource management and deployment.
Innovation Solution
A method and device that generate user movement tracks by deriving known and unknown path points, connecting them in a temporal sequence, and using these tracks to create a user mobility model that balances network loads across location areas, thereby improving resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a user mobility model is created to manage huge numbers of users, then network resource utilization can be improved, but no appropriate method has been provided in the prior art to create such a model
Solution Approach 1:
The patent segments the user movement trajectory into multiple segments: known path points (where user events are recorded) and unknown path points (inferred between known points). This segmentation allows the system to handle huge numbers of users by processing only the known points while inferring the unknown portions, thereby creating a manageable mobility model without requiring complete tracking of all user movements.
Solution Approach 2:
The patent introduces an intermediary approach by using unknown path points as placeholders and inference targets. Instead of directly tracking every user movement detail, the system uses the temporal sequence of known path points as intermediaries to infer the unknown movements, thus enabling mobility model creation without complete user behavior data.
2Measurement precision
If known path points are derived and connected to form user movement tracks, then user mobility can be modeled, but unknown path points must be filled in between adjacent known path points
Solution Approach 1:
The patent performs preliminary action by first deriving all known path points and their temporal sequences before filling in the unknown portions. The known path points and their connections are established in advance, and only then are the unknown path points inferred and inserted between adjacent known points, simplifying the overall process.
Solution Approach 2:
The patent changes the parameter approach by transitioning from complete trajectory tracking to a two-state representation (known/unknown path points). This parameter change allows the system to maintain measurement precision for observed movements while managing complexity by inferring unobserved portions based on temporal relationships and spatial constraints.
3Productivity
If cells are re-divided into location areas based on user mobility model, then network loads can be balanced, but location update flows are still triggered when users move across location areas
Solution Approach 1:
The patent applies local quality by creating location areas with non-uniform boundaries based on actual user mobility patterns. Instead of uniform cell division, the location area boundaries are optimized locally according to user movement characteristics, resulting in areas that balance network loads while minimizing boundary-crossing movements that trigger location updates.
Data Source
AI summary
A method for generating a user mobility model includes: firstly deriving user movement tracks of all the users in a cellular network and then generating from the user movement tracks a user mobility model involving a user traffic between home cells of any two of path points in the cellular network. Thus, a cell can be re-divided into respective location areas according to the generated user mobility model to thereby deploy reasonably various network resources in the GSM network according to the re-divided positional zones. The invention further discloses a method and device for generating a user movement track and a device for generating a user mobility model.