Indoor Positioning Model Splitting Global and Local Signal Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building models and interpolating/extrapolating measurement data in indoor scenarios with vertical dimensions is challenging due to differences in structural properties between horizontal and vertical directions, and complex signal propagation with high shadowing effects, requiring specialized techniques for reliable modeling.
Innovation Solution
A method that splits the model into a first location-specific function describing global trends and a second location-specific function describing local variations, allowing for two-stage determination of model parameters, which includes considering shadowing effects and measurement covariance, and does not require heuristic parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If measurement data is collected extensively through crowd-sourcing and measurement campaigns, then the coverage and completeness of the database is improved, but the time and resources required for data collection and continuous updating increase significantly
Solution Approach 1:
The patent performs preliminary action by collecting and storing measurement data in advance during a training phase, creating a database that can be reused for multiple positioning operations. The system pre-processes and stores signal strength measurements along with location information, so that when positioning is needed, the data is already available without requiring new measurement campaigns.
Solution Approach 2:
The patent creates a model that copies the characteristics of the physical environment based on collected measurement data. Instead of continuously collecting real measurement data, the system creates a virtual representation (model) of the signal propagation characteristics that can be reused for positioning multiple users without additional field measurements.
2Reliability
If measurement data is collected from restricted access areas, then the completeness of the database is improved, but the ability to collect data in those areas is prevented
Solution Approach 1:
The patent uses an intermediary approach by employing model-based estimation to bridge the gap between areas with and without measurement data. The system creates a mathematical model that can estimate signal characteristics in restricted areas based on measurements from accessible areas, allowing the database to be complete without physically accessing restricted zones.
Solution Approach 2:
The system creates a virtual model that replicates the signal propagation characteristics of restricted areas based on data from accessible areas. This model copy allows the database to represent the entire coverage area including restricted zones without requiring actual measurements from those inaccessible locations.
3Adaptability or versatility
If interpolation and extrapolation methods are used to estimate data in areas without measurements, then the positioning capability outside coverage areas is improved, but the accuracy of position estimates may be reduced
Solution Approach 1:
The patent applies parameter changes by using different statistical parameters and model approaches for different regions. The system adjusts the model parameters and estimation methods based on the available measurement data and the specific characteristics of the area being estimated, optimizing the balance between coverage and accuracy for different locations.
Solution Approach 2:
The system applies local quality by using different estimation strategies for different areas. In areas with dense measurement data, the system uses direct model fitting, while in areas with sparse data, it employs different interpolation or extrapolation methods. Each region receives the most appropriate processing method for its specific data characteristics.
4Device complexity
If a single global model is used for positioning, then the device complexity is reduced, but the accuracy in complex indoor environments with vertical dimensions is insufficient
Solution Approach 1:
The patent segments the positioning model by separating the signal strength prediction into distinct components: a path loss model for general signal attenuation and a shadowing model for local variations. This segmentation allows the system to handle complex indoor environments with vertical dimensions by treating different propagation effects separately while maintaining manageable model complexity.
Solution Approach 2:
The system uses parameter changes by introducing additional parameters specific to indoor environments, such as floor loss parameters and shadowing standard deviations. These parameters allow the model to adapt to vertical dimensions and complex indoor structures without requiring a completely different modeling approach, thus maintaining reasonable complexity while improving accuracy.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method performed by at least one apparatus is inter alia disclosed, said method comprising: obtaining measurement data on a location-specific quantity of a signal transmitted by a transmitter; obtaining location information associated with said measurement data on said location-specific quantity; and determining, based on said obtained measurement data and said obtained location information, one or more model parameters of a model describing said location-specific quantity in dependence of location, wherein said model assumes a location dependence of said location- specific quantity in form of a combination of a first location-specific function and a second location-specific function.