Location Filtering via Device Mobility Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The time-varying nature of RF signals leads to inaccuracies in tracking the location of both stationary and mobile devices, despite the use of filtering techniques, resulting in significant errors in determining the actual path of mobile devices.
Innovation Solution
A method that classifies devices as stationary or mobile based on location information, adjusting filter parameters such as measurement noise and process noise covariance for Kalman filters to improve location estimation accuracy, by determining whether a device is likely to be stationary or mobile, thereby selecting appropriate filter settings to enhance tracking precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filtering is applied to location data to reduce measurement noise, then location estimation accuracy improves for stationary devices, but tracking precision deteriorates for mobile devices due to time-varying RF signal characteristics
Solution Approach 1:
The system dynamically adjusts filter parameters based on device mobility detection. When a device is detected as mobile, the system switches from using stationary device filters to mobile device filters, allowing the filtering characteristics to adapt to the time-varying nature of RF signals from moving devices. This dynamic adaptation resolves the contradiction by enabling optimal filtering for each device state rather than using a fixed filtering approach.
Solution Approach 2:
The invention changes filter parameters (such as measurement noise covariance and process noise covariance) based on the detected mobility state of the device. By adjusting these parameters according to whether the device is stationary or mobile, the system optimizes location estimation accuracy for each state, thereby resolving the contradiction between filtering effectiveness for stationary devices and tracking precision for mobile devices.
2Device complexity
If a single filter configuration is used for all devices, then system complexity is reduced, but location tracking accuracy deteriorates due to inability to adapt to different mobility states
Solution Approach 1:
The system segments the device population into stationary and mobile categories based on mobility detection. By dividing devices into these distinct groups, the system can apply different filter configurations to each segment, improving location tracking accuracy for both stationary and mobile devices while maintaining manageable system complexity through clear classification categories.
Solution Approach 2:
The system dynamically selects appropriate filter configurations based on real-time mobility detection. Rather than maintaining multiple static filter configurations simultaneously, the system uses mobility detection to dynamically switch between filter types, reducing system complexity while adapting to different device states to maintain high location tracking accuracy.
3Measurement precision
If filtering parameters are optimized for stationary devices, then location estimation accuracy improves, but tracking errors increase for mobile devices due to inappropriate filter settings
Solution Approach 1:
The system changes filter parameters (measurement noise covariance, process noise covariance) based on the detected mobility state. For mobile devices, the system adjusts these parameters to account for the time-varying nature of RF signals, thereby reducing tracking errors. This parameter adaptation allows the system to maintain high location estimation accuracy across both stationary and mobile devices without the tracking errors that would result from using stationary-optimized filters for mobile devices.
Data Source
AI summary
In one embodiment, a method includes obtaining location information associated with a remote device, and processing the location information. Processing the location information includes determining if the remote device is mobile. The method also includes configuring a filter such that at least one parameter indicates that the remote device is mobile if the remote device is mobile, and configuring the filter such that the at least one parameter indicates that the remote device is approximately stationary if the remote device is not mobile. The filter is applied to the location information to generate a filtered location estimate which is arranged to estimate a location of the remote device.