Cooperative Target Tracking Using Mobile Sensor RSS Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing target tracking systems face challenges in accurately localizing mobile targets in complex indoor environments due to non-line-of-sight measurements and the need for costly site surveys, and they often fail to consider the mobility and cooperation of targets and sensors.
Innovation Solution
The system employs mobile sensor devices that use received signal strength matrices to cooperatively track targets by rebroadcasting beacons and leveraging particle filters, such as the Rao-Blackwellized particle filter, to refine sensor locations and generate RSS matrices without explicit site surveys, thereby improving tracking accuracy and coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional target tracking systems are used in complex indoor environments, then they require costly site surveys and explicit signal propagation models, but they fail to achieve accurate localization due to non-line-of-sight measurements and environmental complexities
Solution Approach 1:
The system enables sensors to automatically learn signal propagation characteristics by observing beacon transmissions and receptions among themselves, eliminating the need for manual site surveys. Each sensor contributes to building the RSS matrix through its own measurements and observations, making the system self-configuring and adaptive to the specific environment.
Solution Approach 2:
The system performs preliminary learning of signal propagation characteristics during an initial phase where sensors collect RSS measurements from beacon transmissions. This pre-learning process builds the RSS matrix before actual target tracking begins, allowing the system to account for environmental factors like walls and obstacles in advance, improving subsequent localization accuracy.
2Measurement precision
If mobile sensor devices are deployed for target tracking, then tracking coverage and accuracy improve, but the system must handle dynamic sensor positions and mobility-induced measurement variations
Solution Approach 1:
The system is designed to handle dynamic sensor positions by continuously updating the RSS matrix as sensors move. The particle filter algorithm adapts to changing sensor locations by incorporating motion models that predict sensor displacement, allowing the system to maintain tracking accuracy despite mobility-induced variations in measurement geometry and signal propagation paths.
Solution Approach 2:
The system uses feedback from actual RSS measurements to continuously refine the learned signal propagation model. As mobile sensors move and collect new measurements, these observations feed back into updating the RSS matrix, allowing the system to adapt to both environmental characteristics and dynamic positioning, improving tracking accuracy over time.
3Area of stationary object
If cooperative tracking with beacon rebroadcasting is implemented, then tracking coverage extends to areas with fewer direct sensor observations, but the system complexity increases due to multi-hop signal propagation
Solution Approach 1:
The system uses intermediate sensors as mediators to relay beacon information across the network. When a sensor cannot directly observe a target, other sensors that do observe the target can rebroadcast beacon information, and the RSS matrix learning mechanism accounts for multi-hop propagation paths. These intermediate nodes enable tracking coverage extension without requiring complex manual configuration of signal propagation models.
Solution Approach 2:
The system automatically learns multi-hop signal propagation characteristics through observations of beacon rebroadcasting among sensors. Rather than requiring explicit modeling of complex multi-path propagation, the RSS matrix is built empirically from actual measurements, allowing the system to self-adapt to the effective signal propagation in the specific environment including wall penetrations and reflections.
4Measurement precision
If particle filters are used to refine sensor locations and generate RSS matrices, then tracking accuracy improves significantly, but computational requirements and processing time increase
Solution Approach 1:
The system uses particle filters with a limited number of particles (e.g., 100 particles) rather than exhaustive sampling, providing sufficient tracking accuracy for practical applications while constraining computational requirements. This partial action approach achieves the necessary precision for mobile target tracking without the excessive computational burden of more sophisticated filtering methods, balancing accuracy with energy consumption in mobile sensor systems.
Data Source
AI summary
An architecture is provided for cooperative target tracking and signal propagation learning using mobile sensors. A method can comprise as a function of sensing data representative of a location of a target device at a first defined moment and model data relating to a motion model representing a probability density function, determining, by a system comprising a processor, a group of locations for the target device at a second defined time point, wherein the probability density function facilitates determining, based on the location of the target device at the first defined moment, a current location of the target device at a third defined moment; and as a function of the group of locations, generating, by the system, a data structure representing a matrix of received signal strength values; and identifying, by the system, a location of the group of locations for the target device at the third defined moment based on the data structure.