Multipath CSI Location Mapping Without Anchor Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing location estimation techniques in wireless communications systems, such as triangulation and fingerprinting, fail to leverage multipath components effectively, leading to inaccurate positioning and high computational overhead, especially in environments with insufficient anchors or varying radio propagation characteristics.
Innovation Solution
A machine learning model trained on channel state information (CSI) measurements to predict device and virtual anchor locations using multipath timing and angle of arrival information, enabling accurate location estimation without anchor coordination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional triangulation or fingerprinting techniques are used for location estimation, then location prediction can be performed, but the techniques fail to leverage multipath components effectively leading to inaccurate positioning and high computational overhead
Solution Approach 1:
The patent converts multipath components, which were traditionally considered harmful interference degrading location estimation accuracy, into beneficial information sources. By treating multipath signals as virtual anchors, the system transforms signal reflections into useful positioning data, thereby improving location prediction accuracy while reducing computational overhead compared to traditional methods
Solution Approach 2:
The patent changes the fundamental parameter interpretation of multipath components from noise to useful signal. By extracting timing and angle of arrival information from multipath components and treating them as virtual anchor positions, the system transforms the parameter space to enable accurate location estimation without requiring additional physical anchors or complex coordination
2Ease of operation
If fingerprinting based on RSSI or CSI measurements is used, then location prediction can be performed, but timing coordination constraints are imposed on anchors in the network
Solution Approach 1:
The patent enables the system to self-determine anchor positions and device location using only signal measurements, without requiring external coordination or pre-configured anchor position data. The multipath components themselves provide the reference information needed, making the system independent of network-wide timing coordination while maintaining high prediction accuracy
3Adaptability or versatility
If triangulation or trilateration is used with insufficient anchors, then location estimation can be performed, but the techniques are specific to a given spatial environment and require anchor coordination
Solution Approach 1:
The patent creates virtual copies of anchor positions from multipath components. Each multipath reflection is treated as a virtual anchor, effectively multiplying the available reference points without adding physical anchors. This copying mechanism enables accurate location estimation in environments with insufficient physical anchors while removing environment-specific constraints through the machine learning model's generalization capability
Data Source
AI summary
Certain aspects of the present disclosure provide methods, apparatus, and systems for predicting a location of a device in a spatial environment using a machine learning model. An example method generally includes measuring a plurality of signals received from a network entity at a device. A channel state information (CSI) measurement is generated from the measured plurality of signals. Generally, the CSI measurement includes a multipath component. Positions of one or more anchors in a spatial environment are identified based on a machine learning model trained to identify the positions of the one or more anchors based on the CSI measurement. A location of the device is estimated based on the identified positions of the one or more anchors.