User Equipment Positioning With Inertial NLoS Bias Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The lack of Line of Sight (LoS) communications, or Non Line of Sight (NLoS), introduces significant bias in estimating the position of User Equipment (UE) in wireless communication networks, leading to inaccurate location estimation due to biased range and angle measurements, which is particularly challenging in cluttered environments with limited LoS paths and requires costly and scalable expert systems for data collection.
Innovation Solution
A method leveraging inertial measurements from UEs, combined with machine learning models trained on crowdsourced data, to correct NLoS errors by backtracking UE positions from LoS anchors, using inertial navigation systems and neural networks to predict NLoS biases, reducing the need for expensive expert systems and enabling continuous data collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If NLoS communication is used for positioning, then positioning coverage is improved, but positioning accuracy deteriorates due to biased range and angle measurements
Solution Approach 1:
The patent introduces an intermediary correction mechanism that uses inertial measurement units (IMUs) and machine learning models to estimate and compensate for NLoS bias. The system collects data from multiple UEs with IMUs, trains neural networks to predict bias patterns, and applies corrections to positioning measurements, thereby maintaining accuracy while using NLoS paths for coverage.
Solution Approach 2:
The system implements feedback by continuously collecting positioning data from UEs, comparing actual positions with predicted positions, and using the differences to retrain and refine the neural network models. This iterative feedback process improves the accuracy of bias correction over time.
2Measurement precision
If expert systems are used to collect positioning data, then data collection accuracy is improved, but system cost and complexity increase
Solution Approach 1:
The patent enables UEs to serve themselves by equipping them with inertial measurement units that automatically collect and transmit positioning data without requiring external expert systems. The UEs autonomously participate in the data collection process, eliminating the need for costly professional surveying equipment and personnel.
Solution Approach 2:
Instead of using expensive expert systems, the patent creates a distributed copy of positioning data collection across multiple UEs. Each UE with an IMU becomes a data collection node, replicating the function of expensive surveying equipment but at minimal cost, thereby achieving comprehensive data collection without high system complexity.
3Productivity
If multiple UEs with IMUs are deployed for data collection, then data scalability is improved, but device requirements and deployment complexity increase
Solution Approach 1:
The patent makes IMUs universal by integrating them into existing UE hardware platforms, allowing any UE with an IMU to participate in data collection. This multi-functional approach enables smartphones, tablets, and other mobile devices to serve as positioning data collection nodes, greatly improving scalability without requiring specialized expensive equipment.
Solution Approach 2:
The system leverages the existing self-service capabilities of UEs by using their built-in or attachable IMUs to automatically collect positioning data during normal operations. This eliminates the need for complex deployment infrastructure and expert system coordination, allowing spontaneous participation from any UE in the network.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively mitigates NLoS errors by leveraging existing infrastructure for continuous data collection, providing accurate UE positioning with reduced costs and scalability issues, while maintaining high generalization capabilities through neural networks.
Implementation Method 1
obtaining, using one or more inertial sensors of the UE, first sensor location information indicative of movement of the UE between the first time and a second time
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Corrected User Equipment (UE) positioning is based at least in part on signaling that is associated with UE inertial measurements. Such measurements are used in some embodiments to track UE locations along trajectories of UE movement and develop or obtain a model to subsequently predict UE positioning corrections based on channel estimates of a wireless channel. For example, reference signaling may be transmitted by a UE, and the UE then receives positioning corrections or corrected positioning in response to the reference signaling.