Anchor-Level Signal Classification for Accurate Radio Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing localization systems based on radio technologies face challenges in achieving high positional accuracy due to signal attenuation, reflections, and Non-Line-of-Sight conditions, leading to inaccuracies and increased data traffic in central positioning engines.
Innovation Solution
Implementing decentralized signal classifiers at each anchor in the radio network to filter out low-quality radio signals based on machine learning, ensuring only high-quality signals are forwarded to the central positioning engine for position estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all radio signals are forwarded to the central positioning engine for processing, then the system can process all available signals, but the data traffic increases and positional accuracy decreases due to processing low-quality signals
Solution Approach 1:
The signal classifier performs preliminary classification of radio signals at the anchor before forwarding to the positioning engine. By pre-filtering signals based on their quality (direct path vs. multipath), the system reduces the volume of data transmitted while ensuring only high-quality signals are processed, thereby improving positional accuracy and reducing data traffic.
Solution Approach 2:
The invention extracts and separates low-quality signals (multipath signals) from the data stream at the anchor level. By identifying and excluding these low-quality signals before they reach the central positioning engine, the system reduces data traffic while maintaining high positional accuracy through processing only reliable direct path signals.
2Measurement precision
If low-quality radio signals are processed by the central positioning engine, then all signals are utilized, but positional accuracy decreases due to multipath effects and signal attenuation
Solution Approach 1:
The signal classifier performs preliminary identification and classification of signal quality at the anchor before forwarding to the positioning engine. By pre-filtering out low-quality multipath signals and only forwarding high-quality direct path signals, the system simplifies the processing burden on the central positioning engine while maintaining high positional accuracy.
Solution Approach 2:
The invention extracts and removes low-quality signals from the processing pipeline at the anchor level. By separating multipath signals from direct path signals and excluding the former from further processing, the system reduces the complexity of signal processing required at the central engine while improving positional accuracy.
3Productivity
If signal classification is performed at the central positioning engine, then centralized processing is simplified, but data traffic increases and processing time is consumed
Solution Approach 1:
The invention segments the signal classification function from the central positioning engine and distributes it to individual anchors. Each anchor performs local signal classification to determine signal quality, which reduces the data traffic to the central engine and improves processing efficiency by performing classification at the source rather than centrally.
Solution Approach 2:
The anchors perform self-service by classifying their own received signals locally before forwarding to the positioning engine. This self-classification capability at the anchor level reduces the burden on the central processing engine, decreases data traffic, and improves overall processing efficiency.
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 reduces data traffic and enhances positional accuracy by filtering out low-quality signals, thereby improving the performance of the central positioning engine.
Implementation Method 1
the signal classifier is configured to determine a signal quality class out of a plurality of signal quality classes for the received radio signal based on signal amplitude information of the received radio signal
Implementation Method 2
a position estimator in a positioning server is provided for estimating the position of the object based on propagation delays of radio signals received by different anchors in response to a transmission event of a radio tag on the object
Implementation Method 3
radio signals can experience attenuation, reflections and scattering, resulting in multipath effects
Implementation Method 4
when radio signals bounce off walls, floors and objects, those signals can interfere with the direct Line-of-Sight (LOS) path
Implementation Method 5
radio signals can experience attenuation, reflections and scattering, resulting in multipath effects
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a computer-implemented method for generating a predictor for a position (pos) of an object (O) within a predetermined environment (EN) by the use of a radio network (RN), where the radio network (RN) comprises a plurality of anchors (a1, ..., a5) being receivers for radio signals (rs) within the radio network (RN), wherein a position estimator (PE) in a positioning server (PS) is provided for estimating the position (pos) of the object (O) based on propagation delays of radio signals (rs) received by different anchors (a1, ..., a5) in response to a transmission event of a radio tag (t) on the object (O), the method comprising the following steps for a respective anchor (a1, ..., a5) of the plurality of anchors (a1, ..., a5): i) training for the respective anchor (a1, ..., a5) a signal classifier (SC) and locally storing in the respective anchor (a1, ..., a5) the trained signal classifier (SC) for being executed on the respective anchor (a1, ..., a5), the signal classifier (SC) being configured to classify the signal quality of a radio signal (rs) received by the respective anchor (a1, ..., a5) by determining a signal quality class (QC1, ..., QCn) out of a plurality of signal quality classes (QC1, ..., QCn) for the received radio signal (rs) based on signal amplitude information (CIR) of the received radio signal (rs); ii) configuring the respective anchor (a1, ..., a5) such that the respective anchor (a1, ..., a5) forwards a received radio signal (rs) to the positioning server (PS) for a position estimation by the position estimator (PE) only in case that the signal quality class (QC1, ..., QCn) determined by the signal classifier (SC) of the respective anchor (a1, ..., a5) fulfills a threshold criterion (TC) with respect to a minimum signal quality.