RFID Tag Positioning by Best-Sensor Selection and AOA Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RFID systems face challenges in accurately determining the position of passive RFID tags due to attenuation, dispersion, multipath propagation, and manual errors, leading to multiple and often incorrect position estimates without additional information.
Innovation Solution
An RFID system selects the best sensor or set of sensors for estimating the position of each RFID tag based on metrics derived from detected replies, using measures of dispersion such as variance and standard deviation to filter out spurious angles and positions, and averaging or correlating position estimates from preferred sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are used to estimate RFID tag position, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent segments the sensor network into individual sensor units, each independently estimating tag positions. The system divides the complex task of position estimation into multiple independent sensor perspectives, allowing the appliance to collect and compare estimates from different sensors to improve overall accuracy while maintaining manageable complexity at each sensor node.
Solution Approach 2:
The patent merges position estimates from multiple sensors at the appliance level. By combining measurements from multiple independent sensors and applying statistical analysis (comparing estimates to determine consistency), the system achieves improved measurement precision through data fusion while keeping individual sensor units relatively simple.
2Measurement precision
If all sensors process position estimates, then measurement precision improves, but use of energy increases
Solution Approach 1:
The patent extracts the heavy processing function from individual sensors and centralizes it at the appliance. Sensors perform only lightweight local estimation and transmit results to the appliance, which performs the energy-intensive task of comparing estimates and determining consistency. This extraction of processing functions reduces energy consumption at distributed sensor nodes while maintaining centralized analytical capability.
3Loss of time
If position estimates are averaged without selection, then processing time reduces, but measurement precision deteriorates
Solution Approach 1:
The patent performs preliminary filtering of position estimates by comparing them for consistency before final processing. The appliance identifies and eliminates outliers or inconsistent estimates from individual sensors before computing the final position, ensuring that only reliable estimates are averaged. This preliminary action maintains precision while enabling efficient processing of the filtered set.
Solution Approach 2:
The system implements feedback by comparing position estimates from multiple sensors to assess their consistency. This comparison feedback mechanism allows the appliance to identify reliable versus unreliable estimates, adjusting the processing accordingly to maintain accuracy while optimizing processing time through selective averaging of consistent estimates.
Data Source
AI summary
Radio-frequency identification (RFID) systems use readers to query and locate passive RFID tags in stores, warehouses, and other environments. A signal from the reader powers up the tag, which modulates and backscatters the signal toward the reader. The reader or an appliance coupled to the reader can estimate the tag's position based on the angle of arrival (AOA) of the backscattered signal. In some situations, AOA measurements by different readers may yield different position estimates for the same tag. If these position estimates are close enough to each other (e.g., within the expected imprecision or error radius), they can be averaged to improve precision. If not, the appliance can measure the variance or another measure of dispersion for each reader's position estimates, then pick the reader with the lowest dispersion as the preferred or best sensor for locating that tag, improving precision and reducing processing time.