EMI De-Embedding for Large Grain Bins Using Partial VNA Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electromagnetic imaging (EMI) systems using partial Vector Network Analyzers (VNAs) face challenges in de-embedding cable and switching effects due to varying cable lengths and temperature fluctuations, which are not feasible to measure in the field, leading to inaccurate grain moisture content measurements in large storage containers.
Innovation Solution
A 2-port network de-embedding technique using S11 and S21 measurements to model and remove the effects of cables and switching channels, estimating S22 and S12 parameters, and calibrating the system for accurate imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard de-embedding methods are used that depend on knowing cable parameters, then accurate calibration can be achieved, but cable measurements must be performed before installation which is not feasible for custom-cut cables
Solution Approach 1:
The patent performs preliminary characterization of the measurement system components (VNA, switch, cables) during manufacturing when cable parameters can be measured. This creates a de-embedding model that can be applied later in the field without requiring physical cable measurements, thus resolving the contradiction between needing accurate parameters and the inability to measure custom-cut cables on-site.
Solution Approach 2:
The patent creates a virtual model (copy) of the measurement system's electromagnetic characteristics that can be used for de-embedding operations. Instead of requiring physical cable measurements in the field, the system uses a pre-created digital representation of the measurement chain that can be applied repeatedly without physical intervention.
2Measurement precision
If automatic calibration with calibration loads is performed, then measurement accuracy is improved, but the process cannot be performed in the field due to systematic cable attachment requirements
Solution Approach 1:
The patent extracts the calibration process from the physical field operations and relocates it to the manufacturing/initial setup phase. By performing the calibration characterization when the system is accessible in a controlled environment, the need for physical calibration loads and systematic cable attachments is eliminated from field operations.
Solution Approach 2:
The system performs self-characterization during manufacturing where all components can be systematically connected and measured. This initial self-service calibration creates a permanent de-embedding model that the system then uses autonomously in the field without requiring external calibration equipment or complex setup procedures.
3Measurement precision
If full S-parameters are measured for de-embedding, then complete system characterization is achieved, but partial VNA systems that only measure S11 and S21 cannot perform standard de-embedding
Solution Approach 1:
The patent changes the approach from measuring all four S-parameters to using only two parameters (S11 and S21) that are available from partial VNA systems. By developing de-embedding methods that work with this reduced parameter set, the system achieves complete characterization without requiring full VNA capabilities, thus resolving the contradiction between measurement completeness and device complexity.
4Measurement precision
If cable parameters are measured before installation, then de-embedding accuracy is improved, but cable lengths vary during installation making pre-measurement impractical
Solution Approach 1:
The patent performs preliminary measurement of cable parameters during manufacturing when cables are still accessible and can be characterized. This creates a baseline de-embedding model that accounts for the specific cable lengths and characteristics used in the installation, allowing the system to maintain accuracy even with varying cable lengths since each installation's cables are measured and characterized during the initial setup phase.
Data Source
AI summary
In one embodiment, a system comprising: a container configured to store commodity; a measurement system comprising a vector network analyzer (VNA), a switch module, a plurality of cables, the container, and a plurality of antennas coupled to interior walls of the container, the switch module configured to switch signals transmitted to and received from the plurality of antennas via a plurality of channels, the VNA configured to measure scattering parameters (S-parameters) of all of the plurality of channels; a non-transitory computer readable medium comprising software; and a processor configured by the software to: de-embed a combined effect of the measurement system based on a 2-port network de-embedding technique using only a subset of the S-parameters; and provide an image of the commodity using an inversion algorithm based on input of a calibrated S-parameter after the de-embedding.