Related Experiment Video
Updated: Aug 3, 2026
14:14
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
Published on: August 12, 2018
8.9K
Fast Prediction Method for Scattering Parameters of Rigid-Flex PCBs Based on ANN
Jingling Mei1,2,3, Haiyue Yuan4, Xinxin Guo4
1Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, China.
Sensors (Basel, Switzerland)
|April 13, 2024
Summary
A new neural network model accurately predicts rigid-flex printed circuit board (PCB) parameters for Indium Gallium Arsenide (InGaAs) detectors. This innovation speeds up the design of aerospace infrared systems, improving signal quality.
Area of Science:
- Electrical Engineering
- Aerospace Engineering
- Materials Science
Background:
- Indium Gallium Arsenide (InGaAs) detectors are crucial for aerospace short-wave infrared (SWIR) applications requiring high signal-to-noise ratios.
- Rigid-flex printed circuit boards (PCBs) are integral to InGaAs detector systems, with their ground plane design critically influencing parasitic capacitance and weak signal integrity.
- Traditional simulation and experimental methods for optimizing PCB designs are time-consuming and expensive.
Purpose of the Study:
- To develop a rapid and precise method for evaluating and optimizing rigid-flex PCB designs for InGaAs detection systems.
- To investigate the impact of rigid-flex board parameters on weak infrared analog signals in aerospace applications.
- To introduce a novel neural network-based approach for predicting scattering parameters of rigid-flex boards in InGaAs detection links.
Main Methods:
- Utilized software simulations to generate a comprehensive dataset of rigid-flex board scattering parameters.
- Trained a backpropagation (BP) neural network model using the simulated sample data.
- Validated the BP neural network model on a real-world rigid-flex board from an aerospace SWIR mission.
Main Results:
- The neural network model accurately predicted high-speed interconnect scattering parameters for various rigid-flex board configurations.
- Achieved a prediction error of less than 1% when compared to a 3D field solver.
- Demonstrated the model's efficiency in overcoming the limitations of iterative design optimization.
Conclusions:
- The developed neural network model offers an efficient and accurate solution for predicting rigid-flex PCB scattering parameters in InGaAs detection circuits.
- This approach significantly improves the design quality and reduces the optimization time for critical aerospace infrared systems.
- This study represents the first investigation into the specific effects of rigid-flex boards on weak infrared detection signals.
