Waveguide Combiner Grid Simulation for Faster Grating Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ray-tracing simulations for light propagation in waveguide combiners are computationally intensive and time-consuming, making them inefficient for rapid design iterations in waveguide-based display systems.
Innovation Solution
A discretized grid-based waveguide model simulates light propagation by treating it as a viscous liquid flowing through cells, using a neural network-like approach without machine learning, where k-vectors are discretized into bins and computations are performed in sequential steps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ray-tracing techniques are used for light propagation simulation, then accuracy is improved, but computation time increases significantly
Solution Approach 1:
The waveguide geometry is divided into discrete grid cells, transforming the continuous ray-tracing problem into a segmented grid-based simulation. This segmentation allows light propagation to be modeled as discrete state transitions between cells, dramatically reducing computation time while maintaining sufficient accuracy for design optimization purposes
Solution Approach 2:
Instead of simulating individual light rays with complex interactions, the patent creates a simplified copy model where light propagation is represented as probability distributions across grid cells. This copied model captures essential propagation behavior without the computational burden of full ray-tracing physics
2Loss of information
If conventional ray-tracing techniques are used, then detailed light interaction analysis is improved, but design iteration speed deteriorates
Solution Approach 1:
The patent extracts only the essential light propagation characteristics needed for design optimization, removing unnecessary computational details of full ray-tracing. By taking out only the critical interaction information and representing it through grid cell transitions, the system achieves fast iteration while retaining sufficient detail for design decision-making
Solution Approach 2:
The simulation approach changes parameters from continuous ray coordinates and interaction points to discrete grid cell states and transition probabilities. This parameter transformation enables faster computation while preserving the essential light-matter interaction information needed for evaluating waveguide performance
3Productivity
If fast grid-based simulation is used, then computation time is reduced, but accuracy compared to ray-tracing decreases
Solution Approach 1:
The grid-based simulation dynamically adapts its resolution and computational effort based on the specific design phase and requirements. During early exploration, coarser grids provide fast results, while finer grids can be applied when higher precision is needed, making the accuracy-speed tradeoff flexible rather than fixed
Data Source
AI summary
A computer-based simulation of light propagation and interactions with a waveguide and optical elements in a waveguide combiner uses a model based on a neural network that inherits its shape and properties from a grid structure superimposed on a waveguide combiner. Machine learning is not utilized as weights between nodes in the network are based on physical and geometrical rules which removes the need for training. The waveguide combiner is modeled as a stack of two-dimensional layers that are divided into cells. The k-vector space describing direction and wavelength of diffracted beams for the waveguide combiner is adapted to be non-continuous such that k-vectors are discretized into individual bins that are respectively associated with the different layers. Simulation computations are carried out in a sequence of discrete steps directed to interactions among the cells in which light energy is exchanged with their neighbors from within and between layers.