SAW FEM Block Cascading for Semi-Infinite Substrate Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The finite element method (FEM) simulations for surface acoustic wave (SAW) devices face challenges due to difficulties in modeling semi-infinite substrates and high memory consumption, leading to limited simulation accuracy and long computation times, making it impractical for complex designs and frequent re-simulations.
Innovation Solution
The implementation of hierarchical cascading in FEM simulations, which partitions SAW devices into unit blocks, allowing for independent calculation of sub-models at different frequency ranges and parallelization, reducing memory consumption and simulation times while avoiding synchronization issues in parallel computing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If FEM simulation is applied to SAW devices with arbitrary materials and complex structures, then simulation accuracy and generality are improved, but memory consumption and computation time increase significantly
Solution Approach 1:
The patent applies segmentation by dividing the SAW device structure into multiple unit blocks that can be simulated independently. Each unit block represents a portion of the overall device, allowing the simulation to process smaller sub-problems separately and combine results, thereby reducing memory consumption and computation time while maintaining simulation accuracy for complex structures with arbitrary materials
Solution Approach 2:
The patent introduces a hierarchical dimension to the simulation approach by organizing unit blocks into a tree structure with different levels. This hierarchical organization allows the simulation to manage complexity across multiple scales, enabling accurate modeling of complex structures without proportionally increasing computational burden
2Measurement precision
If FEM simulation models semi-infinite substrate crystal, then simulation accuracy is improved, but modeling difficulty and computation complexity increase
Solution Approach 1:
The patent segments the semi-infinite substrate modeling problem into finite unit blocks with defined boundaries. By representing the effectively infinite substrate as a collection of manageable unit blocks, the patent reduces modeling complexity while preserving the essential physics of semi-infinite substrate behavior through proper boundary condition implementation
Solution Approach 2:
The patent uses periodic boundary conditions that effectively copy the unit block structure to represent the extended semi-infinite substrate. This copying approach allows accurate modeling of infinite substrate behavior using a finite computational domain, reducing both modeling difficulty and computation complexity
3Productivity
If hierarchical cascading partitions SAW devices into unit blocks for independent calculation, then memory consumption and computation time are reduced, but simulation complexity increases
Solution Approach 1:
The patent implements segmentation by partitioning the SAW device into unit blocks organized in a hierarchical tree structure. This segmentation enables independent calculation of each unit block, reducing memory consumption and computation time. The hierarchical organization manages the inherent complexity by providing a systematic framework for combining results from multiple blocks
Solution Approach 2:
The patent merges individual unit block solutions through hierarchical combination to produce the overall device response. This merging process systematically integrates results from segmented calculations, maintaining accuracy while managing complexity through the hierarchical framework that organizes the combination operations
4Productivity
If parallelization is applied to accelerate simulation, then productivity is improved, but synchronization issues and computation complexity increase
Solution Approach 1:
The patent segments the simulation into independent unit block calculations that can be executed in parallel. Each unit block represents an independent computational task with well-defined inputs and outputs, enabling parallelization without complex synchronization. The hierarchical structure naturally organizes these parallel tasks across multiple processing units
Solution Approach 2:
The patent implements self-service through the hierarchical cascading structure where each unit block independently processes its own calculation and automatically provides results to parent blocks in the hierarchy. This self-service approach eliminates the need for complex centralized synchronization, as each computational unit operates autonomously and combines results through predefined hierarchical relationships
Data Source
AI summary
The present disclosure provides systems and methods for scalable and parallel computation of hierarchical cascading in finite element method (FEM) simulations of surface acoustic wave (SAW) devices. Different computing units of a cluster or cloud service may be assigned to independently model different core blocks or combinations of core blocks for iterative cascading to generate a model of the SAW devices. Similarly, frequency ranges may independently be assigned to computing units for modeling and analysis of devices, drastically speeding up computation.