RF Circuit Modeling with Parallel RL Networks for Impedance Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating SPICE models for frequency-dependent transmission lines in power distribution networks, global clock trees, and coplanar waveguides struggle to accurately represent frequency-dependent resistance and inductance, leading to inaccuracies in signal delay, slew, and noise analysis.
Innovation Solution
A computer-implemented method using a coordinate transformation to determine the values of resistive and inductive elements in a passive circuit, ensuring exact matching of DC, low-frequency, and high-frequency impedance targets through a series connection of parallel RL elements, employing an amplitude-phase coordinate system to solve for the elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If discrete resistive and inductive elements are used to build SPICE models, then the model can be simplified for transient simulations, but the accuracy in representing frequency-dependent resistance and inductance deteriorates
Solution Approach 1:
The patent segments the frequency-dependent impedance behavior into multiple discrete parallel RL elements, where each element represents a specific frequency range. This segmentation allows the model to accurately capture frequency-dependent characteristics across different bands while maintaining simplicity in transient simulations.
Solution Approach 2:
The patent changes the parameters (resistance and inductance values) of the discrete elements based on frequency-dependent target values. By adjusting these parameters according to measured or simulated frequency responses, the model achieves high accuracy in representing impedance characteristics across multiple frequencies.
2Measurement precision
If frequency-dependent resistance and inductance are measured and translated to SPICE models, then accurate RF models can be generated, but the process complexity and computation time increase
Solution Approach 1:
The patent performs preliminary measurements or simulations to obtain frequency-dependent resistance and inductance target values before model generation. These pre-computed target values are then used to directly determine the discrete element parameters, avoiding time-consuming iterative optimization during the model generation process.
Solution Approach 2:
The patent creates a simplified SPICE model that copies the essential frequency-dependent characteristics from the complex measured or simulated impedance data. By replicating the key behavioral features rather than the complete complexity, the model achieves accurate RF representation with reduced generation time.
3Ease of operation
If SPICE models are used to analyze waveforms including signal delay and coupling, then practical circuit analysis can be performed, but the accuracy of signal delay and noise analysis deteriorates due to model simplifications
Solution Approach 1:
The patent segments the impedance model into multiple parallel RL elements, each optimized for specific frequency ranges. This segmentation enables the model to accurately represent signal delay and coupling effects across different frequency components while maintaining ease of use in standard SPICE circuit analysis.
Solution Approach 2:
The patent replaces complex frequency-dependent transmission line models with an equivalent discrete RL element model that behaves similarly in transient simulations. This substitution maintains practical circuit analysis capability while improving accuracy in signal delay and noise predictions.
Data Source
AI summary
A computer-implemented method is provided for modeling a circuit having a resistive element, an inductive element, and element pairs connected in series. The operations include determining values of the resistive element, the resistive elements, the inductive element, and the inductive elements with respect to a target DC resistance value, a target low-frequency inductance value, and a set of N target resistance values and N target inductance values at a set of N frequency values. The operations include establishing a first set of 2N equations including a first set of 2N unknowns respectively corresponding to the N target resistance values and the N target inductance values. The operations include introducing a coordinate transformation which includes replacing the first set of 2N unknowns with a second set of 2N unknowns in a transformed coordinate system.