A Time-domain Approach to the Design of Coupled-Resonator Microstrip Filters

Andrés AltieriarXiv 2021 · DOI · arXiv

Abstract

Coupled-resonator microstrip filters are among the most versatile filter topologies. A known design approach uses full-wave electromagnetic simulations to determine the coupling coefficient between resonators as a function of their relative position. This could be done using time-domain simulations using a fast Fourier transform (FFT) to extract the couplings from the S parameters obtained from time-domain signals. However, this approach has a poor performance in terms of resolution and specially for weak couplings, leading to unreasonably long simulation times. To overcome this, we introduce a technique to obtain the couplings directly from time signals, without moving to the frequency domain. This procedure works for strong and weak couplings, with much shorter simulation times and a reduced simulation domain over the FFT approach. This technique is used to design coupled resonator filter efficiently from time domain simulations. We implement this procedure using the finite-difference time-domain framework using an open source solver and discuss our implementation. We show that its results are very similar to previously published ones obtained from frequency domain simulations, even in the case of very weak couplings. Finally, we design and measure a filter to show the good performance of the proposed approach.

Introduction

In microwave engineering, coupled-resonator microstrip filters are essential for bandpass applications. Designing these filters requires precise knowledge of the coupling coefficient ($K$) between resonators, which varies with their physical spacing. Traditionally, engineers use full-wave electromagnetic simulations like Finite-Difference Time-Domain (FDTD) and apply a Fast Fourier Transform (FFT) to identify resonant peaks in the frequency domain.

The frequency response |S2,1| of coupled resonators exhibits two distinct peaks, f- and f+, which are used to calculate the coupling coefficient.

As shown in the frequency response, the coupling manifests as a split in the resonant frequency. However, the FFT's resolution is inversely proportional to the simulation time. For narrow-bandwidth filters with weak coupling, these peaks are extremely close, necessitating prohibitively long simulations to resolve them accurately.

Methodology: Super-Resolution in the Time Domain

To address the resolution bottleneck, this paper proposes a method to extract coupling coefficients directly from time-domain signals. This approach bypasses the frequency domain entirely during the extraction phase.

General setup and physical implementation of coupled hairpin resonators for coupling extraction.

The workflow involves:

  1. Direct Excitation: A single resonator is excited with a Gaussian current pulse, reducing the simulation domain by eliminating complex feed lines.
  2. Signal Processing: The voltage response on the adjacent resonator is filtered using a Gaussian bandpass filter to isolate the modes of interest.
  3. ESPRIT Algorithm: Instead of an FFT, the ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques) algorithm is used. ESPRIT is a super-resolution technique that treats the signal as a sum of complex exponentials, allowing for precise frequency estimation even with short data records.

Implementation and Experimental Results

The methodology was implemented using the open-source FDTD solver gprMax. The accuracy of the time-domain extraction was validated against established Method of Moments (MoM) data for hairpin resonators. The results demonstrated a near-perfect match, even for very weak couplings ($K \approx 0.014$) where traditional methods often struggle.

To further verify the approach, a 5th-order Butterworth filter centered at 3.75 GHz was designed and fabricated on a Rogers RO4350 substrate. Measurements from a Vector Network Analyzer (VNA) showed excellent agreement with the simulated design, confirming the practical utility of the proposed time-domain technique.

Conclusion

By leveraging super-resolution algorithms like ESPRIT, the proposed time-domain approach offers a significant improvement in efficiency and accuracy for microstrip filter design. It reduces simulation times and domain sizes while maintaining high precision for both strong and weak resonator couplings. This method provides a powerful tool for the rapid development of high-performance microwave filters.

Summary updated

Abstract

Coupled-resonator microstrip filters are among the most versatile filter topologies. A known design approach uses full-wave electromagnetic simulations to determine the coupling coefficient between resonators as a function of their relative position. This could be done using time-domain simulations using a fast Fourier transform (FFT) to extract the couplings from the S parameters obtained from time-domain signals. However, this approach has a poor performance in terms of resolution and specially for weak couplings, leading to unreasonably long simulation times. To overcome this, we introduce a technique to obtain the couplings directly from time signals, without moving to the frequency domain. This procedure works for strong and weak couplings, with much shorter simulation times and a reduced simulation domain over the FFT approach. This technique is used to design coupled resonator filter efficiently from time domain simulations. We implement this procedure using the finite-difference time-domain framework using an open source solver and discuss our implementation. We show that its results are very similar to previously published ones obtained from frequency domain simulations, even in the case of very weak couplings. Finally, we design and measure a filter to show the good performance of the proposed approach.
BibTeX
@misc{altieri2021time,
  author = {Andrés Altieri},
  title = {{A Time-domain Approach to the Design of Coupled-Resonator Microstrip Filters}},
  year = {2021},
  eprint = {2106.13737},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2106.13737},
}
Figure