Leaky-Wave Antenna Analysis using Multi-Modal Network Theory with Open Periodic Boundaries

arXiv:2604.17571 · physics.optics, physics.app-ph · Submitted 2026-04-19 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Leaky-Wave Antenna Analysis using Multi-Modal Network Theory with Open Periodic Boundaries".

Mira: Here's the gist of the paper. This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs).

Kai: The approach is hybrid, combining analytical techniques with a commercial full-wave solver.

Mira: Dispersion Analysis Framework The first method computes the dispersion diagram of periodic LWAs by being iterative and relying on the full-wave simulation of a single unit-cell coupled with the analytical solution of an eigenvalue problem.

Kai: This method effectively captures both the phase and attenuation constants while using fewer modes than previous methods with commercial frequency-domain solvers.

Title and authors: Kai: So we were just talking about the title "Leaky-Wave Antenna Analysis using Multi-Modal Network Theory with Open Periodic Boundaries" and who wrote it. The authors are Oscar Senlis, Student Member, IEEE, John N. Le, Student Member, IEEE, Anthony Grbic Fellow member of IEEE, Mauro Ettorre Fellow member of IEEE Vincent Laquerbe Senior Member of IEEE and David González-Ovejero Senior Member of IEEE >

Mira: It's a solid team with a good mix of theory and hardware experience there. I'm curious how their specific combination of multi-modal network theory and open periodic boundaries actually tackles the periodicity problem in these antennas >

Lev: From my side, I'm interested in how they handle the coupling between cells when you have this open region; that's where most of the real difficulty lies for us when we try to build things >

Kai: Well, this paper introduces two methods, and it sets up this new framework called OPB-MNT as a way to analyze periodic leaky-wave antennas. It’s presented as a hybrid approach that mixes analytical techniques with using a commercial full-wave solver >

Mira: The implication there is that they are trying to bridge the gap between purely theoretical solutions and the practical reality of simulating complex structures using high-end solvers >

Lev: If this works out, it means we might be able to get faster results for characterizing these antenna structures without having to run massive, computationally expensive simulations from scratch every single time >

Kai: Exactly. It’s about making the analysis more efficient by cleverly using the simulation data in a structured mathematical way >

The paper's summary: Mira: So, if we summarize what this paper is actually doing, they have this first method for calculating the dispersion diagram of periodic LWAs. It’s iterative; you simulate one unit-cell with open boundaries and then solve an eigenvalue problem to find the correct wavenumber >

Kai: And that process is designed to capture both the phase and attenuation constants accurately, which is what we need for designing these antennas >

Lev: But they also have a second, more concrete contribution. They developed a hybrid method based on OPB-MNT specifically for analyzing reception problems in periodic LWAs >

Mira: That's where it gets interesting because they can compare how the unit-cell responds in two different scenarios, transmission versus reception, and look at what those differences actually are >

Kai: It’s not just time-reversal symmetry that makes them different; transmitting an LWA excites an inhomogeneous plane wave while receiving one that is homogeneous, so the comparison isn't simple time reversal >

Lev: So they are trying to quantify the difference between sending a signal out and picking a signal up in a periodic structure, which is crucial for applications where you might be designing both transmitters and receivers on the same chip >

Mira: That sounds like it moves beyond just characterizing the antenna's properties; it starts analyzing its functional response in real-world scenarios >

The paper's improvements: Kai: Now let's talk about what they claim are the specific improvements of this OPB-MNT framework over previous work. They point out that by relying on periodic boundary conditions to account for inter-cell coupling in the open region, they reduce the simulation domain down to just a single unit-cell >

Mira: That sounds like a big simplification because it means their simulation domain size doesn't really matter as much for the final result >

Lev: That’s actually really important from an engineering standpoint. If you can reduce the required domain size, you reduce the computational cost significantly when you move towards building hardware that needs to be fast iterative solvers >

Kai: And another big point is that because of this hybrid nature, they eliminate the need to implement a custom solver for them >

Mira: I think that hybrid nature is key. It means they are not just throwing a complicated math problem at a black box; they are using the solver output to inform the mathematical solution, which makes it more reliable >

Lev: So if we can avoid writing and debugging custom solvers, that’s a huge overhead reduction for anyone trying to get this into production for something like an error correction system >

Conclusion: Kai: So wrapping up this paper on "Leaky-Wave Antenna Analysis using Multi-Modal Network Theory with Open Periodic Boundaries," the main point is that the OPB-MNT framework is a useful tool for designing and characterizing a wide range of LWAs, gratings, reflectors, and metasurfaces >

Mira: They successfully provide an accurate recovery of both the dispersion characteristics and the response in reception by relating those driven and eigen problems >

Lev: And they point out that the differences between those driven and eigen problems come from them not being time-reversal symmetric, which is a necessary caveat for their analysis >

Kai: So, it sounds like this framework successfully relates the input excitation problem to the system's natural modes in a way that gives us accurate results >

Mira: It’s a solid tool for understanding how these structures behave across different operating regimes, and I think we should keep an eye on this for future research into more complex geometries >

Lev: For real hardware implementation, the key will be figuring out exactly how to map this iterative process onto a system that can handle the required convergence speed and fidelity >

Oscar Senlis, John N. Le, Anthony Grbic, Mauro Ettorre, Vincent Laquerbe, David González-Ovejero

University of Michigan · Michigan State University · Centre National d’Etudes Spatiales

physics.optics, physics.app-ph

Submitted: 2026-04-19

Updated: 2026-11-04

Comments: 1st version: 9 pages, 11 figures, submitted to IEEE TAP. 2nd version: 10 pages, 13 figures, added 2 new study cases

License: http://creativecommons.org/licenses/by-nc-sa/4.0/

Importance score: 92/100

The gist: This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs).

Key concepts

OPB-MNT
Multi-Modal Network Theory with Open Periodic Boundaries. This framework uses a combination of iterative solving and eigenvalue problems, leveraging full-wave simulations on a single unit cell with periodic boundary conditions to accurately determine the dispersion characteristics of LWAs.
Dispersion Analysis Framework
A method for finding how the LWA's properties change with frequency. It works by iteratively minimizing the difference between an assumed complex wavenumber and one derived from solving an eigenvalue problem, effectively capturing both phase and attenuation constants.
Receiving Problem Analysis
A hybrid method to analyze how a periodic LWA responds when receiving a plane wave. Unlike transmission analysis, this method compares the unit cell's response to an incident wave by solving a matrix equation derived from full-wave simulation results.
Hybrid Method Advantages
The OPB-MNT framework is advantageous because it reduces the simulation domain to just one unit cell using periodic boundary conditions, eliminating the need for custom solvers and making it independent of the open region's size.

Terminology

Summary

This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs). p. 0

Where the paper says it

1 Leaky-Wave Antenna Analysis using Multi-Modal Network Theory with Open Periodic Boundaries Oscar Senlis, Student Member, IEEE, John N. Le, Student Member, IEEE, Anthony Grbic, Fellow, IEEE, Mauro Ettorre, Fellow, IEEE, Vincent Laquerbe, David González-Ovejero, Senior Member, IEEE Abstract—This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs).

The approach is hybrid, combining analytical techniques with a commercial full-wave solver. p. 0
Where the paper says it

Le, Student Member, IEEE, Anthony Grbic, Fellow, IEEE, Mauro Ettorre, Fellow, IEEE, Vincent Laquerbe, David González-Ovejero, Senior Member, IEEE Abstract—This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs). The approach is hybrid, combining analytical techniques with a commercial full-wave solver.

Dispersion Analysis Framework

The first method computes the dispersion diagram of periodic LWAs by being iterative and relying on the full-wave simulation of a single unit-cell coupled with the analytical solution of an eigenvalue problem. This method effectively captures both the phase and attenuation constants while using fewer modes than previous methods with commercial frequency-domain solvers. p. 0

Where the paper says it

It is iterative and relies on the full-wave simulation of a single unit-cell of a LWA, coupled with the analytical solution of an eigenvalue problem. This method effectively captures both the phase and attenuation constants of periodic LWAs while using fewer modes than previous methods with commercial frequency-domain solvers.

The framework is referred to as OPB-MNT, which yields accurate results while requiring fewer modes than earlier methods. p. 0
Where the paper says it

It yields accurate results while requiring fewer modes than earlier methods [17], and is largely independent of the simulation domain size of the open region. This new framework, referred to as MNT with open periodic boundaries (OPBs) is abbreviated as OPB-MNT.

Receiving Problem Analysis

The second concrete contribution of this work is the development of a hybrid method based on OPB-MNT to analyze a periodic LWA in reception. This method allows us to compare the unit-cell’s response in two different scenarios (reception and transmission) and examine the differences between them. p. 1

Where the paper says it

The second concrete contribution of this work is the development of a hybrid method based on OPB-MNT to analyze a periodic LWA in reception. This method allows us to compare the unit-cell’s response in two different scenarios (reception and transmission) and examine the differences between them.

The receiving case is not simply time-reversed versions of each other, as a transmitting LWA excites an inhomogeneous plane wave while it receives a homogeneous plane wave. p. 0
Where the paper says it

, the case in which the antenna is fed to radiate. Strictly speaking, however, this assumption is inaccurate given that a transmitting LWA excites an inhomogeneous plane wave, while it receives a homogeneous arXiv:2604.

Methodological Advantages

The OPB-MNT framework offers several advantages over previous methods.

by relying on PBCs to account for inter-cell coupling in the open region, the simulation domain is reduced to a single unit-cell

the hybrid nature of the framework eliminates the need to implement a custom solver

The method is largely independent of the simulation domain size of the open region.

Validation and Comparison

The dispersion analysis is validated by computing results for several classic LWA unit-cells and comparing them to those obtained through full-wave simulations of the full-length antenna and other state-of-the-art methods. For example, the OPB-MNT results are in very good agreement with the MoM [29]. p. 4

Where the paper says it

The values calculated using the OPB-MNT, employing a single mode (M = 1) at the wave ports, are compared with those obtained using the finite-length antenna method using 30 cells (black line), the MoM results from [29] (red line) and the MMTMM in [17] (green line). The dispersion obtained from OPB-MNT is in very good agreement with the MoM [29], which is taken as ground truth.

The imaginary part computed with the OPB-MNT exhibits significantly fewer unwanted oscillations compared to other methods. p. 4
Where the paper says it

For the former, a reference solution based on Oliner’s theory is computed (red line), while for the latter, the reference values are obtained from a finite-length antenna simulation (black line) based on the unit cell shown in Fig. constant, the imaginary part computed with the OPB-MNT exhibits significantly fewer unwanted oscillations compared to the MMTMM [17] and the finite-length antenna method, particularly in the lower frequency range (14 to 22 GHz).

Conclusion on Utility

The paper concludes that the multi-modal network theory with open periodic boundaries (OPB-MNT) is expected to be a useful tool for the design and characterization of a wide range of LWAs, gratings, reflectors, and metasurfaces. The framework successfully provides accurate recovery of the dispersion characteristics and response in reception by relating the driven and eigen problems. p. 7

Where the paper says it

The first allows for the eigen-analysis of a transmitting periodic LWA. The problem is solved using an iterative method which enables accurate recovery of the dispersion characteristics of periodic LWAs.

The differences between the driven and eigen problems are due to the fact that they are not time-reversal symmetric. p. 6
Where the paper says it

Finally, at lower frequencies (14 to 22 GHz), Re{ZB+} is slightly higher in the driven problem than in the eigen problem. The differences between the driven and eigen problems are due to the fact that the driven (incident plane wave problem) and the eigen problems are not time-reversal symmetric.

Pruning redundant reasoning tokens

The first step involves using a commercial full-wave solver to simulate the unit-cell with periodic boundary conditions applied to the open region. In the second step, post-processing is performed to solve a multi-modal eigenvalue problem for the guided region of the unit-cell. p. 0

Where the paper says it

The first step involves using a commercial full-wave solver to simulate the unit-cell with periodic boundary conditions applied to the open region. In the second step, post-processing is performed to solve a multi-modal eigenvalue problem for the guided region of the unit-cell.

The iterative process terminates once ∆κ falls below a prescribed convergence threshold ϵκ. p. 2
Where the paper says it

The complex wavenumber imposed across the open region is considered correct when it matches the eigenvalue obtained from (7). The iterative process terminates once |∆κ| falls below a prescribed convergence threshold ϵκ.

The root of this function, κ root, therefore corresponds to the correct wavenumber κ as it implies κ root = κ eig. p. 3
Where the paper says it

This additional step consists of applying a root-finding procedure to the function F(κn) = κ eig n −κn, which represents the difference between the complex wavenumber imposed in the current simulation iteration (κn) and the wavenumber retrieved from the eigenvalue problem (κ eig n ). The root of this function, κ root, therefore corresponds to the correct wavenumber κ as it implies κ root = κ eig.

How the training signal is built

The total port voltages and currents V2, I2 and I3 are computed by solving the matrix equation in (6) for a given voltage excitation Vs and incident angle θ. This requires a full-wave simulation of the setup shown in Fig. p. 2

Where the paper says it

In this case, we terminate the Floquet port with the characteristic impedance of the Floquet modes to recover the correct S-parameters. Alternatively, we can replace the Floquet port in the full-wave simulation setup (Fig.

2 for the specific incident angle θ. p. 5
Where the paper says it

To this end, a fullwave simulation of the setup shown in Fig. 2 is performed for the specific incident angle θ.

The power at ports 1 or 2 (denoted as i) entering the unit-cell for mode n is P(n) i = V(n) i I(n) i∗. p. 5
Where the paper says it

Once the voltages and currents at port 2 are found using (6), V1 and I1 are computed using (2) while V3 is computed using (3). The power at ports 1 or 2 (denoted as i) entering the unit-cell for mode n is P (n) i = V (n) i I (n) i ∗ , (10) and the Bloch impedances of port mode n in the negative and positive direction are given by Z (n) B− = − V (n) 1 I (n) 1 , Z(n) B+ = − V (n) 2 I (n) 2 .

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI system can perform accurate dispersion analysis of periodic leaky-wave antennas using a hybrid iterative method based on multi-modal network theory with open periodic boundaries (OPB-MNT). This allows for the computation of the dispersion diagram by iteratively minimizing the absolute difference between the complex wavenumber imposed on the open region and that obtained from solving an eigenvalue problem.

  2. Improved AI system can analyze reception problems for leaky-wave antennas by determining the response of a unit-cell to an incident plane wave. This is achieved by solving a matrix equation based on full-wave simulation results to find the total port voltages and currents at the wave ports, enabling comparison between the receiving and transmitting problems.

  3. Improved AI system can validate its analysis against various numerical methods by comparing OPB-MNT results with finitelength antenna simulation data for different LWA unit-cell designs. This provides a robust verification of the method's accuracy across a range of frequencies and structures, showing that the OPB-MNT results are in very good agreement with those from the finite-length antenna method.

  4. Improved AI system can provide detailed physical insights into frequency response by identifying specific physical phenomena, such as when the α curve reaches a maximum at a resonant frequency or when additional stop bands are observed, which helps in understanding the behavior of the structure across different regimes.

Abstract

This paper introduces two methods for analyzing periodic leaky-wave antennas (LWAs) within a new framework denoted as multi-modal network theory (MNT) with open periodic boundaries (OPBs). The approach is hybrid, combining analytical techniques with a commercial full-wave solver. The first method computes the dispersion diagram of periodic LWAs. It is iterative and relies on the full-wave simulation of a single unit cell of a LWA, coupled with the analytical solution of an eigenvalue problem. This method effectively captures both the phase and attenuation constants of periodic LWAs while using fewer modes than previous methods with commercial frequency-domain solvers. The method is validated by computing the dispersion of classic LWA unit cells and comparing them to those obtained through full-wave simulations of the full-length antenna and other state-of-the-art methods. The second, also based on OPB-MNT, focuses on LWA driven analysis. Specifically, it determines the response of a unit cell to an incident plane wave. To validate this method, we compute the response of LWAs with different unit cell designs. By comparing these results with the corresponding dispersion analysis, we highlight the differences between the driven case and eigenvalue problems and identify the scenarios in which the former is applicable.

Related papers