Multilayer Dual-polarized Microstrip Antenna Design by Topology Optimization with Enhanced Bandwidth
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.
Rosa: We're looking at the paper: "Multilayer Dual-polarized Microstrip Antenna Design by Topology Optimization with Enhanced Bandwidth".
Dev: Here's the gist of the paper. Dual-polarized (DP) microstrip antennas are utilized in wireless systems for efficient data transmission, and this contribution proposes to design DP microstrip antennas with enhanced bandwidth using a density-based topology optimization approach.
Rosa: Problem Motivation Conventional microstrip antennas typically exhibit a narrow bandwidth which limits their performance in modern systems such as satellite communications, MIMO, and radar systems.
Dev: Traditional approaches to designing DP antennas rely heavily on theoretical analysis and the expertise of design engineers.
Rosa: A common strategy to achieve dual polarizations is to impose structural symmetry on the design
4: –
8: .
Paper summary: Rosa: So, wrapping up the paper "Multilayer Dual-polarized Microstrip Antenna Design by Topology Optimization with Enhanced Bandwidth," the authors show that they achieved two optimized designs operating around five point seven GHz that balance high isolation, which is over forty decibels, with an impedance bandwidth of around ten percent <ref:2608.05712#pg1,Multilayer Dual-polarized Microstrip Antenna Design by Topology Optimization with Enhanced Bandwidth>.
Dev: They also have a second design where they extended the impedance bandwidth to a range from five point five nine GHz to six point zero nine GHz, achieving about nine point nine percent fractional bandwidth while keeping that isolation level between the two ports high at over forty decibels <ref:2608.05712#pg1>.
Taro: The paper suggests that using an optimization algorithm is a supportive tool for designing compact, multiobjective microstrip antennas when you are employing multilayer PCBs. It moves beyond just looking at what works on a single layer and shows how stacking layers can help you achieve better results in terms of bandwidth.
Rosa: Basically, the title highlights that they used topology optimization to enhance the bandwidth of a dual-polarized microstrip antenna by using two layers. It’s about finding that sweet spot where you get good isolation, which is crucial for dual polarization, and wide enough bandwidth for practical use.
Dev: It means that instead of just relying on theoretical analysis or manual design choices, you can use an algorithm to explore the structure of the antenna and find configurations that optimize multiple things at once. The implication is that this method could be useful when designing antennas for systems like MIMO or satellite communications where bandwidth is a real constraint.
Taro: I think what it means for the listener is that complex antenna designs aren't just about knowing all the equations beforehand; you can use an iterative design process to find better physical structures that handle multiple performance requirements simultaneously.
Rosa: That’s right, so this work gives us a method for creating more versatile dual-polarized antennas using topology optimization on multilayer structures. We'll keep an eye on how researchers apply this optimization technique to other antenna types in the future.
Conclusion: Rosa: So, this paper is about taking dual polarized microstrip antennas and using topology optimization on multilayer PCBs to get better bandwidth.
Dev: Yeah, it looks like they focused on how the structure of those layers affects the performance when you optimize for both bandwidth and isolation at the same time.
Taro: It’s not just about stacking things randomly; they are actually designing a specific topology where that dual polarization gets enhanced in terms of how wide its operating range is.
Rosa: The authors, Rosa and Dev, used this optimization approach to minimize reflection coefficients and mutual coupling between the ports.
Dev: They found that by optimizing these density variables, specifically those related to the material presence or absence at each layer, they could achieve a significant improvement over traditional methods.
Taro: Their results show two specific designs. Design one gave them about five point eight percent bandwidth in one design, and Design two pushed that much wider to nearly ten percent bandwidth.
Rosa: And what’s really interesting is that the mutual coupling between the two ports stayed below negative twenty decibels for most of the operating frequency band across both optimized designs.
Dev: That’s a solid number for isolation, especially when you're dealing with dual polarization where those coupling issues can really mess with things.
Taro: It suggests that this algorithm-based design is a useful tool when you need compact, multilayer antennas that perform well in terms of bandwidth and separation.
Rosa: It moves the goal from just making a single layer antenna better to using the whole stack of layers as one optimization variable for better performance.
Dev: So, what this means for practical applications is that we can design these complex multi-objective antennas without relying only on heavy theoretical analysis or guesswork.
Taro: The implication is that when you’re building systems like MIMO or radar where bandwidth and isolation are both tight constraints, this optimization method offers a way to explore the design space more effectively.
Rosa: It shows that using topology optimization isn't just for abstract theory anymore; it can directly inform the physical structure of a multilayer antenna.
Dev: But we still have to consider how this translates to real-world manufacturing and how robust those designs are when you put them into a production environment.
Umea University · Karlstad University · Proant AB
eess.SY, cs.CE, cs.SY
Submitted: 2026-08-06
Updated: 2026-11-16
Comments: This work has been submitted for possible publication
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 81/100
The gist: Dual-polarized (DP) microstrip antennas are utilized in wireless systems for efficient data transmission…
Key concepts
- Dual-polarized (DP) Microstrip Antennas
- These are antennas designed to transmit signals using two different polarizations simultaneously. This is crucial in modern wireless systems like MIMO and radar because it allows for more efficient data transmission and better signal reception.
- Topology Optimization
- This is an advanced design method where the physical structure of the antenna is determined by minimizing specific performance criteria, such as reflection loss or coupling. It lets the computer automatically determine the best material distribution within a given space.
- Density-based Topology Optimization
- In this approach, the antenna's shape is controlled by a density vector (p), where '1' means material is present and '0' means it is absent. This method allows for complex, optimized structures that are tailored to achieve specific goals like maximizing bandwidth.
Terminology
Summary
Dual-polarized (DP) microstrip antennas are utilized in wireless systems for efficient data transmission, and this contribution proposes to design DP microstrip antennas with enhanced bandwidth using a density-based topology optimization approach. p. 0 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. X, 2026 1
Multilayer Dual-polarized Microstrip Antenna Design by
Topology Optimization with Enhanced Bandwidth
Pan Lu, Eddie Wadbro, Viktor Lundstrom, Jonas Starck, Martin Berggren, Emadeldeen Hassan ¨
Abstract—Dual-polarized (DP) microstrip antennas are utilized in
wireless systems for efficient data transmission.Where the paper says it
Problem Motivation
Conventional microstrip antennas typically exhibit a narrow bandwidth which limits their performance in modern systems such as satellite communications, MIMO, and radar systems. Traditional approaches to designing DP antennas rely heavily on theoretical analysis and the expertise of design engineers. p. 0 Therefore, the design of DP microstrip antennas with
enhanced bandwidth is of great significance. Traditional approaches to designing DP antennas rely heavily on
theoretical analysis and the expertise of design engineers. Traditional approaches to designing DP antennas rely heavily on
theoretical analysis and the expertise of design engineers. A common
strategy to achieve dual polarizations is to impose structural symmetry on the design [4]–[8].Where the paper says it
Where the paper says it
Optimization Formulation
The optimization problem is formulated to minimize the reflection coefficients and the mutual coupling between the feeding ports, and a gradient-based optimization algorithm is applied to solve it. The objective function is formulated as min σ log D1,1 (σ) D2,1 (σ) D1,p (σ) subject to the governing equations. p. 1 The antenna system satisfies the following energy balance,
X2
i=1
W+
i + Win,Ω∞ = Wout,∞ +
X2
i=1
W−
i + WΩ∞, (2)
where W+
i
and Win,Ω∞ denote the supplied energies at port i and
from a far-field source, respectively; Wout,Ω∞ denotes the outgoing
energy radiated by the system; and WΩ∞ denotes the energy loss in
the system. We formulate the optimization problem
min
σ
log
D1,1 (σ) D2,1 (σ)
D1,p (σ)
, (3)
subject to the governing equations, where σ ∈ [σmin, σmax] denotes
the conductivity in the design domains. Without
this term, the reflection coefficient D1,1 could be trivially minimized
by employing lossy designs [18]. The minimization of the term D2,1
maximizes the isolation between the two ports.Where the paper says it
Where the paper says it
Optimization Methodology
The design variables are typically associated with the entries of a density vector p such that pi ∈ [0, 1], with 0 and 1 denoting the absence and presence of the design material. Filtering techniques are commonly employed to prevent convergence to low-performing local optima or to impose feature size control that might be useful for imposing manufacturing constraints. p. 1 In density-based topology optimization, the design variables are
typically associated with the entries of a density vector p such
that pi ∈ [0, 1], with 0 and 1 denoting the absence and presence
of the design material, respectively. So-called filtering techniques
are commonly employed to prevent convergence to low-performing
local optima or to impose feature size control that might be useful
for imposing manufacturing constraints. , K, acting on the design vector p [17],
p˜ = F
K
F
K−1
. F
1
(p)
(5)
The filtered design variables p˜ are mapped to the physical conductivity used in the simulation through
σ = 109p˜−4
. Based on
Yee’s scheme [20], the discretized form of (4) is given by [18]
∂W−
i
∂σe
= −(h)
3 XN
n=1
E
N−n
e
E
∗n− 1
2
e + E
∗n+ 1
2
e
2
∆t, (7)
where σe is the conductivity at edge e in the mesh, h and ∆t are the
spatial and temporal discretization steps, respectively, Ee and E
∗
e are
the electric and adjoint fields, respectively, and N is the number of
total simulation time steps. From expression (7) and the mapping
p 7→ σe described above, the gradient of the objective function
with respect to the density variables is computed using the chain
rule.Where the paper says it
Where the paper says it
Where the paper says it
Performance Results
The optimized designs exhibit low reflection, high isolation, and enhanced bandwidth. For Design I, the −10 dB impedance bandwidth spans the frequency range 5. p. 2 59 GHz with |S11| ≈
−14 dB. The −10 dB impedance bandwidth spans the frequency
range 5. We present two antenna designs
operating around 5. 7 GHz, which show a compromise in performance
between a high isolation (more than 40 dB) and enhanced impedance
bandwidth (around 10%). 43 GHz to 5. 75 GHz, corresponding to a fractional bandwidth
of approximately 5. 67 GHz and 6. 00 GHz in
simulation and the −10 dB bandwidth extends to a wider range from
5. We present two antenna designs
operating around 5. 7 GHz, which show a compromise in performance
between a high isolation (more than 40 dB) and enhanced impedance
bandwidth (around 10%). 43 GHz to 5. 75 GHz, corresponding to a fractional bandwidth
of approximately 5. Although the −10 dB impedance bandwidth is moderate, the |S11| remains below −5 dB over a wider frequency range. The magnitude of the mutual coupling |S21| between the two ports
remains lower than −20 dB over most of the operating frequency
band, which confirms the high isolation between the two ports.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Conclusion
Our investigations suggest that a good performance can be achieved when the excitation ports are connected to only one layer. Our results demonstrate the effectiveness of algorithm-based design as a supportive tool for employing multilayer PCBs to design compact, multiobjective microstrip antennas. p. 3 In particular, better results, particularly with
respect to broadband response, are obtained when the outer design
layer is actively connected to the probes. Our results demonstrate
the effectiveness of algorithm-based design as a supportive tool
for employing multilayer PCBs to design compact, multiobjective
microstrip antennas.Where the paper says it
TABLE I
Method Layer Size BW(%)
|---|---|---|---|
theoretical analysis(TA), optimization algorithms (OA) 1 1.44 × 1.44 × 0.024 1.8%
OA 2 [5.46,6.03] OA 2 0.50 × 0.50 × 0.031 9.9%
REFERENCES
[1] P. K. Mishra, D. R. Jahagirdar, and G. p. 3 REFERENCES
[1] P. Jahagirdar, and G. Jahagirdar, and G. Kumar, “A review of broadband
dual linearly polarized microstrip antenna designs with high isolation
[Education Column],” IEEE Antenn.Where the paper says it
Where the paper says it
[2] Y. Zhang, P. Li, X. Qin, P. Liang, K. Wei, and Y. p. 3 238–251, 2014. Wei, and Y. Wei, and Y. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol. Li, “Dual-polarized
microstrip antenna with single-dimension-constraint resonant frequency,”
IEEE Trans. Antennas Propag, vol.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[3] B. Feng, L. Li, J.-C. Cheng, and C. p. 4 X, 2026
5
[3] B. Cheng, and C. Cheng, and C. Sim, “A dual-band dualpolarized stacked microstrip antenna with high-isolation and band-notch
characteristics for 5G microcell communications,” IEEE Trans.Where the paper says it
Where the paper says it
[4] C. Chen, “A single-layer single-patch dual-polarized high-gain crossshaped microstrip patch antenna,” IEEE Antennas Wirel. p. 4 4506–4516, 2019. Chen, “A single-layer single-patch dual-polarized high-gain crossshaped microstrip patch antenna,” IEEE Antennas Wirel.Where the paper says it
[5] X. Zhou, F. Wu, Y. Tong, L. Tian, and Z. p. 4 2417–2421, 2023. Tian, and Z. Tian, and Z. Liu, “Ultrawideband
linear-polarized and dual-polarized ME dipole and their array for 5G/6G
applications,” IEEE Antennas Wirel.Where the paper says it
Where the paper says it
[6] S.-H. Zhu, X.-S. Yang, L.-L. Wang, and J. p. 4 149–153, 2026. Wang, and J. Wang, and J. Wang, “Hybrid topology
optimization of dual-polarized wide-beam antenna and its application
in wide-angle scanning array,” Int. Wang, “Hybrid topology
optimization of dual-polarized wide-beam antenna and its application
in wide-angle scanning array,” Int. -Aided Eng. Wang, “Hybrid topology
optimization of dual-polarized wide-beam antenna and its application
in wide-angle scanning array,” Int. -Aided Eng.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[7] Y. He and Y. Li, “Dual-polarized microstrip antennas with capacitive via fence for wide beamwidth and high isolation,” IEEE Trans. p. 4 e23011, 2022. Li, “Dual-polarized microstrip antennas with capacitive via
fence for wide beamwidth and high isolation,” IEEE Trans.Where the paper says it
[8] Q. Wu, Y. Shi, and L. p. 4 5095–5103, 2020. Shi, and L. Shi, and L. Li, “Wideband dual-polarized differential-fed
filtering microstrip patch antenna with high suppression and wide
stopband,” Int.Where the paper says it
Where the paper says it
[9] W. T. Li, W. X. Cai, J. G. He, Y.-Q Hei, and X. p. 4 6682038, 2023. Hei, and X. Hei, and X. Shi, “Low-profile
wideband dual-polarized antenna for millimeter-wave beam steering
applications,” IEEE Trans.Where the paper says it
Where the paper says it
[10] S. Koziel and A. p. 4 7741–
7751, 2023. Koziel and A. Koziel and A. Pietrenko-Dabrowska, “Tolerance-aware multiobjective optimization of antennas by means of feature-based regression
surrogates,” IEEE Trans.Where the paper says it
Where the paper says it
[11] J. Li, W.-Q Qiu, P. Xiao, Z.-Liu, G. Li, W. T Joines, and S Liao, “An adaptive evolutionary neural network-based optimization design method for wideband dual-polarized antennas,” IEEE Trans. p. 4 Joines, and S. Liao, “An
adaptive evolutionary neural network-based optimization design method
for wideband dual-polarized antennas,” IEEE Trans.Where the paper says it
[12] L.-L Wang, X. p. 4 149–153, 2026. Wang, and J. 8165–8172, 2023. Yang, and C. Yang, and C. Ma, “An efficient gradient-based hybrid parameter-topology optimization for antenna design,” IEEE Trans.Where the paper says it
Where the paper says it
Where the paper says it
[13] P Lu, E Wadbro, J Starck, M Berggren, and E Hassan, “Topology optimization of decoupling feeding networks for antenna arrays,” IEEE Trans. Antennas Propag., vol. 74, no. 1, pp. 467–479
[14] A H Bokhari, E Hassan, and E Wadbro, “Topology optimization of microwave frequency dividing multiplexers,” Struct. Multidiscip. p. 4 Wadbro, “Topology optimization of
microwave frequency dividing multiplexers,” Struct. Multidiscip. Wadbro, “Topology optimization of
microwave frequency dividing multiplexers,” Struct. Multidiscip.Where the paper says it
Where the paper says it
[15] E Hassan, B Scheiner, F Michler, M Berggren, E Wadbro, F Rohrl, Z Zorn, R Weigel, and F Lurz, “Multilayer topology optimization of wideband SIW-to-waveguide transitions,” IEEE Trans. Microw. Theory Techn. p. 4 Lurz, “Multilayer topology optimization of
wideband SIW-to-waveguide transitions,” IEEE Trans. Theory
Techn.Where the paper says it
[16] J Gedeon, I Allayarov, A Cala Lesina, and E Hassan, “Time-domain topology optimization of power dissipation in dispersive dielectric and plasmonic nanostructures,” IEEE Trans. Antennas Propag., vol. 73, no. 5, pp. 3079–3094
[17] E Hassan, E Wadbro, L Hagg, and M Berggren, “Topology optimization of compact wideband coaxial-to-waveguide transitions with minimum-size control,” Struct. Multidiscip. Optim., vol. 57, pp. 1765–1777
[18] E Hassan, E Wadbro, and M Berggren, “Topology optimization of metallic antennas,” IEEE Trans. Antennas Propag., vol. 62, no. 5, pp. 2488–2500
[19] J A Roden and S D Gedney, “Convolution PML (CPML): An efficient FDTD implementation of the CFS–PML for arbitrary media,” Microw. Opt. Technol. Lett., vol. 27, no. 5, pp. 334–339, 2000
[20] A Taflove and S C Hagness, Computational electrodynamics: the finitedifference time-domain method, 3rd ed. Norwood: Artech House, 2005
[21] K Svanberg, “A class of globally convergent optimization methods based on conservative convex separable approximations,” SIAM J. Optim., vol. 12, no. 2, p. 555–573, 2002
[22] High Performance Computing Center North (HPC2N), 2026, accessed: https://www.hpc2n.umu.se/
| Method | Layer Size | BW |
|---|---|---|
theoretical analysis(TA) 1.44 × 1.44 × 0.024 1.8%
OA [5.46,6.03] OA 2 0.50 × 0.50 × 0.031 9.9%
| Method | Layer Size | BW |
|---|---|---|
theoretical analysis(TA) [5.46,6.03] OA 2 0.50 × 0.50 × 0.031 9.9%
| Method | Layer Size | BW |
|---|---|---|
theoretical analysis(TA) [5.46,6.03] OA 2 0.50 × 0.50 × 0.031 9.
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
-
Improved AI system can design planar multilayer dual polarized microstrip antennas with enhanced bandwidth using a densitybased topology optimization approach to minimize reflection coefficients and mutual coupling while maximizing outgoing energy for low lossy material selection.
-
The improved system can perform design iterations by calculating the objective function gradient via the adjoint field method, as described by equation (4), enabling gradient-based optimization of the design vector p through the globally convergent method of moving asymptotes (GCMMA).
-
The AI system can incorporate manufacturing constraints into the optimization process using filtering techniques, specifically applying a nonlinear filter that
reduces intermediate densities and controls the minimum feature size
to prevent convergence to low-performing local optima. -
The system can achieve better broadband performance by simultaneously optimizing copper distribution in two layers of an FR4 stack-up, allowing both layers to develop different resonance modes when one layer is connected and the other is disconnected.
Abstract
Dual-polarized (DP) microstrip antennas are utilized in wireless systems for efficient data transmission. However, their bandwidth is typically very limited. In this contribution, we propose to design DP microstrip antennas with enhanced bandwidth using a density-based topology optimization approach. We formulate an optimization problem that simultaneously accounts for feeding port matching, the ports' isolation, and far-field dual-polarized performance. To enhance the bandwidth, we employ an FR4 stack-up, in which the copper on two layers is optimized simultaneously. We present two antenna designs operating around 5.7 GHz, which show a compromise in performance between a high isolation (more than 40 dB) and enhanced impedance bandwidth (around 10%). The optimized designs are experimentally validated, showing an excellent agreement between the simulated and measured performance.
Related papers
- Industrial Load Modeling and Optimization for Market-Based Interaction with Power Systems
- Exact and Approximate Convex Reformulation of Linear Stochastic Optimal Control with Chance Constraints
- Certified Set Convergence for Piecewise Affine Systems
- Symbolic Control for Stochastic Systems via Finite Parity Games
- Multiobjective optimization-based design and dispatch of islanded, hybrid microgrids for remote, off-grid communities in sub-Saharan Africa
- Physiology-informed layered sensing for intelligent human-exoskeleton interaction