A machine learning-assisted framework for the inverse design of pixelated millimetre-wave patch antennas targeting the 22--30 GHz band is presented. The antenna surface is represented as a 19x23 binary pixel grid on a Rogers RT/duroid 5880 substrate, where each pixel is either metal or empty, with a continuous electrical path from the feed enforced by design. An initial dataset of approximately 6,000 full-wave CST simulations was collected from structured random pixel patterns, of which only around 40% achieved a resonance with |S11| <= -10 dB anywhere in the band, resulting in an imbalanced dataset. To improve simulation efficiency, an XGBoost binary classifier was trained on this data to distinguish resonant from non-resonant patterns before simulation. Using the classifier as a pre-simulation filter, an additional 4,000 patterns were selected and simulated, raising the overall proportion of resonant designs in the combined 10,000-sample dataset from approximately 40% to 52%. A hybrid CNN-BiLSTM forward surrogate was then trained on this augmented dataset to predict the full complex S11 response across 801 frequency points, using a physics-guided composite loss that explicitly emphasises resonance dip accuracy. Finally, an inverse design model was developed that optimises in a compact 64-dimensional latent space using gradient descent to generate pixel patterns matching a desired S11 specification. The results show good agreement between the surrogate-predicted and CST-simulated |S11| responses for the generated designs and demonstrate the feasibility of automatically designing and reconfiguring antenna structures.
Figures & tables
Fig. 1: Overview of an AI-assisted inverse patch antenna design process
Fig. 2: Full pixelated patch antenna surface on Rogers RT/duroid 5880 ( 16×12.5 mm, εr=2.2 , h=0.508 mm). All 300 optimisable metal pixels are shown in black. The fixed feed line (cols. 1–11, row 10) is shown in red. Grey regions are structurally excluded.
Fig. 3: End-to-end system pipeline. An XGBoost classifier screens possible patterns to augment the CST dataset. A CNN-BiLSTM forward surrogate and a convolutional AE prior are trained on the augmented dataset. The inverse optimiser searches the 64-d AE latent space through the frozen surrogate to generate antenna patterns satisfying the target S11 specification.
Model
Acc.
F1
AUC
Train-Val Gap
Random Forest
80.1%
0.796
0.874
14.7%
XGBoost
77.4%
0.768
0.844
4.4%
LightGBM
77.2%
0.772
0.839
3.4%
SVM
75.9%
0.761
0.827
4.1%
Logistic Reg.
71.0%
0.706
0.790
3.6%
TABLE I: Classifier benchmark on the 6K balanced dataset. Best result per model across five random seeds. Bold : best per column.
Fig. 4: CNN-BiLSTM forward surrogate architecture. The spatial encoder extracts geometric features from the binary pixel pattern, the frequency projection head reshapes them into a per-frequency sequence, and the bidirectional LSTM models inter-frequency correlations to output the full complex S11 spectrum.
Full-wave electromagnetic (EM) simulation enables accurate patch-antenna analysis but is computationally expensive for large-scale forward prediction and inverse design. We present a mesh-native, physics-augmented graph-learning framework that treats radiation-pattern prediction as signal reconstruction on an irregular surface mesh. For the forward problem, a GPS graph transformer is trained with Physics-Augmented Intermediate Supervision (PAIS), an auxiliary node-level objective that predicts complex surface currents, the physical intermediate linking geometry to radiation. PAIS improves multiple GNN backbones at no inference-time cost, while shuffled-current and non-physical controls show the gain comes from physical correspondence. Direction-conditioned decoding and a differentiable radiation-integral consistency loss further exploit this structure. On an 80,000-sample CST benchmark, GPS+PAIS reaches MSE 0.17 / PSNR 19.67, generalizes to a PCA split, and transfers zero-shot to canonical patches. For inverse design, surrogate-filtered diffusion beats nearest-neighbor retrieval by 32% relative MSE.
Development of modern antennas is a cognitive process that intertwines experience-driven determination of topology and tuning of its parameters to fulfill the performance specifications. Alternatively, the task can be formulated as an optimization problem so as to reduce reliance of geometry selection on engineering insight. In this work, a bi-stage framework for automatic generation of antennas is considered. The method determines free-form topology through optimization of interconnections between components (so-called pixels) that constitute the radiator. Here, the process involves global optimization of connections between pixels followed by fine-tuning of the resulting topology using a surrogate-assisted local-search algorithm to fulfill the design re-quirements. The approach has been demonstrated based on two case studies concerning development of broadband and dual-band monopole antennas.
Khadijeh Askaripour, Adrian Bekasiewicz, Slawomir Koziel
Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Narutowicza 11/12, 80-233 Gdansk, Poland · Department of Engineering, Reykjavik University, Menntavegur 1, 102 Reykjavik, Iceland
Future wireless systems are expected to transform the surrounding space from a passive propagation medium into a smart electromagnetic environment, where engineered surfaces control wave propagation, support wireless sensing, and create programmable electromagnetic fingerprints. A key challenge in realizing this vision is the inverse design of metasurfaces for tailored electromagnetic propagation. While forward analysis evaluates the response of a known geometry, the inverse task starts from a prescribed scattering signature and seeks a physically realizable structure that produces it. This inverse task is inherently nonlinear and often high-dimensional, while candidate solutions may be non-unique and provide no direct indication of practical realizability. Here, we introduce a conditional diffusion framework for inverse design of dielectric resonator metasurfaces from target angular scattering patterns. Trained on T-matrix simulated geometry-response pairs, the model learns a conditional distribution of geometries instead of a deterministic mapping, enabling multiple candidate designs for the ill-posed inverse problem. The best generated metasurface achieves a mean percentage error of 1.39%, outperforming CMA-ES optimization (4.1% after 10 h) while requiring only about one minute for after-training inference. The model also produces lower error distributions than deterministic neural baselines for out-of-distribution spectra, highlighting the potential of diffusion models for efficient metasurface design.