Intelligent antenna structure design

Abstract

Embodiments can include, for example: an antenna structure including one or more antenna patch structure. The one or more antenna patch structure can include an active antenna element and / or a parasitic antenna element.

Description

INTELLIGENT ANTENNA STRUCTURE DESIGNCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Patent application No. 63 / 650,733, filed May 22, 2024, entitled, “ANTENNA STRUCTURE”, which is incorporated herein by reference in its entirety. This application also claims priority to U.S. Patent Application No. 63 / 738,044, filed December 23, 2024, entitled “ANTENNA STRUCTURE”, which is incorporated herein by reference in its entirety.BACKGROUND

[0002] Embodiments herein are related generally to antenna structures and specifically to intelligent antenna structure designs.

[0003] Antennas are often characterized by several performance metrics. The radiation pattern of an antenna indicates the distribution of emitted energy into space. An omnidirectional antenna emits energy uniformly across all directions, whereas a directional antenna focuses more energy towards specific directions. The gain of an antenna is a measure of its ability to direct energy more efficiently in a particular direction than a reference antenna, typically an isotropic antenna. The polarization of an antenna refers to the orientation of the electric field in the emitted wave. Antenna impedance is the ratio of voltage to current at its input terminals. The operational bandwidth of an antenna is the range of frequencies over which it functions effectively. The efficiency of an antenna is calculated as the ratio of power radiated to the total power input.

[0004] An antenna can be provided as part of an antenna array, which is a setup consisting of multiple antennas arranged in a particular geometric pattern to act as a single unit. Such arrays enhance features like gain, directivity, and beamforming capabilities that are challenging to achieve with a single antenna. Beamforming in antenna arrays is a signal processing technique that manipulates the directionality of the array’s radiation pattern.

[0005] Multi-band antenna structures can be configured for operation in multiple bands.BRIEF DESCRIPTION

[0006] Embodiments can include, for example: an antenna structure including one or more antenna patch structure. The one or more antenna patch structure can include an active antenna element and / or a parasitic antenna element.

[0007] Embodiments can include, for example: an antenna structure including one or more antenna patch pattern. The one or more antenna patch pattern can include active antenna elements and / or parasitic antenna elements.

[0008] Embodiments herein can include, for example, disposing an antenna patch pattern over a ground plane, the antenna patch pattern spaced apart from the ground plane, wherein the antenna patch pattern has been designed with use of a process that includes: generating a plurality of candidate patch patterns; evaluating the plurality of candidate patch patterns; and selecting one of the plurality of candidate patch patterns as the antenna patch pattern in dependence on the evaluating.

[0009] Embodiments herein can include, for example, disposing an antenna patch pattern over a ground plane, the antenna patch pattern spaced apart from the ground plane, wherein the antenna patch pattern has been designed with use of a process that includes generating a plurality of candidate patch patterns, evaluating the plurality of candidate patch patterns, and selecting one of the plurality of candidate patch patterns as the antenna patch pattern in dependence on the evaluating.

[0010] Embodiments can include, for example: an antenna structure including a ground plane and an antenna patch pattern spaced apart from the ground plane. The antenna patch pattern can include one or more active antenna element in wireline communication with a transmitter / receiver and one or more parasitic antenna element.

[0011] Embodiments can include, for example: an antenna structure including a ground plane and an active antenna patch pattern spaced apart from the ground plane, the active antenna patch pattern having at least one active antenna element. The antenna structure can include a tuning antenna patch pattern disposed in stacked relation with the active antenna patch pattern, the tuning antenna patch pattern consisting of one or more parasitic antenna element.

[0012] Embodiments can include, for example: an antenna structure including a ground plane and an antenna patch structure spaced apart from the ground plane. The antenna patch structure can include an antenna patch pattern comprising one or more active antenna element in wireline communication with a transmitter / receiver and one or more parasitic antenna element.

[0013] Embodiments can include, for example: an antenna structure including a ground plane and an active antenna patch structure spaced apart from the ground plane, the active antenna patch structure having at least one active antenna element. The antenna structure can include a tuning antenna patch structure disposed in stacked relation with the active antenna patch structure, the tuning antenna patch structure consisting of one or more parasitic antenna element.

[0014] Embodiments herein can include, for example, a method for retrofitting a previously deployed antenna structure, the method including generating a plurality of antenna structure configurations, wherein respective ones of the antenna structure configurations include one or more component of the previously deployed antenna structure and one or more variable antenna patch structure that is varied between the antenna structure configurations; simulating performance of respective ones of the antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the antenna structure configurations an antenna structure simulation dataset; processing antenna structure simulation data of respective ones of the antenna structure simulation datasets; outputting a design antenna structure in dependence on the processing; and adding a mechanical component fabricated in accordance with the design antenna structure to the previously deployed antenna structure.

[0015] Embodiments herein can include, for example, generating a plurality of antenna structure configurations, wherein respective ones of the antenna structure configurations include one or more variable antenna patch structure that is varied between the antenna structure configurations; simulating performance of respective ones of the antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the antenna structure configurations an antenna structure simulation dataset; processing antenna structure simulation data of respectiveones of the antenna structure simulation datasets; and outputting a design antenna structure in dependence on the processing.

[0016] Embodiments herein can include, for example, receiving, through a user interface, user defined structural attribute data specifying one or more structural attribute of an antenna structure and target performance data specifying one or more target performance characteristic of the antenna structure; generating, in dependence on the user defined structural attribute data, a plurality of antenna structure configurations, wherein respective ones of the antenna structure configurations include one or more variable antenna patch structure that is varied between the antenna structure configurations; simulating performance of respective ones of the antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the antenna structure configurations an antenna structure simulation dataset; and outputting a design antenna structure in dependence on the target performance data and on simulation data of respective ones of the antenna structure simulation datasets.

[0017] Embodiments can include, for example: a method comprising receiving, through a user interface, user defined data associated with an antenna structure. The method can include generating a plurality of candidate antenna structure configurations in dependence on the user defined data. The method can also include simulating performance of the candidate antenna structure configurations, producing simulation datasets corresponding to the simulated performance of the respective candidate antenna structure configurations, and outputting a selected antenna structure configuration in dependence on the user defined data and on the simulation datasets.

[0018] Embodiments can also include, for example: a method comprising receiving, through a user interface, user defined data associated with an antenna structure. The method can include generating a plurality of candidate antenna structure configurations in dependence on the user defined data, and outputting a selected antenna structure configuration in dependence on the user defined data.

[0019] Additional features are realized through the techniques set forth herein. Other embodiments and aspects, including but not limited to methods, computer program product and system, are described in detail herein and are considered a part of the claimed invention.BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Fig. 1 is an exploded assembly view of an antenna structure according to one embodiment.

[0021] Fig. 2 is an exploded assembly view of an antenna structure according to one embodiment.

[0022] Fig. 3 is an exploded assembly view of an antenna structure according to one embodiment.

[0023] Fig. 4 is a top view of an antenna patch pattern according to one embodiment.

[0024] Fig. 5 is a top view of an antenna patch pattern according to one embodiment.

[0025] Fig. 6 is a side view of an alternative antenna structure according to one embodiment.

[0026] Fig. 7 is a perspective view of an alternative antenna structure according to one embodiment.

[0027] Fig. 8 is a top view of an alternative antenna structure according to one embodiment.

[0028] Fig. 9 is a side view of an alternative antenna structure according to one embodiment.

[0029] Fig. 10 is a perspective view of an alternative antenna structure according to one embodiment.

[0030] Fig. 11 is a top view of an alternative antenna structure according to one embodiment.

[0031] Fig. 12 depicts an antenna fabrication system according to one embodiment.

[0032] Fig. 13 is an exploded assembly view of an antenna structure according to one embodiment.

[0033] Fig. 14 is an exploded assembly view of an antenna structure according to one embodiment.

[0034] Fig. 15 is an exploded assembly view of an antenna structure according to one embodiment.

[0035] Fig. 16A depicts a conductive material (e.g., copper) configuration on the single-layer conventional antenna designed using the inverse design technique wherein a 1 shows that a conductive material (e.g., copper) pixel exists and 0 shows there is no conductive material (e.g., copper) pixel in that location.

[0036] Fig. 16B depicts a 3D demonstration;

[0037] Fig. 16C depicts a corresponding simulation return loss result.

[0038] Fig. 16D depicts a corresponding radiation pattern.

[0039] Fig. 16E depicts the conductive material (e.g., copper) configuration on the dual-layer antenna designed using parasitic antenna elements and the inverse design technique.

[0040] Fig. 16F depicts a 3D demonstration;

[0041] Fig. 16G depicts the corresponding simulation return loss result where the conductive material is copper.

[0042] Fig. 16H depicts a corresponding radiation pattern.

[0043] Fig. 17A depicts an antenna patch structure of an antenna structure.

[0044] Fig. 17B depicts an antenna patch structure of an antenna structure.

[0045] Fig. 17C depicts an antenna patch structure of an antenna structure.

[0046] Fig. 17D depicts an antenna patch structure of an antenna structure.

[0047] Fig. 17E depicts an antenna structure having multiple antenna patch structures interconnected with vias spatial regions.

[0048] Fig. 17F depicts an antenna structure having multiple antenna patch structures interconnected with vias spatial regions.

[0049] Fig. 17G depicts an antenna structure having multiple antenna patch structures interconnected with vias spatial regions.

[0050] Fig. 18A is a flowchart illustrating input of generated antenna structure configurations into an artificial intelligence model for determining a design antenna structure for fabrication.

[0051] Fig. 18B depicts providing of matrix inputs mapping to antenna patch patterns of antenna patch structures.

[0052] Fig. 18C depicts aggregation of fitness evaluation across multiple factors.

[0053] Fig. 19A is a flowchart illustrating input of generated antenna structure configurations into a machine learning model for determining a design antenna structure for fabrication.

[0054] Fig. 19B depicts training of a machine learning model.

[0055] Fig. 19C depicts an artificial neural network (ANN).

[0056] Figs. 20A-20B depict an antenna geometry with a thin cross-shape top view footprint configured to facilitate optimized low band (LB) performance.

[0057] Figs. 20C-20D depict performance characteristics associated to the antenna geometry of Fig. 20A-20B.

[0058] Fig. 20E depicts an antenna geometry with a thin cross-shape top view footprint configured to facilitate optimized low band (LB) performance.

[0059] Figs. 21 A-21B depicts an antenna geometry with a thin cross-shape top view footprint configured to facilitate optimized low band (LB) performance.

[0060] Figs. 21C-21D depict performance characteristics associated to the antenna geometry of Fig. 21A-21B.

[0061] Figs. 22A-22D depicts an antenna geometry with a mid thickness cross-shape top view footprint configured to facilitate optimized mid band (MB) performance.

[0062] Figs. 22C-22D depict performance characteristics associated to the antenna geometry of Fig. 22A-22B.

[0063] Figs. 23A-23B depicts an antenna geometry with a rectangular-shape top view footprint configured to facilitate optimized high band (HB) performance.

[0064] Figs. 23C-23D depict performance characteristics associated to the antenna geometry of Fig. 23A-23B.

[0065] Figs. 24A-24B depicts an antenna geometry with a circle-shape top view footprint configured to facilitate optimized high band (HB) performance.

[0066] Figs. 24C-24B depict performance characteristics associated to the antenna geometry of Fig. 24A-24B.

[0067] Figs. 25A-25B depicts an antenna geometry with a diamond-shape top view footprint configured to facilitate optimized high band (HB) performance.

[0068] Figs. 25C-25D depict performance characteristics associated to the antenna geometry of Fig. 25A-25B.

[0069] Figs. 26A-26B depicts an antenna geometry with a thick cross-shape top view footprint configured to facilitate optimized high band (HB) performance.

[0070] Figs. 26C-22D depict performance characteristics associated to the antenna geometry of Fig. 26A-2B6.

[0071] One or more aspects of the present invention are particularly pointed out and distinctly claimed as examples in the claims at the conclusion of the specification. The foregoing and other objects, features, and advantages of the invention are apparent from the following detailed description taken in conjunction with the accompanying drawings in which:DETAILED DESCRIPTION

[0072] An antenna structure 100 defining an antenna system is shown in Fig. 1 . Antenna structure 100 can include ground plane layer 102 having ground plane 105. Ground plane 105 can include conductive material, e.g., metal, metal alloy or other conductive material such as a doped semiconductor. Ground plane 105 can have formed therein opening (aperture) 106 for accommodation of transmission line 101 that provides electrical communication between one or more active antenna element defining an active antenna layer and transmitter and / or receiver 200 (hereinafter transmitter / receiver 200). Transmission line 101 can include, e.g., one or more of a coaxial cable, a conductive via, a feedline embedded in a dielectric substrate, a wire, a balun and / or the like. Ground plane 105, in one embodiment, can be formed of conductive material, e.g., of metal, metal alloy or other conductive material such as a doped semiconductor.

[0073] Embodiments herein recognize that transmission line 101 throughout the views can represent one or more transmission line and that opening 106 of ground plane 105 can represent one or more opening. In some embodiments, for example, different active antenna layers herein can be connected to transmitter / receiver 200 and in wireline communication with transmitter / receiver 200 via respective different transmission lines according to transmission line 101. Such multiple different transmission lines represented by transmission line 101 connected to transmitter / receiver 200 can extend through a common opening of a ground plane according to opening 106, or such multiple different transmission lines can extend through respective different openings of ground plane 105 according to opening 106.

[0074] In a further aspect of antenna structure 100 shown in Fig. 1, antenna structure 100 can include antenna layer 108 stacked over ground plane 105. Antenna layer 108 can include spacer 109 and antenna patch structure 110 (antenna patch). Antenna patch structure 110 can include conductive material, e.g., metal, metal alloy or other conductor such as a doped semiconductor. Antenna patch structure 110 can be formed of conductive material, e.g., of metal, metal alloy or other conductor such as a doped semiconductor. Antenna patch structure 110 can be formed on spacer 109, i.e., on material defining spacer 109. Antenna patch structure 110 herein can have a patch pattern featuring certain characteristics as are set forth herein. Antenna structure 100 as shown in Fig. 1 can include a single antenna layer, i.e., antenna layer 108, and can be regarded as a single layer antenna structure.

[0075] Another antenna structure 100 is shown in Fig. 2. Tn Fig. 2, antenna structure 100 includes components set forth in reference to antenna structure 100 described in reference to Fig. 1, but can include an additional antenna layer, namely antenna layer 112 shown in Fig. 2. Antenna layer 112 can be stacked upon antenna layer 108. Antenna layer 112 can include features according to antenna layer 108. Specifically, antenna layer 112 can include spacer 109 that spaces the layer from its under-layer and antenna patch structure 110. In one aspect, antenna patch structure 110 of antenna layer 112 can be configured differently than antenna patch structure 110 of antenna layer 108.

[0076] As depicted in Fig. 3, antenna structure 100 having features in common with antenna structure 100 of Fig. 1 and Fig. 2 can be expanded to include N antenna layers. In reference to Fig. 3, antenna structure 100 can include antenna layer 150 stacked over antenna layer 112 and zero or more additional antenna layers disposed between antenna layer 112 and antenna layer 150. Antenna layer 150 can define an Nth antenna layer of antenna structure 100 which includes first antenna layer 108 and second antenna layer 112. Nth antenna layer 150, like first antenna layer 108 and second antenna layer 112, can include spacer 109 and antenna patch structure 110. Antenna patch structure 110 of antenna layer 150 can be configured differently from antenna patch structure 110 of antenna layer 112 and antenna patch structure 110 of antenna layer 108. Antenna layers herein can be characterized by an antenna patch structure 110 supported by a spacer. Antenna layers of an antenna structures 100 herein such as antenna layer 108, antenna layer 112, and antenna layer 150 of antenna structure 100 shown in Fig. 3 can share a common aperture, meaning that the radiating elements of the different layers are confined to the same physical area X-Y plane dimension in a specific geometric plane. Where antenna layers herein share a common aperture, the X-Y dimension of the common aperture can be defined by the X-Y dimension of ground plane 105.

[0077] In one aspect, the various antenna layers of antenna structure 100 can be configured to include differentiated design frequency bands. In one embodiment, first antenna layer 108 can have a first design frequency band of any arbitrary band, second antenna layer 112 can have second design frequency band of any arbitrary band and Nth antenna layer 150 can include a third design frequency band of any arbitrary band.

[0078] In one illustrative use case, the first design frequency band, the second design frequency band, and the third design frequency band can be differentiated. In one illustrative use case, first antenna layer 108 can have a low band (LB) design frequency band, second antenna layer 112 can have a mid-band (MB) design frequency band and Nth antenna layer 150 can include a high band (HB) design frequency band. In another illustrative use case, first antenna layer 108 can have a high band (HB) design frequency band, second antenna layer 112 can have a mid-band (MB) design frequency band and Nth antenna layer 150 can include a low band (LB) design frequency band. In another use case, at least one of the first, second and third design frequency bands can be a currently un-allocated frequency band.

[0079] Embodiments herein recognize that new antenna bands for cellular networks, Wi-Fi, satellite communications, and other wireless services are expected to become available. Government agencies, e.g., the Federal Communications Commission (FCC) of the United States, periodically allocate new frequency bands for various types of communication. Embodiments herein recognize that antenna structure 100 of a stacked design can accommodate the introduction of new antenna layers designed for new targeted design frequency bands without an increase in the X-Y dimensions of antenna structure 100. Embodiments herein, in one aspect, facilitate retrofitting and upgrading of a previously deployed antenna structure to support operation in a new one or more design frequency band.

[0080] Embodiments herein recognize that challenges to designing antenna structure 100 can increase as additional antenna layers are introduced. Embodiments herein recognize that maximum power delivered to antenna structure 100 can be yielded when an impedance of antenna structure 100 is impedance matched to transmission line 101. Embodiments herein recognize that impedance matching of antenna structure 100 to transmission line 101 can become increasingly challenging when antenna layers are added due to complex electromagnetic interactions between layers including coupling effects, variations in material properties, frequency-dependent behavior, geometric complexity. Additionally, manufacturing tolerances, thermal and environmental factors, and challenges in accurate simulation and measurement can become increasingly more challenging as additional antenna layers are added. Achieving optimal impedance matching can involve, e.g., advanced design techniques, extensive simulations, iterative testing, and use of impedance matching networks or tuners.

[0081] Embodiments herein can include both structural and fabrication methodology features for impedance matching of antenna structure 100. In one aspect, an optimization solver can be used to evaluate a candidate solution space of candidate impedance matched antenna structure designs that can be evaluated based on one or more performance characteristic. In one aspect, an optimization solver can be used to evaluate a candidate solution space of candidate impedance matched antenna structure designs that can be evaluated not only based on an impedance matching constraint, but on other constraints as well, such as constraints involving, e.g., targeted bandwidths and / or the like. In another aspect, parasitic antenna elements can be proactively introduced into candidate antenna structure designs for evaluation. In another aspect, one or more parasitic tuning antenna layer defined by a spacer supported antenna patch can be introduced into antenna structure 100.

[0082] In general, antenna layers having lower design frequency bands can include relatively larger active antenna elements and antenna layers having higher design frequencies can include relatively smaller active antenna elements. In one embodiment, the antenna patch structure 110 of the respective antenna layers 108, 112, and 150 can be positioned approximately a distance of one quarter of a wavelength from ground plane 105, where the wavelength is the wavelength of the design frequency. In some use cases, designing antenna layers 108, 112, and 150 to be positioned approximately one quarter of a wavelength from ground plane 105 can simplify the design process, e.g., resulting in targeted radiation properties more easily achieved. In one embodiment, each respective antenna layer 108, 112, and 150 can be positioned at an arbitrary distance from ground plane 105.

[0083] In Fig. 4, there is shown an antenna layer comprising antenna patch structure 110 having a certain configuration. Antenna patch structure 110 can include active antenna elements 1102 wireline connected to transmitter / receiver 200 for wireline communication with transmitter / receiver 200 and parasitic antenna elements 1104 not wireline connected to transmitter / receiver 200 for wireline communication with transmitter / receiver 200. Antenna patch structure 110 can be supported by spacer 109 that includes a material capable of structurally supporting antenna patch structure 110. Active antenna elements 1102 of an antenna patch structure herein can be wireline connected to transmitter / receiver 200 via transmission line101 . Active antenna elements 1 102 and parasitic antenna elements 1104 herein can be of any arbitrary shape.

[0084] In Fig. 5, there is shown an antenna layer comprising an antenna patch structure 110 consisting of parasitic antenna elements 1104. The structure of Fig. 5 can be stacked within a range of distances from another antenna layer and in one embodiment can be stacked in close proximity to another antenna layer. The structure of Fig. 5 can be stacked at any arbitrary spacing distance with respect to another antenna layer. Antenna patch structure 110 can be supported by spacer 109 that includes a material capable of supporting antenna patch structure 110. Parasitic antenna elements 1104 herein can be of any arbitrary shape.

[0085] Antenna layers defining an antenna structure 100 herein can include active antenna layers and / or tuning antenna layers. Active antenna layers can include one or more active antenna element and can include zero or more parasitic antenna elements. Tuning antenna layers herein can include one or more parasitic antenna element and can be absent of any active antenna element. The antenna layer of Fig. 4 depicts an example of an active antenna layer. The antenna layer of Fig. 5 depicts an example of a tuning antenna layer. Antenna layers herein can include active antenna layers and / or tuning antenna layers.

[0086] Antenna patch structures 110 herein can include active antenna patch structures and / or tuning antenna patch structures. An antenna patch structure 110 provided by an active patch herein can include at least one active antenna element and zero or more parasitic antenna elements. An antenna patch structure 110 provided by a tuning patch herein can be absent of any active antenna elements and can include one or more parasitic antenna element. The antenna layer of Fig. 4 depicts an example of an active antenna layer having an antenna patch structure 110 that includes active antenna elements 1102 and parasitic antenna elements 1104. The antenna layer of Fig. 5 depicts an example of a tuning antenna layer having an antenna patch structure 110 that includes parasitic antenna elements 1104 but which is absent of any active antenna elements 1102.

[0087] In embodiments herein, antenna patch structure 110 throughout the views can be formed as a layer on spacer 109 which spacer 109 can include one or more layer. Antenna patch structure 110 throughout the views can feature a low profile and a thickness that is aboutconstant throughout its distribution when formed on spacer 109. Antenna patch structure 110 throughout the views can be planar in construction. Antenna patch structure 110 throughout the views can be supported on spacer 109. Spacer 109 can be defined partially, entirely, or substantially entirely, by air. Spacer 109 can be defined partially, entirely, or substantially entirely, by dielectric material. Spacer 109 can be defined partially, entirely, or substantially entirely, by an air dielectric. Spacer 109 can be defined partially, entirely, or substantially entirely, by magnetic material.

[0088] Spacer 109 in one embodiment can include first and second layers. The first layer can be a rigid dielectric layer and can be configured to mechanically support antenna patch structure 110. The second layer can be provided by air to define an air dielectric layer.

[0089] Spacer 109 in one embodiment can include one or more layer of dielectric material. Spacer 109 in one embodiment can consist of a single layer of dielectric material. Spacer 109 in one embodiment can include one or more layer of magnetic material. Spacer 109 in one embodiment can consist of a single layer of magnetic material.

[0090] Antenna layers 108, 112, 150 depicted in Figs. 1-3 are shown as being defined by spaced apart antenna patch structures 110, wherein respective ones of the antenna patch structures 110 are planar in construction and extend at a certain elevation. The spaced antenna patch structures 110 can include active antenna patch structures and / or tuning antenna patch structures. Various processes can be employed for forming an antenna patch structure 110 on spacer 109. Various deposition techniques can be employed, e.g., photolithography and etching, screen printing with conductive ink, additive manufacturing using conductive material, e.g., 3D printing with conductive ink, electroplating with use of a seed layer.

[0091] An alternative architecture for antenna structure 100 is set forth in reference to Figs. 6 through 8. Referring to the side view of Fig. 6, antenna structure 100 can include ground plane 105 and antenna layers 108, 112, 116, 120, 124 and 150. In reference to the antenna layers 108, 112, 116, 120, 124 and 150, antenna layers 108, 116 and 124 can be active antenna layers i.e., defined by active antenna patch structures supported on a layer defining spacer 109. The described active antenna layers 108, 116 and 124 can each be associated to and configured for operation in a certain band in the described architecture of Fig. 6 through 8. Antenna layers 108,116 and 124 can be configured for first band operation, second band operation, and third band operation respectively, and the first, second, and third band can be any arbitrary bands. In one particular embodiment, antenna layers 108, 116 and 124 can be configured for high band operation (HB), mid band operation (MB), and low band operation (LB) respectively. However, any order of bands of possible, and antenna structure 100 can include any number of antenna layers. In a further aspect, based on bandwidth improvement provided with embodiments herein, the bandwidth of any given active antenna layer can be made more likely to accommodate multiple authority-allocated bands. Where an antenna layer herein supports operation in multiple authority-allocated bands, transmitter / receiver 200 can be configured to separate signals of the respective allocated bands in software.

[0092] In a further aspect of antenna structure 100, the described active antenna layers 108, 116 and 124 of Fig. 6 through 8 can respectively have an associated tuning antenna layer that can be devoid of active antenna patch elements.

[0093] Antenna structure 100 of Figs. 6-8 can include antenna layer 112 provided as a parasitic tuning antenna layer associated to antenna layer 108 provided as an active antenna layer, antenna layer 120 provided as a parasitic tuning antenna layer associated to antenna layer 116 provided as an active antenna layer, antenna layer 150 provided as a parasitic tuning antenna layer associated to antenna layer 124 provided as an active antenna layer. In one aspect of being associated to a particular active antenna layer, a tuning antenna layer herein can be positioned closer in proximity to its associated active antenna layer than any other active antenna layer within antenna structure 100. In the embodiment of antenna structure 100 as set forth in Figs. 6 through 8, there is provided one tuning antenna layer for each active antenna layer. In the embodiment of Figs. 6 through 8, there are provided first, second and third active antenna layers 108, 116 and 124 and associated first, second and third tuning antenna layers 112, 120, and 150, i.e., one tuning antenna layer for each of the three bands of operation. The various antenna layers 108, 112, 116, 120, 124 and 150 set forth in reference to Figs. 6-8 can include respective antenna patch structures 110 defining an antenna patch pattern. In one embodiment, each tuning antenna layer can induce capacitive coupling with its nearest neighbor associated active antenna layer and one or more additional active antenna layer. Any active antenna layer herein canalternatively be provided as a parasitic tuning layer and any parasitic tuning layer herein can alternatively be provided as an active antenna layer.

[0094] Another embodiment of antenna structure 100 is set forth in reference to Figs. 9-11. Referring to the side view of Fig. 9, antenna structure 100 can include ground plane 105 and antenna layers 108, 112, 116 and 150. In reference to the antenna layers 108, 112, 116 and 150, antenna layers 108, 112, and 116 can be active antenna layers, i.e., defined by active antenna patch structures supported on a spacer material capable of mechanically supporting the antenna patch structures, and antenna layer 150 can be a parasitic tuning antenna layer that is absent of any active antenna elements wireline electrically connected to transmitter / receiver 200. Alternatively, antenna layer 150 can be provided as an active antenna layer having one or more active antenna element and one or more parasitic antenna element. Any active antenna layer herein can alternatively be provided as a parasitic tuning layer and any parasitic tuning layer herein can alternatively be provided as an active antenna layer.

[0095] Antenna layers 108, 112 and 116 can be configured for first band operation, second band operation, and third band operation respectively. The first, second, and third bands referred to herein can be any arbitrary bands. Referring to Figs. 9-11, antenna structure 100 can include ground plane 105, antenna layer 108, antenna layer 112, antenna layer 116, and antenna layer 120. In the embodiment of Figs. 9 -1 1, antenna layers 108, 1 12, and 1 16 can be active antenna layers, while antenna layer 120 can be a tuning parasitic antenna layer absent of any active antenna elements. In the design of Figs. 9-11, antenna layer 150 provided in such embodiment as a parasitic tuning antenna layer can be defined by a spacer material supported antenna patch structure 110. Antenna patch structure 110 of antenna layer 150 provided as a parasitic tuning antenna layer can be absent of active antenna elements 1102 wireline connected to and in wireline communication with transmitter / receiver 200. Antenna layer 150 provided as a parasitic tuning antenna layer can capacitively and inductively couple with one or more of antenna layers 108, antenna layer 112, and antenna layer 116. Embodiments herein recognize that antenna elements 1102 wireline connected to transmitter / receiver 200 are in wireline communication with transmitter / receiver 200. Embodiments herein recognize that active antenna elements 1102 wireline connected to a feedline wireline connected to transmitter / receiver 200 are wireline connected to and in wireline communication with transmitter / receiver 200.

[0096] The inclusion of one or more parasitic tuning layer in the embodiments of Figs. 1-11 can encourage capacitive and inductive coupling between the tuning antenna layer and the one or more active antenna layer of antenna structure 100.

[0097] Embodiments herein recognize that for maximum power delivery and best performance of antenna structure 100, the impedance of antenna layers defining antenna structure 100 should be impedance matched to an impedance of transmission line 101 that delivers power to the antenna layers defining antenna structure 100.

[0098] Embodiments herein recognize that providing one or more parasitic tuning antenna layer can increase the degrees of freedom in which a solution space of impedance matched antenna structures can be explored. Embodiments herein recognize that when multiple antenna layer antenna structures are designed, it can be increasingly challenging to perform impedance matching between an incoming feedline for powering the antenna layers.

[0099] Embodiments herein recognize that these complexities become more severe as additional bands and active antenna layers are added to an antenna structure. Embodiments herein can feature use of one or more tuning parasitic antenna layer defined by a spacer material supported tuning parasitic antenna patch structure 110 as shown with respect to antenna layers 112, 120, 150 of the embodiments of Figs. 6-8, and antenna layer 150 of Figs. 9-11. Embodiments herein recognize that the deployment of one or more parasitic tuning antenna layer induces capacitive and inductive coupling between the tuning parasitic layer in one or more active antenna layers add degrees of freedom to a solution space of candidate impedance matched antenna structure designs that can be explored.

[0100] In reference to the embodiments of Fig. 6-8, antenna patch structure 110 of antenna layer 108 can be concentrated into 16 separate periodically repeating pattern spaced apart subareas as seen in Figs. 7-8, each defining a patch antenna. The multiple spaced apart subareas (patch antennas) define a first band antenna array. Antenna patch structure 110 of antenna layer 116 can be concentrated into 4 separate periodically repeating pattern subareas as seen in Figs. 7-8, each defining a patch antenna. The multiple subareas (patch antennas) of antenna layer 116 define a second band antenna array. Antenna patch structure 110 of antenna layer 124 can be concentrated into 2 separate periodically repeating pattern subareas as seen in Figs. 7-8, eachdefining a patch antenna. The multiple subareas (patch antennas) of antenna layer 124 define a third band antenna array. As seen in Fig. 7 and 8 an antenna patch structure 110 can define an antenna array that comprises multiple spaced patch antennas. As best seen in Fig. 7, the antenna patch structure 110 at antenna layer 108 can define a 4x4 array of spaced apart patch antennas, the antenna patch structure 110 at antenna layer 116 can define a 2x2 array of patch antennas and the antenna patch structure 110 of antenna layer 124 can define a 2x1 array of patch antennas. The first, second, and third bands referred to herein can be any arbitrary bands.

[0101] In reference to the embodiments of Fig. 9-11, antenna patch structure 110 of antenna layer 108 can be concentrated into 16 separate periodically repeating pattern spaced apart subareas as seen in Figs. 10-11, each defining a patch antenna. The multiple subareas (antennas) define a first band antenna array. Antenna patch structure 110 of antenna layer 112 can be concentrated into 4 separate periodically repeating pattern subareas as seen in Figs. 10-11, each defining an antenna. The multiple subareas (antennas) of antenna layer 112 define a second band antenna array. Antenna patch structure 110 of antenna layer 116 can be concentrated into 2 separate periodically repeating pattern subareas as seen in Figs. 10-11, each defining an antenna. The multiple subareas (antenna) of antenna layer 116 define a third band antenna array. As seen in Figs. 9 and 10 an antenna patch structure 110 can define an antenna array that comprises multiple spaced apart patch antennas. As best seen in Fig. 10 the antenna patch structure 110 at antenna layer 108 can define a 4x4 array of patch antennas, the antenna patch structure 110 at antenna layer 112 can define a 2x2 array of patch antennas and the antenna patch structure 110 of antenna layer 116 can define a 2x1 array of patch antennas. The first, second, and third bands referred to herein can be any arbitrary bands.

[0102] While not shown for focus on antenna patch features, antenna patch structures 110 defining antenna layers of Figs. 6-11 are understood as being supported by a spacer 109 as described in Figs. 1-3. Further, active antenna patch structures 110 defining the active antenna layers of Figs. 6-11 are understood as having active antenna patch elements 1102 wireline connected to and in wireline communication with transmitt er / receiver 200 via transmission line 101 as depicted in Fig. 6-11. As set forth herein, active antenna layers herein can further include parasitic antenna patch elements 1104 that are not in wireline communication with transmitter / receiver 200.

[0103] Embodiments herein recognize that transmission line 101 throughout the views can represent one or more transmission line and that opening 106 of ground plane 105 can represent one or more opening. In some embodiments, for example, different active antenna layers herein can be connected to transmitter / receiver 200 and in wireline communication with transmitter / receiver 200 via respective different transmission lines according to transmission line 101. Such multiple different transmission lines represented by transmission line 101 connected to transmitter / receiver 200 can extend through a common opening of ground plane according to opening 106, or such multiple different transmission lines can extend through respective different openings of ground plane 105 according to opening 106.

[0104] In the case where an antenna patch structure 110 defines multiple spaced apart patch antennas, transmission line 101 represents a wireline connection providing wireline communication between each respective patch antenna and transmitter / receiver 200. A wireline connection providing wireline communication between multiple respective patch antenna and transmitter / receiver 200 can include, e.g., a power divider, or a feed network where an antenna layer is configured for phased array operation.

[0105] Embodiments herein recognize that the inclusion of one or more tuning antenna layer defined by a spaced antenna patch structure can expand a candidate solution space for solving an impedance matched condition characterized by an impedance of multiple antenna layers defining antenna structure 100 being matched with respect to an impedance of a transmission line 101 supplying power to the multiple layers.

[0106] In a further aspect, embodiments herein can include employing an optimization solver for identifying solutions to the described solution space expanded by use of one or more tuning antenna layer. Embodiments herein recognize that using an optimization solver can provide various advantages, including increased efficiency and accuracy in returning solutions to complex mathematical problems, which significantly reduces computation time and minimizes errors compared to manual calculations. Embodiments herein recognize that optimization solvers can be highly scalable, capable of handling large-scale problems with numerous variables and constraints, making them ideal for both industrial and research applications. Embodiments hereinrecognize that optimization solvers can provide flexibility by supporting various optimization problem types, such as linear, nonlinear, integer, and combinatorial optimization.

[0107] Embodiments herein recognize that optimization solver technologies can include use of, e.g., linear and mixed-integer programming solver technologies, nonlinear optimization technologies, constraint programming optimization technologies, metaheuristic solver technologies, genetic algorithm technologies, simulated annealing technologies, and / or quadratic programming technologies.

[0108] Antenna structure 100, in one aspect, can include features that facilitate modeling and evaluation of antenna structure designs by an optimization solver or other artificial intelligence (Al) model. In one aspect, providing antenna layers herein, e.g., parasitic tuning antenna layers, and / or active antenna layers can include configuring antenna patch structure 110 as a pixelated patch structure. Where antenna patch structure 110 is provided as a pixelated patch structure, a spatial area delimiting antenna patch structure 110 can include a grid of spatial regions having positions which can be termed pixel positions, as seen, e.g., in the views of Figs. 7-8, 10-11, 16A, and 16E, as well as Figs. 20A-26D. At each pixel position, the spatial area delimiting an antenna patch structure 110 either includes or does not include a conductive material formation. The spatial regions at the various pixel positions can be rectilinear, e.g., square in shape, or can be another shape. Where antenna patch structure 110 is provided as a pixelated patch structure, candidate design variations can be easily modeled and evaluated, e.g., changing any pixel position from a “occupied” to “not occupied” state (alternatively referred to as “covered” and “not covered” states) defines a new candidate antenna structure configuration for evaluation.

[0109] For evaluation of candidate solutions defining a candidate solution space, the respective candidate solutions can be modeled as a matrix which represents the pattern of an antenna patch structure on a spacer (each element of the matrix represents the existence of a conductive material formation at a pixel position, where “1” expresses a pixel position occupied by conductive material and “0” (zero) expresses a pixel position not occupied by conductive material). For example, a 10x10 pixelized matrix can be reshaped into a 100x1 vector. When using multiple layers, vectors of each layer can be concatenated, resulting in a single vector with the total number of elements equal to the sum of all pixels in an antenna structure configuration.

[0110] In one aspect, pixelated antenna elements can be employed for providing an antenna patch structure 110 as set forth herein. As best seen in Figs. 7-8 and 10-11, antenna patch structures 110 that can be defined by pixelated pattern of defining Is and 0s where Is refer to a pixel position having conductive material and 0 corresponds to a pixel position without conductive material. In one example, the pixel positions can be defined on a grid having rows and columns of pixel positions.[OHl] A system 1000 for use in designing and / or fabricating an antenna structure is shown in Fig. 12. System 1000 can include processing circuitry 310 configured to send, e.g., webpages defining user interface 1202 for display on a display, e.g., of processing circuitry 310 or an external user equipment (UE) device associated to a user, and fabrication device 3112, e.g., a three-dimensional (3D) printer.

[0112] In reference again to Fig. 12, system 1000 can include processing circuitry 310. Processing circuitry 310 can include, according to one example, one or more processors 3101, memory 3102, and one or more input / output interface 3103. One or more processor 3101, memory 3102 and one or more input / output interface can be connected via system bus 3104. Memory 3102 can include a combination of system memory and storage memory. Memory 3102, according to one example, can store one or more programs for facilitating processes that are set forth herein. One or more processors 3101 can run one or more programs stored in memory 3102 to facilitate processes as is set forth herein. Memory 3102 can define a computer readable medium. Processes described herein may be performed by one or more computer systems or other processing devices, as provided, in a server as set forth herein. An example computer system to incorporate and use aspects described herein is depicted and described with reference to Fig. 10. Processing circuitry 310 can include one or more processor 3101, memory 3102, and one or more VO interface 3103, which may be coupled to each other by busses and other electrical hardware elements (not depicted). Processor(s) 3101 can include any appropriate hardware component s) capable of implementing functions, for instance executing instruction(s) (sometimes alternatively referred to as code, firmware and / or software) retrieved from memory 3102. Execution of the instructions causes the processing circuitry 310 to perform processes, functions, and / or the like, such as those described with reference to the methods set forth herein. One or more computer systems configured according to processing circuitry 310 can host one ormore virtual machine (VM) such as a hypervisor based virtual machine, and / or a container based virtual machine as set forth herein. In some examples, aspects described herein are performed by a plurality of homogenous or heterogeneous computer systems coordinated to collectively perform processes, functions, and / or the like, such as those described herein.

[0113] Memory 3102 can include hardware components or other storage devices to store data such as programs of instructions for execution, and other data. The storage devices may be magnetic, optical, and / or electrical based, as examples. Hard drives, field-programmable gate arrays (FPGAs), magnetic media, compact disks (CDs), digital versatile disks (DVDs), and flash memories are example storage devices. Accordingly, memory 3102 may be volatile, nonvolatile, or a combination of the two. As a specific example, memory 3102 includes one or more hard drives and one or more random-access memory (RAM) devices for, respectively, nonvolatile, and volatile storage of data. Example programs stored by memory include an operating system and applications that run on the operating system, such as specialized applications to perform functions described herein. Memory 3102 can define a computer readable storage medium. A computer readable storage medium, as used herein, is not to be interpreted as being transitory signals per se. There is set forth herein a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a one or more processor to carry out methods and / or functions set forth herein.

[0114] System 1000 in one embodiment facilitates fabrication of antenna structure 100 with use of one or more solution optimizer program running on processing circuitry 310 and user interface 1210 which permits a user to specify parameters controlling an optimization. In antenna structure parameters area 1204 a user can specify antenna structure parameter values.

[0115] User defined antenna structure parameter values for input into antenna structure parameters area 1204 can include structural attribute antenna structure parameter values and / or antenna structure target performance parameter values. Structural attribute antenna structure parameter values can include, e.g., number of active antenna layers, number of parasitic tuning layers, the dimensions of antenna layers, the materials defining conductive material formations of the antenna layers, spacing distances between antenna layers, ground plane material, groundplane dimensions, relative order of antenna layers, spatial region (e.g., pixel) dimensions, spatial region shape, spacer thickness, spacer material, inclusion of variable vias spatial regions, patch array inclusion, patch array dimensions, patch array wireline connection, fixed component designations, variable component designations, and / or the like.

[0116] Antenna structure target performance parameter values that can be input into antenna structure parameters area 1204 can include, e.g., target performance parameter values specifying attributes, e.g., of loss, gain, S-parameters (impedance matching), radar cross section at a specific band, pattern, design frequency band, bandwidth, polarization, directivity, and / or the like.Antenna structure target performance parameter values that can be input into antenna structure parameters area 1204 can include, e.g., the design frequency band of any specified antenna layer.

[0117] User interface 1202 can be configured so that a user can also or alternatively specify antenna structure parameter values, such as structural attribute antenna structure parameter values and / or antenna structure target performance parameter values using rendering area 1208. Rendering area 1208 can be configured with multi-touch functionality permitting a user, e.g., to add or subtract antenna layers, pinch in or expand antenna layer dimensions, specify design frequency bands and / or bandwidths within labels 1222, 1224, 1226, change an ordering of antenna layers that are indicated with the labels 1222, 1224, 1226, and 1232, and / or the like.

[0118] User interface 1202 can be configured to include optimization parameters area 1206 which can enable a user to input specific optimization parameter values. Table A below illustrates example parameters which a user can specify using optimization area 1206, where system 1000 is configured to run an evolutionary algorithm for identification of a design antenna structure for deployment satisfying specified target performance characteristics. Table A illustrates example values that can be configurable by user input.

[0119] Table A

[0120] In one aspect as set forth in Table A, optimization solver software running on processing circuitry 310 can include evolutionary algorithm software, which can be provided by genetic algorithm software. With antenna structure parameter values and optimization parameter values selected, a user can activate optimize button 1242 of user interface 1202 to activate optimization On activation of button 1242 processing circuitry 310 can run an optimization solver to produce an optimized antenna structure design. Prior to optimization being performed, rendering area 1208 can present a coarse rendering of an antenna structure being designed, e.g., illustrating general dimensions of antenna layers. Subsequent to optimization, rendering area 1208 can present a precise rendering of an optimized antenna structure design, e.g., illustrating the actual state (covered or not covered) of the various spatial region positions, e.g., pixel positions of the various antenna layers defining the optimized design. For performance of rendering in rendering area 1208, processing circuitry 310 can run one or more 3D rendering program.

[0121] On activation of fabricate button 1244, system 1000 can initiate fabrication of an optimized antenna structure design in accordance with the optimized design being rendered in rendering area 1208. In one embodiment, processing circuitry 310 can send parameters defining one or more antenna layer of the optimized design antenna structure for deployment being rendered in rendering area 1208 to fabrication device 3112, e.g., a 3D printer which fabricates (e g., prints) the antenna patch pattern defined by the antenna patch structure 110 of the one ormore antenna layer on a spacer material substrate, e.g., layer, in accordance with the optimized design antenna structure being rendered in rendering area 1208.

[0122] In running a genetic algorithm, processing circuitry 310 can create an initial population of candidate solutions (each candidate solution being a gene) where each solution can be modeled as a vector which represents the pattern of the antenna on the substrate / substrates provided by spacer material (each element of the matrix represents the existence of conductive material, e.g., copper on that position, where 1 means copper pixel and 0 means no copper).

[0123] On creation of an initial population, processing circuitry 310 can set a population size (M), meaning there are M candidate solutions in each generation (also known as iteration). Higher M will result in faster and better convergence (faster in terms of the number of iterations), however it comes at the cost of time and higher computational resources.

[0124] Processing circuitry 310 can evaluate each individual candidate solution (which can be termed chromosomes defined by genes) with use of appropriate simulation software and then considering the desired response, and processing circuitry 310 can calculate an error for each candidate solution using a fitness function that indicates how close the solution is to a desired response. In one aspect, the fitness function evaluates the error for each response. The error can refer to the sum of the deviations between the current response and the desired response. Deviations in response can refer to deviations in simulated performance characteristics of generated antenna structure configurations to specified target performance characteristics, which can be referred to as goals. A user can specify by input into user interface 1202 as set forth herein one or more target characteristic of a design antenna structure for deployment. In one embodiment a specified target characteristic can be the desired return loss (the parameter calculated based on the input impedance that represents the matching criteria for an antenna structure). For example, there can be specified within the given frequency band that return loss to be less than -15 dB, means that the total reflected power of the antenna would be less than 3% of the input power. A user can input target performance characteristics with respect to, e.g., loss, gain, S-parameters (impedance matching), radar cross section at a specific band, pattern, design frequency band, bandwidth, polarization, directivity, and / or the like.

[0125] By inputting multiple target performance characteristics, e.g., system 1000 can output a design antenna structure satisfying the multiple target characteristics.

[0126] A fitness function that evaluates simulated performance of generated antenna structures relative to one or more target performance characteristic can include a mean squared error (MSE) fitness function. Using MSE as the fitness function works effectively because it provides a clear and quantifiable objective to minimize, ensuring that the algorithm focuses on reducing prediction errors. Its continuous nature supports smooth optimization, allowing the genetic algorithm to fine-tune parameters gradually. Additionally, MSE is universally applicable across various domains, such as regression problems, machine learning model training, and parameter estimation tasks. By evolving candidate solutions to minimize the MSE, genetic algorithms can efficiently optimize complex, multidimensional problems.

[0127] The fitness function can evaluate fitness according to one or more factor, e.g., an impedance match factor, a directionality factor. The factor(s) can be established by inputting into user interface 1202 parameter data defining one or more target performance characteristic (goal) of an output design antenna structure for deployment. A lower error value signifies that the solution is closer to the desired outcome, with an error value of 0 being the ultimate goal, representing a perfect match to the desired response. An example of suitable simulation software that can be employed for evaluation of a candidate solution is the Antenna Toolbox™ of MATLAB®.

[0128] With an error value for each candidate solution produced, processing circuitry 310 can cause individual candidate solutions to compete in small groups (tournaments) and the best individual from each group can be selected to pass their genes to the next generation. This promotes the survival of the fittest. In one example, 80% of the new population can be created by combining parts of two parent solutions. Processing circuitry 310 can emulate reproduction, where offspring inherit traits from both parents. Processing circuitry 310 can select pairs of parents from the mating pool, and their genes can be recombined to produce offspring.Processing circuitry 310 accordingly explore new areas of the solution space.

[0129] In one embodiment, processing circuitry 310 can employ a randomness mutation function which makes small random changes to individual genes in the offspring, promoting geneticdiversity. Mutation helps prevent the algorithm from getting stuck in local optima by introducing new genetic material into the population. At a subsequent next stage, processing circuitry 310 can pass top J (e.g., 10) individuals from the current generation to the next generation without any changes. Processing circuitry 310 accordingly can ensure the best solutions are preserved.

[0130] Processing circuitry 310 can repeat the process of selection, crossover, and mutation for up to K (e.g., 200) generations (also known as iteration). In one embodiment, processing circuitry 310 can stop the algorithm early if no improvement is seen in the best solution for 20 generations (MaxStallGenerations) or if the improvement is less than a specified function tolerance P (e.g., le-5) (FunctionTol erance), indicating convergence. User interface 1202 can be configured to facilitate a user adjusting parameter values defining performance of a Genetic Algorithm to optimize its performance.

[0131] In one aspect, processing circuitry 310 for determining a production design for an antenna patch structure 110 can produce multiple generated antenna patch structures 110, wherein respective ones of the generated patch structures 110 include differentiated patch patterns.

[0132] In Figs. 17A-17C there are illustrated differentiated generated antenna patch structures 110 each having and defining a differentiated patch pattern. Figs. 17A-17C illustrate a top view of generated antenna patch structures 110 looking in the direction of the Z axis of the depicted reference coordinate system. Fig. 17A illustrates generated patch pattern 1, Fig. 17B illustrates generated patch pattern 2 and Fig. 17C illustrates generated patch pattern n.

[0133] In Figs. 17A-17C, generated antenna patch structures 110 have differentiated generated antenna patch pattern. Each generated patch pattern can include a predetermined set of spatial regions having respective associated spatial positions. The set of spatial regions can be distributed over a predetermined area. Each spatial region and position can include one of two possible states, an occupied state and an unoccupied state. An unoccupied state refers to the state where a given spatial region is unoccupied by conductive material. An occupied state can refer to the state where a given spatial region is occupied by conductive material.

[0134] In one aspect, each generated patch pattern can include a common set of spatial predetermined spatial regions distributed over a common spatial area, where the spatial area and its defined spatial regions are logically associated to and represent certain locations in actual physical space, e.g., locations on an antenna layer herein. Generated antenna patch structures 110 defining antenna patch structure configurations herein can be subject to artificial intelligence processing for identification of a design antenna patch structure 110. A design antenna structure 110 herein, which can be provided as part of a design antenna structure, can be provided as a design antenna structure for deployment and / or fabrication.

[0135] In one embodiment, as set forth in Figs. 17A-17C, the different spatial regions defining spatial area of a generated antenna patch structure 110 can include an M by N grid of spatial regions. In one aspect, states of one or more spatial region can be differentiated between the generated patch patterns. For example, in pattern 1 (Fig. 17A) the spatial region positions 210 211 and 212 (spatial region positions M22N10, M23N10, M24N10) have the states “unoccupied”, “occupied”, “occupied” whereas the same spatial region positions 210, 211 and 212 (spatial region positions M22N10, M23N10, M24N10) have the states “unoccupied”, “unoccupied”, “occupied” in the generated patch pattern 2 (Fig. 17B).

[0136] While Fig. 17A-17C depict spatial regions distributed in a grid pattern defined by horizonal rows and vertical columns of spatial region positions, the spatial regions need not be distributed in a grid pattern. In the example of Fig. 17A-17C the depicted spatial regions can be referred to as pixel positions. In the embodiment of Figs 17A-17C, each spatial region can include from a top (Z axis direction) view a rectilinear and rectangular shape and specifically is depicted in the view of Fig. 17A-17C a square shape representing an dxd unit dimension in the X-Y plane, where d can be, e.g., d=lmm, 1cm, 1.5cm, 10cm, etc. depending on the application and system configurations configurable with use of user interface 1202 (Fig. 12). In another embodiment the different spatial regions defining an antenna patch structure 110 can be non-rectilinear. While the various spatial regions defining a patch pattern of an antenna patch structure 110 can be any shape and alternate shapes can provide advantages, providing the various spatial regions to be rectangular from a top view can provide advantages in some embodiments. In one example, providing the spatial regions to be rectangular from a top viewcan facilitate the generation of reliable conductive paths defined by adjacent spatial regions each having an occupied state.

[0137] Spatial areas having spatial regions such as MxN grid regions provided throughout the views facilitates ordered transformation of physical world attributes into digital data structures for accommodation of machine processing. Embodiments herein can express attributes defining an MxN grid of antenna patch structure spatial regions as an MxN matrix of binary (1 or 0) values for consumption by a processing circuitry hosted artificial intelligence model, e.g., an optimization solver and / or a machine learning model trained by supervised learning. MxN grid regions provided throughout the views also facilitates ordered translation of changes of an antenna patch structure configuration into changed inputs provided to a processing circuitry hosted antenna simulator.

[0138] Regarding antenna patch structure spatial regions herein throughout the views, it will be understood that spatial regions herein define active antenna elements or parasitic antenna elements depending on whether the spatial regions are in wireline communication with transmitter / receiver 200.

[0139] Figs. 17A-17C depict an antenna patch structure 110 from a top view looking in the direction of the depicted Z axis. Fig. 17D depicts an antenna patch structure 110 from a side view looking in a direction of the reference Y axis. As seen from the side view of Fig. 17D, the different spatial regions 221-226 (M21N7-M26N7) depicted from a top view in Fig. 17A are depicted in a side view in Fig. 17D. Referring to Fig. 17D, it is seen that the various spatial regions can be logically assigned a common height H which maps to an actual physical space height H at which an actual determined design antenna patch structure 110 can be fabricated, e.g., by depositing of conductive material defining an antenna patch pattern onto a spacer material layer such as by printing. The spatial regions 221-226 depicted in Fig. 17D can be supported at bottom elevation B by material of a layer defining spacer 109 as set forth herein. Spatial regions herein such as spatial regions depicted in Figs. 17A-17G can have predetermined widths and lengths. Spatial regions herein such as spatial regions depicted in Figs. 17A-17G can have predetermined heights. Spatial regions herein such as spatial regions depicted in Figs. 17A- 17G can have predetermined positions. Spatial regions herein such as spatial regions depicted inFigs. 17A-17G can include occupied and unoccupied states. When in an occupied state, a spatial region herein can be occupied by conductive material. When in an unoccupied state, a spatial region herein can be absent of conductive material.

[0140] For identification of a design antenna structure having a design antenna patch structure 110, processing circuitry 310 can produce a plurality of generated antenna patch structures 110 defining differentiated patch patterns, e.g., patch patterns 1-n as depicted in Fig. 17A-17C that can be subject to processing by artificial intelligence processing. Artificial intelligence processing for identification of a design antenna patch structure 110 can include, e.g., optimization solver processing, and / or supervised learning machine learning processing, e.g., as can be provided with use of an artificial neural network (ANN).

[0141] A method for performance by processing circuitry 310 for determination of a design antenna structure 100 for fabrication is described in reference to the flow chart of Fig. 18A.

[0142] At block 1802 processing circuitry 310 can generate a plurality of antenna structures 100 defining antenna structure configurations in the manner set forth in reference to Fig. 17A-17C. That is, at block 1802, processing circuitry 310 can produce a plurality of differentiated generated antenna structures 100 each having one or more antenna patch structure such as antenna patch structure 110.

[0143] In the case where a generated antenna structure 100 defining a generated antenna structure configuration includes a single antenna layer, antenna structures generated at block 1802 can include a single antenna layer with a single antenna patch structure 110 as set forth herein. In the case where an antenna structure design is to include multiple antenna layers each having an associated antenna patch structure 110, the generated antenna structures 100 generated at block 1802 can include multiple antenna patch structures 110, each differentiated antenna patch structure 110 associated to a different antenna layer.

[0144] At block 1804, processing circuitry 310 can record one or more simulator simulation performance parameter dataset for each respective antenna structure 100 defining an antenna structure configuration generated at generate block 1802. Each parameter dataset can include one or more parameter value of a particular parameter domain. Parameter domains can include, e.g.,S-parameter, gain, RCS, bandwidth, beamwidth, etc. For producing a simulation, processing circuitry 310 can input each generated antenna structure 100 generated at block 1802 into a simulator. A set of one or more simulation performance parameter dataset for a respective antenna structure configuration can define an antenna structure simulation dataset.

[0145] When inputting an antenna structure 100 defining a generated antenna structure configuration into a simulator, processing circuitry 310 can input physical data input parameters according to the antenna structure configuration. A physical data input parameter can specify, e.g., a physical dimension and / or a physical material. Where a certain spatial region position of an antenna patch structure of an antenna structure configuration is in an occupied state, a physical data input parameter input into the simulator for evaluation will specify that conductive material is included in the certain spatial region position. Where the certain spatial region position of an antenna patch structure of an antenna structure configuration is in an unoccupied state, a physical data input parameter input into the simulator for evaluation will specify that conductive material is not included in the certain spatial region position. Referring to block 1804 and block 1904, processing circuitry 310 can alter input antenna structure physical data input parameters input into a simulator in dependence on the output antenna structure configuration, e.g., lengthening or shortening conductive paths defining an antenna patch structure in accordance with the output configuration. In one embodiment, generating antenna structure configurations at generate block 1802 can include randomly changing states of randomly selected one or more antenna patch structure 110 defining the antenna structure configurations.

[0146] The generated antenna structures 100 defining generated antenna structure configurations generated at block 1802 can include one or more fixed component and one or more variable component.

[0147] In one common use case, the one or more fixed component can include a ground plane, and the one or more variable component can include one or more antenna layer. In such a use case, the physical data input parameter input into the simulator, e.g., defined by processing circuitry 310, for the ground plane do not vary between the input antenna structures but do in fact vary for the one or more antenna layer (the antenna patch structure 110 for the antenna layers can be varied between the input antenna structure configurations). In one example with reference toFigs. 6 and 7, such a use case can include ground plane 105 being specified as a fixed component, and antenna layers 108, 112, 116, 120, 124, and 150 being designated as variable components.

[0148] In one common use case, the one or fixed component can include a ground plane 105 and a first one or more antenna layer, and the one or more variable component can include a second one or more antenna layer. In one example with reference to Figs. 6 and 7, such a use case can include ground plane and antenna layers 108, 112, 116, 120 being designated as fixed components that are not varied between the generated configurations subject to simulation, and antenna layer 124 and antenna layer 150 being designated as variable components which are varied (e.g., in terms of their patch pattern defined by their antenna patch structure) between the generated configurations. In another example with reference to Figs. 6 and 7, such a use case can include ground plane and antenna layers 108, 112, 116, 120, 124 being designated as fixed component that are not varied between the generated configurations subject to simulation, and antenna layer 150 being designated as variable component which is varied (e.g., in terms of its patch pattern defined by its antenna patch structure) between the generated antenna structure configurations.

[0149] Accordingly, it can be observed that embodiments herein can be employed to retrofit and upgrade a previously deployed antenna structure 100. The retrofitting can include use of one or more additional antenna layer (active and / or tuning) that fits within a common aperture of ground plane 105 and any existing antenna layer defining the antenna structure.

[0150] In reference to Figs. 6 and 7, a previously deployed antenna structure 100 can include the components depicted in Figs. 6 and 7 except for antenna layer 124 and antenna layer 150. For upgrading and retrofitting such an antenna structure to transmit and / or receive wireless signals in an additional authority-allocated band (which can be a band within a current or next generation), processing circuitry 310 can be configured to generate at generate block 1802 antenna structures 100 defining antenna structure configurations with antenna layer 124 provided as an active antenna layer (and optionally antenna layer 150) being specified as variable components having input physical data parameters being varied for each antenna structure configuration input into a simulator and with ground plane 105 and antenna layers 108, 112, 116, and 120 specified asfixed components with input physical data parameters remaining fixed for each configuration input into the simulator.

[0151] In reference to Figs. 6 and 7 according to another use case, a previously deployed antenna structure 100 can include the components depicted in Figs. 6 and 7 except for antenna layer 150, but performance of the antenna structure has changed from its designed performance due to one or more environmental condition change. For upgrading and retrofitting such an antenna structure, processing circuitry 310 can be configured to generate at generate block 1802 antenna structures 100 defining antenna structure configurations with antenna layer 150 provided as a parasitic tuning antenna layer being specified as a variable component having input physical data parameters being varied for each antenna structure configuration input into a simulator and with ground plane 105 and antenna layers 108, 112, 116, 120, and 124 specified as fixed components with input physical data parameters remaining fixed for each configuration input into the simulator.

[0152] In reference to Figs. 6 and 7, according to another use case, a previously deployed antenna structure 100 can include the components depicted in Figs. 6 and 7 except for antenna layer 150. For upgrading and retrofitting such an antenna structure to transmit and / or receive wireless signals in an additional authority -allocated band (which can be a band within a current or next generation), processing circuitry 310 can be configured to generate at generate block 1802 antenna structures 100 defining antenna structure configurations with antenna layer 150 provided as a parasitic tuning antenna layer being specified as a variable component having input physical data parameters being varied for each antenna structure configuration input into a simulator and with ground plane 105 and antenna layers 108, 112, 116, 120, and 124 specified as fixed components with input physical data parameters remaining fixed for each configuration input into the simulator. Using user interface 1202 target performance parameter datasets can be defined that specify expanding an operating bandwidth of one or more active antenna layer of layers 108, 112, 116, 120, and 124. In such an embodiment, where a certain antenna layer supports operation in multiple authority-allocated bands, transmitter / receiver 200 can be configured to separate signals of the various bands in software.

[0153] In one embodiment, ground plane 105 can be specified as a variable component. In one embodiment, ground plane 105 can be configured to include variable pattern of conductive material spatial regions which can be differentiated between generated configurations of ground plane 105. In such an embodiment, the ground plane defines an antenna layer herein.

[0154] The generated antenna structures 100 defining generated antenna structure configurations generated at block 1802 can include zero or more fixed component and one or more variable component. In one embodiment every layer of antenna structure 100, including ground plane 105 can be specified as a variable component.

[0155] In reference to Figs. 6 and 7, according to another use case, a previously deployed antenna structure 100 can include the components depicted in Figs. 6 and 7 except for antenna layer 150. For upgrading and retrofitting such an antenna structure to transmit and / or receive wireless signals in an additional authority-allocated band (which can be a band within a current or next generation) and / or for improved performance of any specified factor, processing circuitry 310 can be configured to generate at generate block 1802 antenna structures 100 defining antenna structure configurations with antenna layer 150 provided as a parasitic tuning antenna layer being specified as a variable components having input physical data parameters being varied for each antenna structure configuration input into a simulator and with ground plane 105 and antenna layers 108, 1 12, 116, 120, and 124 specified as fixed components with input physical data parameters remaining fixed for each configuration input into the simulator. Using user interface 1202 target performance parameter datasets according to target performance characteristics described herein can be defined that specify expanding an operating bandwidth of one or more active antenna layer of layers 108, 112, 116, 120, and 124. In such an embodiment, where a certain antenna layer supports operation in multiple authority-allocated bands, transmitter / receiver 200 can be configured to separate signals of the various bands in software.

[0156] If a suitable design antenna structure is not determined on a first attempt in reference to any use case herein described, a user can adjust inputs to user interface 1202 in order to expand a solution space. In one example, a user can specify a second to nth tuning antenna layer according to antenna layer 150 provided as a parasitic tuning antenna layer and can specify each tuning layer as a variable component that can be varied between generated antenna structureconfigurations. Additionally or alternatively, a user can specify variable vias spatial regions interconnecting antenna patch structures of different elevations. Additionally or alternatively for expansion of a solution space, a user can specify ground plane 105 as a variable component that can be varied between generated antenna configurations.

[0157] In one embodiment, system 1000 can apply constraints to specified variable components in dependence on user defined configuration data defined by a user using user interface 1202 entered in area 1204. In one example, system 1000 can constrain dimensions of individual spatial regions MmNn (pixels) as shown in Fig. 17A-17C in dependence on a selected design frequency band for an antenna layer. In another example, system 1000 can constrain an antenna patch structure 110 to define a certain number and layout of patch antennas (e.g., a 4x4 array of patch antennas for antenna layer 108 depicted in Fig. 10) depending on a design frequency band selected for a layer. In one aspect, system 1000 can be configured so that a user can configure (e.g., activate, deactivate) any constraint.

[0158] An antenna simulator herein can be defined by an appropriately programmed processing circuitry 310. An antenna simulator herein can employ computational algorithms and numerical methods to model the electromagnetic behavior of antennas, predicting their performance under specific operating conditions. A user can define and input as physical data input parameters antenna structure geometry, including physical structure, material properties, and surrounding environment, and specifying operating parameters such as frequency, polarization, and feeding mechanisms. The simulator can create a mesh, dividing the antenna structure into smaller elements to enable detailed calculations. Using numerical techniques like the Method of Moments (MoM), Finite Element Method (FEM), or Finite-Difference Time-Domain (FDTD), the simulator can solve Maxwell’s equations to determine how electromagnetic fields interact with the antenna and its environment. This process can generate key performance metrics, including radiation patterns (2D / 3D distributions of power), gain, directivity, impedance, design frequency band, bandwidth, and efficiency, which are analyzed and visualized through tools such as polar plots, Smith charts, surface current distributions, and field plots. Simulators also allow for optimization by adjusting design parameters to improve performance. Advanced simulators can model real-world environments, incorporating factors like ground effects or nearby structures, to assess antenna behavior in practical scenarios. These tools leverage highcomputational power, often using GPUs for complex simulations, and provide essential insights for applications such as wireless communication, radar, satellite systems, and aerospace technologies. By enabling detailed virtual prototyping, antenna simulators minimize the need for costly physical prototypes, streamline the design process, and help identify potential performance issues early, ensuring efficient, accurate, and effective antenna design and optimization.

[0159] An antenna simulator herein can incorporate tools of MATLAB® which includes an Antenna Toolbox™ for designing, simulating, and analyzing antennas and antenna arrays for applications in wireless communication, radar, and electromagnetics. It offers predefined antenna types like dipoles, patches, and horns, as well as custom antenna design using parametric shapes. Users can design linear, planar, or conformal arrays with control over element spacing and orientation, and simulate far-field and near-field radiation patterns, impedance, gain, and directivity. The toolbox features visualization tools for 2D / 3D radiation patterns, surface currents, and field distributions, along with polar plots and Smith charts for impedance analysis. It integrates with other MATLAB toolboxes, such as the RF Toolbox for circuit analysis and the Phased Array System Toolbox for beamforming and radar simulations, and supports numerical techniques like Method of Moments (MoM) for precise electromagnetic modeling. Examples include designing antennas for specific frequencies, analyzing antenna arrays, and plotting impedance over frequency ranges, making it ideal for developing wireless communication systems, optimizing array configurations, and simulating radar systems. MATLAB® is a registered trademark of MathWorks, Inc. The described simulator can additionally or alternatively incorporate antenna tools of, e.g., Ansys HFSS™, a 3D electromagnetic field simulator for RF and wireless design, CST Studio Suite™ by Dassault Systemes, offering comprehensive 3D electromagnetic simulation capabilities, Altair Feko™, which addresses a broad set of high-frequency electromagnetic simulation applications, and TICRA Tools™, specializing in antenna analysis and synthesis software. The recorded datasets recorded at block 1804 can be recorded into a memory 3102 of processing circuitry 310.

[0160] On completion of record block 1804, processing circuitry 310 can proceed to block 1806.At block 1806, processing circuitry 310 can apply a fitness function. In one embodiment, the mean square error (MSE) fitness function can be applied, as shown in Eq. 1.

[0162] Where n is the total number of data points or observations in the dataset, i is the index of a specific data point (ranging from 1 to n), yi is the target value (the design target) of the i-th data point, and yGiis the predicted value for the i-th data point (the simulation output). Processing circuitry 310 can apply a fitness function according to Eq. 1 for each performance parameter dataset recorded at block 1806.

[0163] On completion of block 1806, processing circuitry 310 can proceed to block 1808. At block 1808, processing circuitry 310 can select elite nominations based on the fitness function applied at block 1806. Processing circuitry 310 with the fitness function applied at block 1806 assigns scores to individuals, and the top-ranking solutions with the highest fitness are preserved as elite nominations, bypassing crossover and mutation to ensure their quality is retained in the next generation. This process, known as elitism, helps preserve the best solutions, accelerates convergence toward the optimal result, and provides stability by reducing the risk of losing high- performing individuals due to random changes. By ranking individuals based on their fitness, the genetic algorithm ensures that only the top performers are directly carried over, while the rest of the population undergoes evolutionary operations to explore new solutions.

[0164] On completion of block 1808, processing circuitry 310 can proceed to block 1810. At block 1810, processing circuitry 310 can generate a new population using crossover and / or mutation. In a Genetic Algorithm (GA), generating a new population using crossover and mutation involves first selecting parents based on their fitness scores using methods such as roulette wheel, tournament, or rank-based selection. Crossover, which combines genetic material from two parents to produce offspring, can be applied to a proportion of the population (e.g., 70-90%) and can be executed using techniques like single-point crossover, where a single split point is chosen, and genes beyond it can be swapped; two-point crossover, which swaps genetic material between two points; uniform crossover, which randomly selects genes from each parent; or arithmetic crossover, which computes offspring genes as weighted combinations of parent genes. Mutation introduces random changes to offspring with a low probability (e.g., 1-5%) to maintain genetic diversity and prevent premature convergence, using methods such as bit-flip mutation for binary representations, swap mutation for positional adjustments, or Gaussianmutation for continuous variables. The resulting output can replace some or all of the current population, with replacement strategies such as generational replacement, where the entire population is replaced, or steady-state replacement, where only the weakest individuals are substituted. Throughout this process, elites — individuals with the highest fitness scores — are often preserved without modification to ensure quality solutions persist, while crossover and mutation explore new potential solutions. This balance of preservation and variation drives the GA’s evolution toward optimal solutions.

[0165] Expressing generated antenna structure configuration with matrices of Is and Os (e g., binary string vectors) can facilitate advantages in the determination of a design antenna structure. In Genetic Algorithms using binary encoding, for example, the processes of crossover and mutation play crucial roles in evolving candidate solutions by manipulating the binary strings that represent them. Crossover refers to a mechanism through which two parent binary strings can be combined to produce new offspring, blending their genetic information to explore potentially better solutions. Crossover can be performed using methods such as single-point crossover, where a random point is selected, and the segments before and after that point are swapped between the parents, or two-point crossover, where two crossover points define a segment that is exchanged. Another common method is uniform crossover, where each bit in the offspring is chosen randomly from either parent with equal probability, allowing for a more granular combination of the parents' genetic material. Mutation can serve as a mechanism to introduce variability and prevent premature convergence by flipping individual bits (changing Os to Is or Is to Os) at randomly chosen positions within a binary string. This process is governed by a mutation rate, typically a small probability, to ensure that the structure of good solutions is not excessively disrupted while still enabling exploration of new areas of the solution space. Together, crossover and mutation work synergistically: crossover exploits existing solutions by recombining their traits, while mutation explores the solution space by introducing randomness. This balance between exploitation and exploration is fundamental to the effectiveness of a Genetic Algorithm, ensuring that the population evolves toward higher fitness over successive generations. Employing an optimization solver such as a Genetic Algorithms (GAs) can conserve computing resources by converging quickly to a solution. Genetic Algorithms (GAs) can natively conserve computing resources by using a population-based search that focuses on promising areas of the solution space, avoiding exhaustive evaluations. Their evolutionaryprocess reuses knowledge from previous generations, while their stochastic nature explores a subset of possibilities rather than the entire solution space. Additionally, GAs are inherently suitable for parallel computation and prioritize finding approximate solutions, further reducing computational demands even without explicit optimization. In embodiments herein, the rapid exploration of candidate solution spaces can include alteration of bits of a binary matrix, which can be provided as a binary string (Kxl matrix defining a vector), wherein a pattern of Is and Os represents one or more antenna patch pattern of one or more antenna patch structure.

[0166] At block 1810, processing circuitry 310 can simulate performance of respective new antenna structure configurations generated at block 1802 in the manner of running a simulation set forth in reference to block 1804 and further at block 1810 can perform recording of one or more simulation performance parameter dataset for the respective new antenna structure configurations generated at block 1810 in the manner set forth in reference to block 1804.

[0167] On completion of block 1810, processing circuitry 310 can proceed to criterion block 1812. At criterion block 1812 processing circuitry 310 can ascertain whether a termination condition has been satisfied. The termination condition can be satisfied, e.g., on a satisfactory solution being identified or on a maximum number of generations being satisfied. On the determination that termination condition has been satisfied, processing circuitry 310 can proceed to block 1814 to output the best performing solution as the design antenna structure. A design antenna structure herein can be a design antenna structure for fabrication and / or deployment.

[0168] On the determination that a termination criterion has not been satisfied, processing circuitry 310 can return to block 1806 where processing circuitry 310 can evaluate the new population generated at block 1810 using the fitness function of Eq. 1. In one embodiment, block 1810 can additionally or alternatively be performed after a “no” determination at block 1812 prior to a next iteration of fitness function application at block 1806. Processing circuitry 310 can iteratively perform the loop of blocks 1806 to 1812 until a termination criterion satisfied at block 1812.

[0169] Figs 18B and 18C depict performance of the method of Fig. 18A, where recorded simulation datasets recorded at block 1804 are recorded for a plurality of parameter domains and where antenna structure 100 included multiple antenna layers 108, 112, 116 specified as variableantenna layers. Tn each antenna layer of the antenna structure as depicted in Fig. 18B, the spatial regions defining an antenna patch structure and patch pattern, e.g., which can include MxN spatial regions as shown in Figs. 17A-17C, can be represented by MxN matrices 1852, where the size of these matrices can vary between antenna layers depending on the specific application. In the MxN matrices, the state of the respective spatial regions of each antenna patch structure and pattern can be represented with the value “1” or “0” in the matrices, e.g., “1” can represent “occupied” and “0” can represent unoccupied, or vice versa.

[0170] Next, processing circuitry 310 can transform the set of MxN matrices into Kxl matrices 1854 defining a set of vectors. Once transformed, processing circuitry 310 can combine the vectors into a single large vector for input into a genetic algorithm interface wherein candidate solutions (chromosomes) are expressed as binary strings defined by a string of Is and 0s. Depending on the artificial intelligence model applied for optimization, an MxN matrix can be applied as an input to the artificial intelligence model without transformation of the MxN matrix into a Kxl matrix defining a vector.

[0171] After feeding the combined binary string vector to the machine defined by an appropriately programmed processing circuitry 310, the model begins optimizing the parameters. In the first step, processing circuitry 310 at block 1802 can generate random values as the initial population, representing potential solutions. These random values are used to extract the required simulation performance parameter datasets for each input antenna configuration, corresponding to the matrix patterns in the different layers, using any EM simulation software (e.g., MATLAB® Antenna Toolbox™). From this population, the GA model defined by an appropriately programmed processing circuitry 310 can calculate simulations outputs based on predefined simulation parameter dataset domains (it can be s-parameter, gain, RCS, bandwidth, beamwidth, etc.), allowing evaluation of the performance. Processing circuitry 310 can compare these results with target values across different frequencies using a fitness function at block 1806 to guide the optimization process.

[0172] The target, or goal evaluated at apply fitness function block 1806 can include several objectives — such as maximizing gain, minimizing certain parameters, or optimizing radar crosssection (RCS). To evaluate the GA model’s performance, processing circuitry 310 can calculatea Figure of Merit (FOM) for each parameter dataset using an appropriate error calculation method, such as the Mean Squared Error (MSE) fitness function described in reference to Eq. 1, which can be employed for evaluating fitness of each generated antenna configuration for each parameter dataset. Processing circuitry 310 can combine these individual FOMs to produce a total FOM, FOMtotal as depicted in Fig. 18C, through a weighted summation, allowing prioritization of certain objectives over others.

[0173] Once the total FOM is calculated, processing circuitry 310 can check whether required performance criteria are satisfied (block 1812). If it does, the process is complete. If not, processing circuitry 310 can continue exploring the solution space, iterating through the optimization process including by performance of block 1810 until, e.g., the desired FOM is achieved, the maximum number of generations is satisfied, or a point is reached where further iterations no longer yield improvements.

[0174] In reference to Fig. 18A an artificial intelligence method was described where generated antenna structure configurations were applied to an optimization solver model defined by a genetic algorithm, and where there was returned an optimized antenna structure design selected for fabrication.

[0175] In reference to Figs. 18A-18C, there is set forth herein generating, e.g., at block 1802 and / or block 1808 a plurality of antenna structure configurations, wherein respective ones of the plurality of antenna structure configurations include one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations; simulating performance of respective ones of the plurality of antenna structure configurations; producing, from the simulating, at block 1804 multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the plurality of antenna structure configurations an antenna structure simulation dataset; processing, e.g., at block 1806 antenna structure simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations; and outputting at block 1814 a design antenna structure in dependence on the processing.

[0176] In one embodiment, the producing, from the simulating, at block 1804 can include recording respective ones of the antenna structure simulation datasets into data storage, e.g.,provided by memory 3102, in association with a matrix representation of the respective ones of the plurality of antenna structure configurations, wherein the matrix representation is defined by a pattern of binary (1 or 0) values, wherein the binary values represent one or more antenna patch pattern and specify occupied or unoccupied states of the spatial regions of the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations. In one embodiment, the matrix representation can be provided by a binary string representing one or more antenna patch pattern, and processing circuitry 310 can by alteration of binary values of the binary string at block 1810 (by crossover and / or mutation) generate new populations of antenna structure configurations.

[0177] In reference to Fig. 19A, there is described another artificial intelligence method by which generated antenna structure configurations can be applied to a machine learning model, and based on the application of generated antenna structure configurations to the machine learning model an optimized design antenna structure can be output. A design antenna structure herein can be a design antenna structure for fabrication and / or deployment.

[0178] At block 1902 processing circuitry 310 can generate antenna structure configurations. In one embodiment, processing circuitry 310 can perform block 1902 in the manner described with reference to block 1802 of the flowchart of Fig. 18 A.

[0179] On completion of block 1902, processing circuitry 310 can proceed to block 1904. At block 1904, processing circuitry 310 can record one or more simulation performance parameter dataset. In one embodiment, processing circuitry 310 at block 1904 can perform recording in the manner of recording block described in reference to block 1804 of the flowchart of Fig. 18A.

[0180] On completion of block 1904, processing circuitry 310 can proceed to block 1906. At block 1906, processing circuitry 310 can train a predictive model using supervised learning. For supervised learning the configurations generated at generate block 1902 and the one or more simulation performance parameter dataset recorded at block 1904 can be applied as training data.

[0181] On completion of block 1906, processing circuitry 310 can proceed to inferencing block 1908. At inferencing block 1908, processing circuitry 310 can perform inferencing of the trained predictive model for identification of a finalized antenna structure design.

[0182] Operations of the flowchart of Fig. 19A are described further in reference to predictive model 902 set forth in reference to Fig. 19B. Referring to predictive model 902 set forth in reference to Fig. 19B, predictive model 902 can be trained with use of supervised machine learning and once trained, predictive model 902 can be subject to inferencing with use of inferencing data. There can be applied as inputs to predictive model 902 training data and inferencing data.

[0183] In one aspect, predictive model 902 can be trained with iterations of training data, wherein an iteration of training data can be applied for each antenna structure configuration generated at generate block 1902.

[0184] Training data for each training iteration can include input training data and outcome training data. The input training data for each training data iteration can include configuration data that specifies a generated antenna structure configuration. In one embodiment, a given antenna structure configuration can be expressed as a set of one or more MxN matrix (as shown by 1852 of Fig. 18B) mapping to the one or more MxN spatial regions of one or more antenna patch structure 110 that define the given antenna structure configuration. In the MxN matrix or matrices applied as training data into predictive model 902, the state of the respective spatial regions of each antenna patch structure and pattern can be represented with the value “1” or “0” in the matrices, e.g., “1” can represent “occupied” and “0” can represent unoccupied, or vice versa.

[0185] With further reference to predictive model 902 of Fig. 19B, outcome training data defining a supervised learning training label for each training iteration can include for the given antenna structure associated to the iteration the one or more antenna structure simulation performance parameter dataset recorded at block 1904.

[0186] Trained as described, predictive model 902 can be configured to be responsive to inferencing data. Trained as described, predictive model 902 can learn a relationship between antenna structure configurations, e.g., determined using random generation, and antenna structure simulation performance parameter datasets. Predictive model 902 in one embodiment can be provided by an artificial neural network (ANN).

[0187] Inferencing data for inferencing predictive model 902 can include one or more target antenna structure performance parameter dataset. In response to application of the described inferencing data, predictive model 902 can output to one or more antenna structure configuration capable of producing the one or more target antenna structure performance parameter dataset. Processing circuitry 310 can select an output antenna structure configuration output by predictive model 902 as the output design antenna structure 100. A design antenna structure herein can be a design antenna structure for fabrication and / or deployment.

[0188] In reference to Figs. 19A-19C, there is set forth herein generating, e.g., at block 1902 a plurality of antenna structure configurations, wherein respective ones of the plurality of antenna structure configurations include one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations; simulating performance of respective ones of the plurality of antenna structure configurations; producing, from the simulating, at block 1904 multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the plurality of antenna structure configurations an antenna structure simulation dataset; processing, e.g., at block 1906 antenna structure simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations; and outputting at block 1908 a design antenna structure in dependence on the processing.

[0189] In one embodiment, the producing, from the simulating, at block 1904 can include recording respective ones of the antenna structure simulation datasets into data storage, e.g. provided by memory 3102, in association with a matrix representation of the respective ones of the plurality of antenna structure configurations, wherein the matrix representation is defined by a pattern of binary (1 or 0) values, wherein the binary values represent one or more antenna patch pattern and specify occupied or unoccupied states of the spatial regions of the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations. Data of the antenna structure simulation datasets in association with the matrix representations of the respective ones of the plurality of antenna structure configurations, can be applied as training data into predictive model 902 as set forth in reference to Fig. 19B, and predictive model 902 can learn a relationship between antenna patch patterns represented by the applied matrices, and the applied simulation data.

[0190] As seen, e.g., in Fig. 9 and 10 an antenna patch structure 1 10 can define an antenna array that comprises multiple patch antennas. As best seen in Fig. 10 the antenna patch structure 110 at antenna layer 108 can define a 4x4 array of patch antennas, the antenna patch structure 110 at antenna layer 112 can define a 2x2 array of patch antennas and the antenna patch structure 110 of antenna layer 116 can define a 2x1 array of patch antennas.

[0191] In a further aspect, generated antenna structure configurations generated at block 1802 and block 1902 can include differentiated vertical antenna structure variable conductive material elements in addition to the described antenna patch structure spatial region elements MmNn which can extend in the horizontal plane.

[0192] In reference to Fig. 17E-17G generated antenna structure configurations generated at block 1802 and 1804 can include vias spatial regions 250 which, like spatial regions MmNn, can have occupied (by conductive material) or unoccupied states. Occupied state via spatial regions 250 are depicted in Figs. 17E-17G (showing regions occupied by conductive material) but unoccupied states for the various depicted antenna structures 100 can be specified, e.g., as shown by vias spatial region 251 of Fig. 17E.

[0193] As shown in Figs. 17E-17G occupied state vias spatial regions 250 can in their respective configurations connect occupied spatial regions MmNn of respective antenna layers extending at differentiated elevations of an antenna structure 100. The positions of the various vias spatial regions 250 having occupied states can be differentiated between generated antenna structures 100 defining antenna structure configurations generated at block 1802 and block 1902. Embodiments herein recognize that occupied state vias spatial regions can in one aspect create current loops that increase a solution space for optimization as set forth in reference to Figs.18A-18C and selection of an optimized solution. Such current loops can also increase a range of simulation performance datasets which in a trained model machine learning training use case as set forth in Figs 19A-19B can increase the likelihood of a suitable configuration being identified by inferencing.

[0194] In one embodiment, the structures depicted in Figs. 17A-17G and throughout the remaining views can have relative dimensions as depicted in Figs. 17A-17G. In Fig. 17E, occupied state vias spatial regions 250 connecting the antenna patch structure 110 of antennalayer 116 and the antenna patch structure 110 of antenna layer 120 can include a height equal to the width and length of an MmNn spatial region defining an antenna patch structure 110 in reference to Figs. 17A-17C, and occupied state vias spatial regions 250 connecting the antenna patch structure 110 of antenna layer 112 and the antenna patch structure 110 of antenna layer 116 can include a height equal to three times the width and length of an MmNn spatial region defining an antenna patch structure 110 in reference to Figs. 17A-17C.

[0195] Various available tools, libraries, and / or services can be utilized for the implementation of trained predictive models herein trained by machine learning. For example, a machine learning service can provide access to libraries and executable code for the support of machine learning functions. A machine learning service can provide access to a set of REST APIs that can be called from any programming language and that permit the integration of predictive analytics into any application. Enabled REST APIs can provide, for example, retrieval of metadata for a given predictive model, deployment of models and management of deployed models, online deployment, scoring, batch deployment, stream deployment, monitoring, and retraining deployed models. According to one possible implementation, a machine learning service can provide access to a set of REST APIs that can be called from any programming language and that permit the integration of predictive analytics into any application. Enabled REST APIs can provide, for example, retrieval of metadata for a given predictive model, deployment of models and management of deployed models, online deployment, scoring, batch deployment, stream deployment, monitoring, and retraining deployed models. Trained predictive models herein can employ the use, for example, of artificial neural networks (ANNs), support vector machines (SVM), Bayesian networks, regression-based models, and / or other machine learning technologies.

[0196] Embodiments herein recognize that in each layer of antenna structure 100, the elements can be represented by MxN matrices, where the size of these matrices can vary between layers depending on the specific application. In a further aspect, these matrices can be transformed into Kxl matrices, i.e., vectors. Once converted, the vectors can be transformed into a single large vector for processing depending on a selected artificial intelligence process. For example, commercially available genetic algorithm tools, e g., MATLAB® can be adapted for processingcandidate solutions (called chromosomes) in vector form. Alternatively, one or more MxN matrix can be fed to an artificial intelligence process without transformation.

[0197] After it performs transformation to provide the combined vector, system 1000 can perform optimizing the parameters. In the first step, it can generate random values as the initial population, representing potential solutions. These random values are used to extract the required parameters for each input, corresponding to the matrix patterns in the different layers, using any EM simulation software (we used MATLAB Antenna toolbox module). From this population, the model calculates the first output based on predefined metrics (it can be s-parameter, gain, RCS, bandwidth, beamwidth, etc.), allowing us to evaluate the performance. System 1000 can compare these results with target values across different frequencies to guide the optimization process.

[0198] The defined target, or goal, can include several objectives — such as maximizing gain, minimizing certain parameters, or optimizing radar cross-section (RCS). To evaluate the model’s performance, we calculate a Figure of Merit (FOM) for each parameter using an appropriate error calculation method, such as the Mean Squared Error (MSE). These individual FOMs are then combined through a weighted summation, allowing us to prioritize certain objectives over others.

[0199] Once the total FOM is calculated, system 1000 can check whether an antenna structure configuration satisfies specified performance criteria. If it does, the process is complete. If not, system 1000 can continue exploring the solution space, iterating through the optimization process, e.g., until either the desired FOM is achieved or a point is reached where further iterations no longer yield improvements.

[0200] Embodiments herein recognize that the advancement of 5G and 6G technologies has significantly enhanced wireless communication, providing faster data speeds and more reliable network connections. Embodiments herein recognize that the implementation of next generations of wireless communication systems can involve incorporating new frequency bands. Embodiments herein recognize that an antenna structure can benefit from the deployment of extra antenna layers as set forth herein capable of operating within such new frequency bands. However, embodiments herein recognize that to ensure that systems for these newer generationsare cost-effective, the addition of antenna layers should not result in an increase in the panel size (aperture size) used by older generations. Embodiments herein also recognize that antenna structure complexity increases as new antenna layers capable of operating and new frequency bands are added. Embodiments herein include features to accommodate antenna layers operating at different bands within limited space. Embodiments herein can ensure optimal performance, characterized by efficient radiation patterns, dual polarization support, and proper return loss over all their bands.

[0201] Given the current industry requirements, embodiments herein recognize that a triple-band antenna system (supporting LB, MB, HB) is capable of meeting the demands of 5G communication systems. Embodiments herein recognize that to achieve a triple-band system, designers typically employ three sets of antenna arrays, each specialized for a distinct frequency band, arranged in various configurations. Embodiments herein recognize that given the constrained space on the panel, a challenge in designing such antenna system is to create a configuration that minimizes interference among the arrays operating at different bands and sharing the same aperture. Embodiments herein recognize that interference management can result in strategic placement and design to ensure optimal performance across all frequency bands. Embodiments herein recognize that due to space constraints, the arrays operating at different frequency bands are stacked vertically at varying distances from the ground plane. Embodiments herein recognize that in some designs, the low band (LB) array is positioned closest to the ground plane, with the mid-band (MB) and high band (HB) antennas situated progressively further away. Embodiments herein recognize that this vertical stacking helps maximize the use of available space, although it subjects the antenna system to increased interference between the bands. Embodiments herein recognize that in contrast, other designs reverse this order, positioning the high-band array nearest to the ground plane and the low-band antenna furthest away. Embodiments herein recognize that over the past few years, numerous configurations have been explored. Embodiments herein recognize that beyond the positioning of the antenna arrays, there is considerable variation in the design of the radiating elements of the antennas, which can significantly affect performance. Embodiments herein recognize that to tackle such a multi-faceted problem, engineers according to a currently employed methodology typically begin with simple structures and constraints — such as using dipolar radiating elements and positioning the antennas a quarter- wavelength from the ground plane drawing on bothanalytical and sometimes physical intuition. Embodiments herein recognize that following this initial setup, engineers according to a currently employed methodology typically employ optimization techniques to fine-tune the basic antenna design, aiming to meet all the complex requirements of the project. Embodiments herein recognize that in these approaches, engineers typically avoid using parasitic antenna elements because their impact is difficult to predict analytically. Embodiments herein recognize that designs that emerge from such methods often possess limited capabilities and fail to meet all requirements. Embodiments herein recognize that with currently employed design methodologies, it is particularly challenging to efficiently control both the matching and the radiation pattern simultaneously.

[0202] Embodiments herein can feature a multiband antenna structure that incorporates parasitic non-conventional radiating elements instead of conventional dipolar radiating elements. Embodiments herein can utilize parasitic elements as controllable aspects of the design process, enabling new generations of antenna systems.

[0203] Employing a physics-aware inverse design and use of artificial intelligence, embodiments herein can provide a new wideband antenna system that benefits from parasitic elements in the form of, e.g., pixelized conductive material, e.g., metallic surfaces. Embodiments herein can combine analytical understanding with the inverse design capabilities of optimization solver software, e.g., running a genetic algorithm, enabling comprehensive use of parasitic elements for engineering matching and radiation patterns.

[0204] Embodiments herein can possess capabilities that surpass those of current state-of-the-art technology in this area, opening up possibilities for enhanced performance and functionality in antenna structure designs. Embodiments herein can enable designers to easily achieve the required return loss and design antennas that can direct beams in any arbitrary targeted direction or shape the radiation pattern as needed.

[0205] In one embodiment, a single band antenna can include a ground plane 105 and an antenna patch structure 110 defining an active antenna layer located within a distance from ground plane 105. The space between the ground plane and the antenna patch defining an active antenna layer can be filled with a spacer 109 including one or more dielectric or magnetic layer.

[0206] The antenna patch structure 110 defining an active antenna layer can include one or more active antenna element wireline connected to a transmission line 101 and can include a pattern defined by a pixelized conductive material, e.g., metallic layer. Furthermore, a parasitic tuning antenna layer that is composed of pixelized conductive material antenna elements can be located within a distance above or below the antenna patch structure 110 defining an active antenna layer. The parasitic tuning antenna layer does not have any source attached to it and can be absent of any active antenna elements. However, through the coupling with the active antenna layer it provides a significant degree of freedom in designing an antenna structure so as to provide the desired antenna characteristics, e.g., return loss level, radiation pattern, bandwidth. In one aspect, a parasitic tuning antenna layer can provide functionality of multiple parallel plate capacitors to expand degrees of freedom and a solution space for exploration and / or generation of training data.

[0207] Embodiments herein include features for achieving directionality constraints of a targeted design. To make an antenna more directive, an antenna layer herein can include multiple subareas, each subarea defining an antenna, and multiple ones of the subareas defining an antenna array. In an active antenna layer, a subarea of an antenna patch can be repeated in a periodic fashion so that the antenna layer defines an antenna array. In another aspect, a parasitic tuning layer disposed in spaced stacked relation to the active layer can include a pattern of subareas having a pattern in common with a pattern of the active antenna.

[0208] Embodiments of antenna structures 100 herein can include several antenna arrays each covering one or more of the frequency spectrum, e.g., as shown in the embodiment of Figs. 6-8, and the embodiment of Fig. 9-11.

[0209] In the embodiment of Figs. 6-8, there is set forth an antenna system where there are three sets of antenna arrays, each designed for a specific design frequency band. Since the space in the antenna system is limited the arrays for each band can be located at a different distance from the ground plane.

[0210] In the embodiment of Figs. 6-8 each antenna layer includes an associated parasitic tuning layer in stacked relation thereto. In another embodiment, as shown in Figs. 9-11, an antenna layer 150 provided by a parasitic tuning antenna layer is disposed as a top (highest elevation)antenna layer of a multi-layer stacked antenna structure. The respective antenna layers of the described antenna structure including the parasitic antenna layer can include pixelized conductive material (e.g., metallic) patterns separated by one or more layer.

[0211] Embodiments of antenna structures herein can include phased array antennas designed for any desired number of frequency bands.

[0212] Embodiments herein can include processing circuitry running optimization solver software that takes the physical space available for an antenna structure, available materials to be used, and the characteristics for the antenna structure (such as bandwidth, return loss level, radiation pattern. . .) as the input from the user and outputs an optimized antenna structure design.

[0213] Embodiments herein recognize that according to a conventional approach for antenna structure design, antennas structures are designed based on the physical understanding of active antenna elements, e.g., the “standard dipole” as the starting point of the design. Embodiments herein recognize that due to difficulty in analytically modeling and characterizing the impact of parasitic elements, parasitic elements are avoided according to the conventional approach.

[0214] Embodiments herein recognize that the conventional approach for antenna structure design produces a limited solution space for a multidimensional problem that often includes multiple constraints. Due to the restricted solution space, embodiments herein recognize that the conventional approach has limited capacity to produce antenna structure designs that simultaneously satisfy multiple constraints.

[0215] Embodiments herein recognize that parasitic antenna elements can significantly enhance candidate solution spaces for an antenna structure design, permitting modeling of significant numbers of differentiated candidate solution antenna structures, and optimization solver selection of an optimized design, even where optimization encompasses multiple constraints.

[0216] Embodiments herein can provide a single band or multiband antenna structure that incorporates non-conventional radiating elements in combination with or in place of conventional dipolar radiating elements. Antenna structures herein can utilize parasitic antenna elements as controllable aspects of the design process, enabling new generations of antenna systems.

[0217] Embodiments herein can employ a physics aware inverse design based on an optimization solver, e.g., an evolutionary algorithm such as a genetic algorithm, or an adjoint variable method. Embodiments herein can employ a physics aware inverse design based on other artificial intelligence methods. Embodiments herein can provide wideband antenna system that benefit from parasitic antenna elements in the form of pixelized metallic surfaces.

[0218] Embodiments herein can combine analytical understanding with the inverse design capabilities of an optimization solver, enabling comprehensive use of parasitic antenna elements for matching and engineering radiation patterns. Antenna structures herein feature capabilities that surpass those of current state of the art technology. Methods herein enable designers to achieve target return loss. Methods herein enable designers to design antennas that can direct beams in any arbitrary direction or shape the radiation pattern as needed.EXAMPLE

[0219] A simulation was initiated with a single-layer setup, depicted in Fig. 16A. In this setup, the pixelized layer was positioned a quarter wavelength (X / 4) above the solid ground plane, aiming to achieve specific return loss and maximum radiation pattern characteristics typical of standard antenna designs. Some of these baselines, such as placing the antenna at a X / 4 distance from the ground plane, are consistent with many traditional antenna configurations that prioritize efficient radiation patterns and good impedance matching for minimal return loss. Introducing a second layer, including a pixelized pattern of parasitic antenna elements, positioned only a tenth of a wavelength (X / 10) from the first, significantly modifies the electromagnetic interaction within the structure.

[0220] The tight proximity between the first and second antenna layers (active antenna layer and the tuning antenna layer having parasitic antenna elements without active antenna elements) induces strong coupling between the pixelized elements across both layers. This coupling can alter the electromagnetic fields and current distribution within the antenna, providing additional degrees of freedom for manipulating the antenna’s performance. The simulation results shown in Figs. 16G-16H indicate that incorporating a parasitic antenna element enhances the maximum gain across the specified frequency band, while the return loss remains largely unchanged. The simulation results shown in Figs. 16G-16H indicate that incorporating a tuning antenna layerenhances the maximum gain across the specified frequency band, while the return loss remains largely unchanged.

[0221] [End of Example]

[0222] Embodiments herein can be used to shape the radiation pattern or achieve any arbitrary response, including adjustments to bandwidth, beamwidth, gain, side lobe level, and back lobe level. The results also indicate that additional layers can be engineered to significantly enhance return loss without negatively impacting the antenna’s radiation pattern. Consequently, this method maintains desirable performance characteristics and / or achieve specific parameters. Embodiments herein optimize performance metrics such as maximum gain or overall radiation pattern, without sacrificing return loss. It is particularly advantageous in scenarios where enhancing or varying the response is necessary but expanding the antenna’s size is impractical.

[0223] In one version, antenna structure 100 can be designed as a stacked antenna. This method involves placing multiple antenna elements over one another, enhancing key performance aspects such as bandwidth, gain, and frequency isolation. Embodiments herein recognize the following advantages of stacked antenna designs: Embodiments herein can use multiple layers to segregate different frequency bands, which results in compact and efficient antenna structures ideal for setups like base station arrays. This layering helps maintain high frequency isolation while still achieving a slim profile and substantial gains. Embodiments herein recognize that compact stacked configurations can also provide benefits in terms of reducing the overall size of antennas, proving beneficial for multi frequency operations, and enhancing bandwidth at particular resonant frequencies. Embodiments herein recognize that certain designs prioritize improvements in properties like dimensions, bandwidth and gain by incorporating features such as parasitic patches or various feeding methods to fine-tune performance. Embodiments herein recognize that stacked designs can be particularly advantageous in applications such as global positioning systems, aiming for high gain and superior performance through the strategic layering of different antenna patch structures. Embodiments herein recognize that overall, stacked antenna designs can offer a technique to boost antenna efficiency for diverse applications by meticulously arranging multiple elements into a condensed structure, thereby enhancing bandwidth, gain, and isolation — key factors in contemporary telecommunication systems.

[0224] Embodiments herein recognize that the ground plane in a stacked antenna design can significantly influence key performance characteristics such as bandwidth, gain, and radiation pattern. Embodiments herein recognize that ground planes can be designed to be small relative to the antenna size, which is beneficial for applications that require compact antenna configurations. Embodiments herein recognize that the design of the ground plane can also be tailored to mitigate multipath effects, particularly for antennas where size and weight are critical. Embodiments herein can include the use of flat conductive and impedance ground planes and the development of vertical stacked structures to enhance radiation pattern Embodiments herein recognize that in more complex designs, ground planes can be used to enhance specific antenna properties. Embodiments herein recognize, for instance, that incorporating a reflecting board connected to the ground plane through via holes can excite new resonant frequencies, thereby enhancing the antenna’s bandwidth. Embodiments herein recognize that advanced stacked designs can incorporate modified ground planes to achieve ultra-wideband performance or improve specific radiation characteristics. Such modifications can include embedding slots or other structures within the ground plane to affect the electromagnetic response and performance of the antenna.

[0225] Embodiments herein recognize that a parasitic antenna element can be defined by a component that is not directly connected to the feedline (the primary power source) but influences the antenna’s radiation pattern through electromagnetic coupling. Embodiments herein recognize that parasitic antenna elements can enhance various performance aspects of the antenna, such as return loss, bandwidth, gain, radiation pattern, polarization, and directivity. Embodiments herein recognize that parasitic antenna elements can improve, e.g., bandwidth, matching, gain, directionality, and loss quality factors, and can enhance overall antenna performance without directly connecting to the antenna’s active components. Embodiments herein recognize that adding parasitic antenna elements can significantly enhance total radiated power and isotropic sensitivity, which can benefit the maintaining strong signal quality in various operational environments. Embodiments herein recognize that parasitic antenna elements can be applied in designing ultra-wideband (UWB) antennas to achieve specific frequency notches, thereby avoiding interference with other systems. Embodiments herein recognize that parasitic antenna elements can modify the antenna’s impedance bandwidth and suppress undesired frequency bands, demonstrating versatility in tailoring antenna characteristics tospecific needs. Embodiments herein recognize that parasitic antenna elements in antenna arrays can enhance isolation between closely spaced antennas, which can be particularly beneficial in high-density configurations like base station arrays. Embodiments herein recognize that parasitic antenna elements can help manage interference and improves the clarity and quality of the received signal.

[0226] Embodiments herein can include disposing an antenna patch over a ground plane, the antenna patch separated from the ground plane by one or more dielectric, where the antenna patch includes an antenna patch pattern, and where the antenna patch pattern has been designed with use of a process that includes generating a plurality of candidate patch patterns, evaluating the plurality of candidate patch patterns, and selecting one of the plurality of candidate patch patterns as the antenna patch pattern in dependence on the evaluating.

[0227] Implementations herein may include one or more of the following features. The method where the evaluating includes employing an evolutionary algorithm. The evaluating includes employing a genetic algorithm. The generating includes applying one or more constraints in the generating the candidate patch patterns. The generating includes applying one or more constraints in the generating the candidate patch patterns, the one or more constraint including a constraint to confine the plurality of candidate patch patterns to pixelated candidate patch patterns. The generating includes automatically generating the plurality of antenna patch patterns. The evaluating includes applying a fitness function using a simulator to simulate performance of the plurality of candidate patch patterns. The evaluating includes employing a genetic algorithm, where the employing the genetic algorithms includes applying a fitness function, performing a mutation process, and performing a crossover process.

[0228] The antenna structure can also include a ground plane; an antenna patch spaced apart from the ground plane by one or more dielectric, where the antenna patch incudes an antenna patch pattern may include one or more active antenna element in wireline communication with a transmitter / receiver, and one or more parasitic antenna element.

[0229] Implementations may include one or more of the following features. The antenna structure where the antenna structure includes a tuning antenna patch stacked in close proximity with the antenna patch, the tuning antenna patch may include parasitic antenna elements.

[0230] The antenna structure can also include a ground plane; an antenna patch spaced apart from the ground plane by one or more dielectric; where the antenna structure includes a tuning antenna patch stacked in close proximity with the antenna patch, the tuning antenna patch may include of parasitic antenna elements.

[0231] An antenna structure 100 defining an antenna system is shown in Fig. 13. Antenna structure 100 can include ground plane layer 102 having feeder substrate 103 and ground plane 105. Ground plane 105 can be formed of conductive material, e.g., of metal, metal alloy or other conductive material such as a doped semiconductor. Feeder substrate 103 can have formed thereon feedline 104 in wireline communication with transmitter and / or receiver 200 (hereinafter transmitter / receiver 200). Ground plane 105 can have formed therein opening (aperture) 106 for accommodation of a conductive via that provides electrical communication between feedline 104 and more antenna elements defining an antenna layer.

[0232] In a further aspect of antenna structure 100 shown in Fig. 13, antenna structure 100 can include antenna layer 108 stacked over ground plane 105. Antenna layer 108 can include spacer 109 and antenna patch structure 110. Antenna patch structure 110 can be formed of conductive material, e.g., of metal, metal alloy or other conductor such as a doped semiconductor. Antenna patch structure 110 can be formed on spacer 109. Antenna patch structure 110 herein can have a patch pattern featuring certain characteristics as are set forth herein. Antenna structure 100 as shown in Fig. 13 can include a single antenna layer, i.e., antenna layer 108. and can be regarded as a single antenna layer antenna structure.

[0233] Another antenna structure 100 is shown in Fig. 14. In Fig. 14, antenna structure 100 includes components set forth in reference to antenna structure 100 described in reference to Fig. 13, but can include an additional antenna layer, namely antenna layer 112 shown in Fig. 14. Antenna layer 112 can be stacked upon antenna layer 108. Antenna layer 112 can include features according to antenna layer 108. Specifically, antenna layer 112 can include spacer 109 and antenna patch structure 110. In one aspect, antenna patch structure 110 of antenna layer 112 can be configured differently than antenna patch structure 110 of antenna layer 108.

[0234] Fig. 15 shows antenna structure 100 having features in common with antenna structure100 of Fig. 13 and Fig. 14 expanded to include N antenna layers. In reference to Fig. 15, antennastructure 100 can include antenna layer 116 stacked over antenna layer 112 and zero or more additional antenna layers. Antenna layer 116 can define an Nth antenna layer of antenna structure 100 which includes first antenna layer 108 and second antenna layer 112. Nth antenna layer 116, like first antenna layer 108 and second antenna layer 112, can include spacer 109 and antenna patch structure 110. Antenna patch structure 110 of antenna layer 116 can be configured differently from antenna patch structure 110 of antenna layer 112 and antenna patch structure 110 of antenna layer 108.

[0235] In one aspect, the various antenna layers of antenna structure 110 can be configured to include differentiated design frequency bands. In one embodiment, first antenna layer can have a low band (LB) design frequency band. Second antenna layer 112 can have a mid-band (MB) design frequency band and Nth antenna layer 116 can include a high band (HB) design frequency band.

[0236] In general, antenna layers having lower design frequency bands can include relatively larger active antenna elements and antenna layers having higher design frequencies can include relatively smaller active antenna elements. In one aspect, the antenna patch structure 110 of the respective antenna layers 108, 112, and 116 can be positioned approximately a distance of one quarter of a wavelength from ground plane 105, where the wavelength is the wavelength of the design frequency.

[0237] In Fig. 4, there is shown an antenna layer comprising patch 110 having a certain configuration. Patch 110 can include active antenna elements 1102 wireline connected to feedline 104 and parasitic antenna elements 1104 not wireline connected to feedline 104.

[0238] In Fig. 5, there is shown an antenna layer comprising a patch 110 consisting of parasitic antenna elements 1104. The structure of Fig. 5 can be stacked in close proximity to another antenna layer.

[0239] Antenna layers herein can include active antenna layers and / or tuning antenna layers. Active antenna layers can include one or more active antenna element and can include zero or more parasitic antenna elements. Tuning antenna layers can be absent of any active antenna element. The antenna layer of Fig. 4 depicts an example of an active antenna layer. The antennalayer of Fig. 5 depicts an example of a tuning antenna layer. Antenna patch structures 110 herein can include active patch structures and / or tuning patch structures. An active patch structure herein can include at least one active antenna element and zero or more parasitic antenna elements. A tuning patch structure herein can be absent of any active antenna elements and can include one or more parasitic antenna elements.

[0240] In embodiments herein, antenna patch structure 110 throughout the views can be formed as a layer on spacer 109 which spacer 109 can include, e.g., a dielectric spacer and / or magnetic spacer. Antenna patch structure 110 (patch) throughout the views can feature a low profile and a thickness that is about constant throughout its distribution when formed on spacer 109. Antenna patch structure 110 throughout the views can be planar in construction. In embodiments set forth herein, spacer 109 can be defined partially, entirely, or substantially entirely, by a dielectric material capable of mechanically supporting an antenna patch structure 110. In embodiments set forth herein, spacer 109 can be defined partially, entirely, or substantially entirely, by an air dielectric. Spacer 109 in one embodiment can include first and second layers. The first layer can be a rigid, e.g., dielectric layer and can be configured to mechanically support antenna patch structure 110. The second layer can be provided by air to define an air dielectric layer. Antenna layers 108, 112, 116 depicted in Figs. 1-3 are shown as being defined by spaced antenna patch structures 110, wherein respective ones of the antenna patch structures are planar in construction and extend at a certain elevation. The spaced antenna patch structures 1 10 can include active antenna patch structures and / or tuning antenna patch structures.

[0241] Fig. 19C is an illustration of an example ANN architecture for trained predictive models herein. One element of ANNs can be the structure of the information processing system, which can include a large number of highly interconnected processing elements (called “neurons”) working in parallel to solve specific problems. ANNs can be trained using a set of training data, with learning that involves adjustments to weights that exist between the neurons. Referring now to Fig. 10, a generalized diagram of a neural network can be shown. Although a specific structure of an ANN can be shown, having three layers and a set number of fully connected neurons, this can be intended solely for the purpose of illustration. In practice, the present embodiments can take any appropriate form, including any number of layers and any pattern or patterns of connections therebetween.

[0242] ANNs can demonstrate an ability to derive meaning from complicated or imprecise data and can be used to extract patterns and detect trends that can be too complex to be detected by humans or other computer-based systems. The structure of a neural network can generally have input neurons 8302 that can provide information to one or more “hidden” neurons 8304. Weighted connections 8308 between the input neurons 8302 and hidden neurons 8304 can be weighted, and these weighted inputs can be processed by the hidden neurons 8304 according to some function in the hidden neurons 8304. There can be any number of layers of hidden neurons 8304, and neurons that perform different functions. There can also exist different neural network structures, such as a convolutional neural network, a maxout network, etc., which can vary according to the structure and function of the hidden layers, as well as the pattern of weights between the layers. The individual layers can perform particular functions and can include convolutional layers, pooling layers, fully connected layers, softmax layers, or any other appropriate type of neural network layer. Finally, a set of output neurons 8306 can accept and process weighted input from the last set of hidden neurons 8304.

[0243] This can represent a “feed-forward” computation, where information propagates from input neurons 8302 to the output neurons 8306. Upon completion of a feed-forward computation, the output can be compared to a desired output available from training data. The error relative to the training data can then be processed in “backpropagation” computation, where the hidden neurons 8304 and input neurons 8302 can receive information regarding the error propagating backward from the output neurons 8306. Once the backward error propagation has been completed, weight updates can be performed, with the weighted connections 8308 being updated to account for the received error. It can be noted that the three modes of operation — feed forward, backpropagation, and weight update — do not overlap with one another. This can represent just one variety of ANN computation, and any appropriate form of computation can be used instead.

[0244] To train an ANN, training data can be divided into a training set and a testing set. The training data can include pairs of an input and a known output, which can be referred to as outcome training data as referenced in connection with predictive models herein. During training, the inputs of the training set can be fed into the ANN using feed-forward propagation. After each input, the output of the ANN can be compared to the respective known output.Discrepancies between the output of the ANN and the known output that can be associated with that particular input can be used to generate an error value, which can be backpropagated through the ANN, after which the weight values of the ANN can be updated. This process can continue until the pairs in the training set can be exhausted.

[0245] After the training has been completed, the ANN can be tested against the testing set to ensure that the training has not resulted in overfitting. If the ANN can generalize to new inputs beyond those on which it was trained, it can be ready for use. If the ANN does not accurately reproduce the known outputs of the testing set, additional training data can be needed, or hyperparameters of the ANN can need to be adjusted.

[0246] ANNs can be implemented in software, hardware, or a combination of the two. For example, the weights of the weighted connections 8308 can be characterized as a weight value that can be stored in a computer memory, and the activation function of each neuron can be implemented by a computer processor. The weight value can store any appropriate data value, such as a real number, a binary value, or a value selected from a fixed number of possibilities, that can be multiplied against the relevant neuron outputs. Alternatively, the weights of the weighted connections 8308 can be implemented as resistive processing units (RPUs), generating a predictable current output when an input voltage can be applied in accordance with a settable resistance.

[0247] As set forth herein, system 1000 can apply constraints to specified variable components in dependence on user defined configuration data defined by a user using user interface 1202 entered in area 1204. In one example, system 1000 can constrain dimensions of individual spatial regions MmNn (pixels) as shown in Figs. 17A-17C in dependence on a selected design frequency band for an antenna layer. In another example, system 1000 can constrain an antenna patch structure 110 to define a certain number and layout of patch antennas (e.g., a 4x4 array of patch antennas for antenna layer 108 depicted in Fig. 10) depending on a design frequency band selected for a layer. In one aspect, system 1000 can be configured so that a user can configure (e.g., active, deactivate any constraint).

[0248] As set forth herein, at block 1802 (Fig. 18A) and / or block 1902 (Fig. 19A), system 1000 can generate a plurality of candidate antenna structure configurations, wherein each candidateconfiguration includes one or more antenna patch structures 110 defining a variable antenna patch structure that can be varied between the generated configurations. As further set forth herein, the antenna patch structure 110 of a given configuration can be defined as one or more matrix of spatial regions (e.g., an MxN grid of pixels), wherein each spatial region is associated with a binary state, either “occupied” (e.g., by conductive material) or “unoccupied.”

[0249] System 1000 can be configured so that states of certain spatial regions defining a variable antenna patch structure are pre-set to the “unoccupied” state, based on configuration data input by a user into user interface 1202, as shown in Fig. 12. For example, in connection with constraining a patch structure 110 to define a specified number and layout of discrete patch antennas (e.g., a 4x4 patch array for antenna layer 108 as depicted in Fig. 10), system 1000 can pre-set spatial regions corresponding to areas between defined patch antennas to the “unoccupied” state across all candidate antenna structure configurations generated at block 1802 and / or block 1902.

[0250] As further set forth herein, antenna layers of antenna structure 100 can be configured for differentiated design frequency bands. In one embodiment, first antenna layer 108 can be configured for a first design frequency band, second antenna layer 112 for a second design frequency band, and Nth antenna layer 150 for a third design frequency band, with each band being of any arbitrary designation. In one illustrative use case, the respective design frequency bands may be defined as a low band (LB), mid band (MB), and high band (HB). In another illustrative use case, at least one of the design frequency bands may correspond to a currently unallocated spectrum band. In each case, the design frequency band of an antenna layer can influence the size, number, spacing, and geometric pattern of the antenna elements supported by the associated antenna patch structure 110.

[0251] As shown throughout the views, and particularly in Figs. 7 and 8, the spatial layout of antenna elements within a patch structure 110 can be differentiated between layers in dependence on their design frequency bands. For example, antenna patch structure 110 of layer 108 can define a 4 - 4 array of spaced-apart patch antennas suitable for LB operation according to one embodiment; patch structure 110 of layer 116 can define a 2x2 array suitable for MB operation according to one embodiment; and patch structure 110 of layer 124 can define a 2x 1array suitable for HB operation according to one embodiment. The layout and geometry of these patch arrays — spatially distinct across the antenna layers — can be automatically enforced by system 1000 during generation of candidate configurations.

[0252] As further corroborated in Fig. 18B, each antenna patch structure 110 across different antenna layers can be digitally represented as an M^N matrix 1852, which can be flattened to form a vector representation 1854 input into an optimization solver. The position-specific binary values of these matrices reflect the spatial configuration of the antenna pattern and directly correlate with physical placement of conductive material on the fabricated antenna layer.

[0253] As further set forth herein, certain antenna geometries characterized by their top-view X-Y footprint can facilitate optimized performance in designated design frequency bands. For example, as illustrated in Figs. 20A-26D, distinct top-view geometries can be pre-associated with respective band designations. Figs. 20A-21E depict a thin cross-shaped geometry optimized for LB operation; Figs. 22A-22D depict a mid-thickness cross-shaped geometry optimized for MB operation; and Figs. 23A-26D depict a variety of rectangular, circular, diamond, and thick cross-shaped footprints optimized for HB operation. Performance metrics associated to antenna patch structure antennas fabricated with use of methods herein and incorporating geometries depicted in Figs. 20A-26D are summarized in Table B.

[0254] Table B

[0255] Accordingly, system 1000 can be configured such that, in response to a user-specified design frequency band selection for a given antenna layer (e.g., input into area 1205 of user interface 1202), system 1000 constrains the candidate antenna structure configurations to conform to one or more predefined geometric templates. Specifically, system 1000 can define for the selected layer a first spatial region subset (variable design area) 2051 and a second spatial region subset (invariable fixed-state area) 2052, as shown in Fig. 20E. Spatial regions in the first subset 2051 can be allowed to vary between “occupied” and “unoccupied” states across candidate configurations, enabling optimization, whereas regions in the second subset 2052 are pre-set and fixed in the “unoccupied” state throughout all candidate configurations.

[0256] Further, in some embodiments, system 1000 can be configured to define a subset of spatial regions within an antenna layer as permanently fixed in the "occupied" state across all candidate configurations generated at block 1802 and / or 1902. This fixed-state constraint can be applied to enforce design continuity and ensure physical or electrical connectivity to key components of the antenna system. For example, system 1000 can assign a predefined collection of spatial regions — such as a contiguous cluster of adjacent grid pixels — to remain in the "occupied" state throughout the design optimization process. In one use case, this collection of spatial regions can include: (i) a first group of spatial regions that define active antenna elements physically contacting conductive structures associated with transmission line 101, and (ii) a second group of spatial regions that define active antenna elements that are in wireline communication with transmitter / receiver 200, by virtue of being electrically connected — via conductive path defined by adjacent “occupied” state spatial regions — to the first group of spatial regions. By fixing such spatial regions in the “occupied” state, system 1000 ensures structuralintegrity and functional viability of antenna elements that require continuous conductive paths to transmission line 101 and transmitter / receiver 200, while allowing optimization to proceed in surrounding or unconstrained areas of the layout.

[0257] As shown in Fig. 20E, these subsets can be delimited by borders 2053, which define the boundary between a configurable variable region and a restricted invariable region of the antenna patch structure 110. In the variable region, spatial region states can be toggled to explore alternative patch patterns. In contrast, spatial regions in the invariable region are constrained from being assigned an “occupied” state and thus effectively excluded from the optimization solution space. This architectural constraint allows system 1000 to tailor the candidate antenna structure space in alignment with frequency-specific design heuristics or performance expectations, thereby reducing unnecessary computation and increasing likelihood of convergence toward a valid optimized solution.

[0258] As additionally evidenced by the example optimization flow of Fig. 18A and associated description in paragraphs, such structural constraints imposed on candidate patch patterns during generation at block 1802 are preserved through simulation, scoring via fitness function (e.g., Eq. 1), and crossover / mutation at block 1810. The geometric restriction strategy described herein ensures that these fixed “unoccupied” state regions are not violated during reproduction of new candidate populations.

[0259] Further, as set forth in the paragraph describing the configuration logic of system 1000 in connection with supervised learning (e.g., block 1906 of Fig. 19A and model 902 of Fig. 19B), the matrix representations including these pre-set invariable “unoccupied” state regions are incorporated into training data used to associate candidate structure configurations to expected performance outcomes. The machine learning model trained with these spatial constraints in place can more accurately infer configuration-to-performance mappings and produce more reliable inferencing results at block 1908.

[0260] In one aspect, the triggering of a particular antenna geometry, such as any of those shown in Figs. 20A through 26D, can be performed not only in response to a user selecting a design frequency band via area 1205 of user interface 1202 in Fig. 12, but also or alternatively in response to other types of user input. For example, system 1000 can be configured toautomatically select a preferred geometric footprint in response to inputs specifying a desired polarization characteristic (e.g., circular vs. linear), a beamwidth constraint, a spatial footprint constraint (e.g., maximum lateral dimension), a return loss target, or a mechanical integration constraint such as aperture shape compatibility with a housing component or mounting surface. In another example, a user could specify that the design antenna structure must co-reside with other modules or sensors in a shared physical volume, prompting the system to select a geometry known to minimize electromagnetic interference or mechanical overlap. Similarly, selection of a geometry may be based on environmental durability constraints (e.g., preferred geometries for thermal expansion or structural rigidity), or even regulatory constraints requiring beam shaping or null directions. The system can also adapt geometry selection based on prior iterations or learned optimization history, leveraging embedded logic or machine learning models to predict geometric layouts that historically yield higher fitness for comparable parameter sets.

[0261] As further corroborated in Fig. 18B, each antenna patch structure 110 across different antenna layers can be digitally represented as an M*N matrix 1852, which can be flattened to form a vector representation 1854 input into an optimization solver. The position-specific binary values of these matrices reflect the spatial configuration of the antenna pattern and directly correlate with physical placement of conductive material on the fabricated antenna layer.

[0262] As set forth in reference to Fig. 17D, embodiments herein recognize that individual spatial regions (e.g., pixels) defining an antenna patch structure 110 can include not only a horizontal position in the X-Y plane but also a vertical dimension along the Z-axis, characterized by a common height H which maps to an actual physical height at which a corresponding antenna patch structure can be fabricated. The logical assignment of a common height H across spatial regions enables a structured transformation between digital representation and physical fabrication. Embodiments herein recognize that this height attribute can be leveraged strategically within antenna structure configurations. For example, different layers of a stacked antenna structure 100 can include antenna patch structures 110 located at differentiated elevations, and the vertical height assigned to spatial regions can be varied among such layers. Embodiments herein recognize that height dimensions of individual spatial regions such as pixels (e.g., 2H, 3H, or nH) can be incorporated into generated antenna structure configurations as variable input parameters within optimization solver processing and / or artificial neural networktraining. In such cases, system 1000 can include vertical height values within the vectorized or matrix-based representation of antenna structure configurations that are subjected to evolutionary algorithm optimization or machine learning-based inference. Accordingly, strategic use of spatial region height can expand the solution space, enabling three-dimensional spatial optimization and allowing for enhanced control over interlayer coupling, resonance tuning, and overall electromagnetic performance. In one example, antenna structure configurations including vias spatial regions that connect antenna patch structures at differentiated elevations (e.g., as shown in Figs. 17E-17G) can include varied vertical heights (e.g., 2H, 3H, etc.) for the respective vias spatial regions depending on which layers they connect. Embodiments herein recognize that systematic inclusion of vertical geometry into design representations permits the design of high-performance multiband or broadband stacked antenna systems, where radiation behavior is influenced by both lateral and vertical characteristics of the antenna patch distribution.

[0263] In one embodiment, system 1000 can be configured such that selection of a particular antenna geometry 2042 from among the geometries illustrated in Figs. 20A-26D can trigger automatic selection of an associated balun geometry 2044. In this embodiment, each antenna geometry 2042, which can be selected based on user input data specifying a design frequency band or another user-defined performance goal, can be logically mapped to a balun geometry 2044 configured to support impedance transition and signal coupling requirements corresponding to that antenna geometry. It is seen in reference to Figs. 20A, 21 A, 22A, 23 A, 24A, 25 A, and 26A that each respective antenna geometry 2042 can be visually and functionally associated with a particular balun geometry 2044 that varies between the respective examples. These variations can include differentiated spatial layouts, trace topologies, and connection structures corresponding to design-specific electromagnetic and mechanical constraints.

[0264] Embodiments herein recognize that a balun (balanced-to-unbalanced transformer) can be configured to convert a balanced signal used by the antenna structure into an unbalanced signal used by the transmitter / receiver 200 or vice versa. In certain implementations, baluns can support impedance matching across broadband or narrowband systems, suppress common-mode currents, and preserve signal integrity across asymmetric feed structures. Embodiments herein recognize that inclusion of balun geometry as a co-optimized component within the designpipeline of system 1000 can improve the capacity to meet user-specified performance targets while reducing system-level electromagnetic interference.

[0265] System 1000 can be configured so that balun parameters are incorporated directly into the optimization process performed at block 1802 and / or block 1902. In one embodiment, system 1000 can represent balun geometries using pixelized layouts that define spatial regions in the X-Y plane analogous to antenna patch structure 110, wherein each spatial region of the balun geometry 2044 can be modeled with binary occupancy states (e.g., “occupied” for conductive material, “unoccupied” for void or dielectric region). These pixelized representations can further include a vertical dimension, e.g., height H, 2H, or nH, consistent with the three-dimensional spatial modeling applied to antenna layers. As such, system 1000 can generate candidate balun configurations by assigning both lateral position and vertical elevation to each spatial region, thereby enabling three-dimensional variation of balun architecture in conjunction with antenna structure design.

[0266] Embodiments herein recognize that integrating balun geometry 2044 into the configuration space of the optimization solver allows for coordinated electrical behavior between the feed network and antenna, improving return loss, gain, and overall system efficiency. In one implementation, system 1000 can perform joint optimization of antenna patch structure 110 and balun geometry 2044, such that candidate antenna structure configurations and associated balun layouts are co-evaluated using simulation performance parameter datasets recorded at block 1804. In another implementation, system 1000 can be configured to support reverse mapping — wherein a user selects or inputs a desired balun geometry (e.g., a preferred trace layout or chipbased structure), and system 1000 restricts or conditions the generation of antenna patch structure configurations to ensure compatibility with the selected balun.

[0267] Further, embodiments herein recognize that balun geometries, like antenna patch structures, can be adapted for full-wave or circuit-based simulation within system 1000. In one embodiment, simulation software integrated with processing circuitry 310 can evaluate impedance transformation behavior of a given balun layout in conjunction with a candidate antenna structure configuration, facilitating co-design of signal transition, impedance matching, and radiation performance. In another embodiment, system 1000 can maintain a library of balungeometry templates (e.g., stripline, coaxial, tapered microstrip, or transformer-type), each expressible using pixelized binary matrices with associated height attributes, and selection of a geometry 2042 can trigger automatic retrieval or generation of one or more balun geometries 2044 from the library.

[0268] Embodiments herein recognize that inclusion of balun geometry in the co-optimization and co-generation processes expands the design space explored by system 1000, allowing for improved performance without requiring post-hoc retrofitting of feed transition elements. In this way, system 1000 enables a cohesive, full-stack design methodology encompassing feed architecture, antenna structure, parasitic tuning elements, and balun interface — each jointly defined in a digital, pixelized, height-aware representation suitable for simulation, machine learning, and fabrication.

[0269] Embodiments herein collectively transform the computer from a passive simulation tool into a physics-aware, Al-powered, optimization-guided, fabrication-aligned design engine that autonomously generates high-performance electromagnetic structures. Through structured data representations, vertical design capabilities, constraint-based geometry control, and Al integration, the system described in the present application improves the functioning of the computer itself, reduces design latency, increases manufacturability, and extends the capabilities of antenna design into regions of complexity and performance that are inaccessible using conventional engineering methods.

[0270] Embodiments herein recognize that a physics-aware breakdown and representation of an antenna patch structure 110 as a structured collection of spatial regions — e.g., an M*N grid of pixel-like unit positions — constitutes a core mechanism by which system 1000 improves the functioning of a computer system for antenna design. This spatial region representation serves not only to express a continuous electromagnetic structure in a form suitable for machine processing, but also to encode relevant physical constraints directly into a format that a computer can operate on efficiently. By organizing the antenna patch structure into discrete spatial regions, each of which can be defined as having an “occupied” state (e.g., conductive material present) or an “unoccupied” state (e.g., conductive material absent), embodiments herein provide aphysics-aware digital representation that aligns with real-world electromagnetic behavior and fabrication constraints.

[0271] From the outset, this structured spatial representation economizes computing resources by converting an unstructured and potentially infinite electromagnetic design space into a finite, tractable configuration space aligned with physically meaningful parameters. The grid structure provides a platform through which the computer system can evaluate candidate antenna structure configurations with reduced memory overhead, more predictable computation time, and high compatibility with digital simulation tools and machine learning processes. Moreover, this representation enables rapid iteration and structured evaluation of design alternatives using matrix operations and binary vector manipulations that are inherently efficient in digital computing environments.

[0272] Embodiments herein further recognize that this grid-based format significantly reduces the degrees of freedom in the solution space relative to unconstrained or free-form modeling techniques. However, unlike arbitrary dimensionality reduction, the reduction in degrees of freedom is implemented in a physics-aware manner that preserves and emphasizes degrees of freedom that matter — namely, those that meaningfully impact antenna performance characteristics such as return loss, gain, pattern shape, or polarization. In doing so, embodiments herein ensure that optimization and machine learning algorithms focus computational effort only on those portions of the configuration space that are likely to produce useful or viable results. This avoids wasteful consumption of computing resources on configurations that are physically implausible, electromagnetically non-functional, or outside of real-world manufacturability.

[0273] The organization of spatial regions into defined grid structures allows system 1000 to enforce meaningful constraints and apply practical design heuristics in a systematic and automated fashion. For instance, as shown in Fig. 20E, system 1000 can define two types of areas within a spatial region grid: a first area 2051 comprising variable state regions that are permitted to toggle between “occupied” and “unoccupied” states during configuration generation, and a second area 2052 comprising fixed state regions that are pre-set to remain in the “unoccupied” state across all configurations. These two areas can be delineated by borders2053, providing the system with an explicit, enforceable geometry that constrains search and optimization processes to relevant spatial zones.

[0274] Embodiments herein recognize that a physics-aware inverse design methodology can significantly improve the functioning of computer systems used for antenna structure design and simulation. One key element of being physics-aware is the breakdown and representation of an antenna patch structure 110 as a spatial grid of discrete regions, e.g., pixels or cells arranged in an M*N matrix. Each region within the grid corresponds to a specific physical location on an antenna layer and can take on a binary state — e.g., “occupied” or “unoccupied” — to represent the presence or absence of conductive material. This representation allows for the underlying electromagnetic physics to be encoded in a format directly consumable by machine-based design tools, optimization solvers, and supervised learning models. As a result, this framework transforms real-world physical design parameters into a structured, finite, and computable space, which economizes computing resources in multiple ways.

[0275] The use of spatial grids to define antenna patch structures immediately confines the design search space to discrete, physically realizable patterns, enabling practical and scalable simulation of large candidate populations. By reducing the complexity of arbitrary geometric variation and constraining design variability to a regularized grid, system 1000 enables precise definition, manipulation, and tracking of candidate configurations. Optimization solvers and Al models need not handle unstructured geometric representations, which would otherwise require costly computation to encode, compare, and simulate. Instead, by representing designs as matrices of binary values (e.g., a K* 1 vector formed from an M*N grid), these models operate on simplified, indexed inputs that are directly traceable to real-world manufacturing outcomes.

[0276] Importantly, the spatial grid framework does more than just simplify representation — it enables a structured way to integrate physics-aware constraints that further reduce unnecessary computation while enhancing optimization relevance. For example, in one embodiment shown in Fig. 20E, spatial regions of a grid defining a candidate antenna patch structure 110 are segmented into a first area 2051 and a second area 2052. The first area 2051 comprises spatial regions designated as variable-state regions meaning their binary occupancy values can change during candidate generation. The second area 2052 comprises fixed-state regions, which are pre-set to remain in an “unoccupied” state across all candidate configurations. A border 2053 may logically or visually separate these two areas. This partitioning serves as a computational gating mechanism — allowing system 1000 to focus evaluation resources on the subset of spatial regions that contribute to the functional behavior of the antenna structure while ignoring irrelevant portions of the design space.

[0277] The restriction of a portion of the grid to a fixed, “unoccupied” state significantly reduces the degrees of freedom. But critically, embodiments herein recognize that this restriction is implemented in a physics-aware manner that actually increases the effectiveness of the design space exploration. Rather than exploring an unconstrained, high-dimensional design space that includes a vast number of non-contributing configurations, system 1000 explores a more compact and targeted space that is bounded by physical insight into which regions of the antenna contribute meaningfully to performance — based on factors such as electromagnetic coupling, current distribution, impedance matching, and radiation pattern control.

[0278] The result is a system in which the computational exploration of antenna designs is both constrained and expanded in a purposeful and practical way — a result that arises from the very use of spatial grid representation itself, and which can be further enhanced by the application of the fixed and variable region functionality illustrated in Fig. 20E. The grid-based modeling of antenna patch structure 110 as a set of discrete spatial regions (e.g., pixels arranged in an M*N matrix) provides an immediate and foundational advantage: it enables physical design variability to be captured in a structured, machine-readable format that dramatically reduces the complexity of the underlying optimization problem. This format facilitates practical processing of highdimensional design spaces by bounding them within discrete and physically relevant configurations, where every element of the grid directly maps to a spatial region in a real-world antenna layer. The very act of expressing antenna configurations using this grid-based framework limits the exploration space to configurations that are physically realizable, thereby reducing computational overhead and simulation burden.

[0279] Moreover, the efficiency, relevance, and utility of this grid-based representation can be enhanced through the use of constraints such as those shown in Fig. 20E. In Fig. 20E, the grid of spatial regions is partitioned into a first area 2051 comprising variable-state spatial regions thatcan toggle between “occupied” and “unoccupied” during candidate generation, and a second area 2052 comprising fixed-state spatial regions that are pre-set to remain “unoccupied” across all configurations. The border 2053 separating these areas provides a clean computational delineation that enables the optimization solver or learning model to bypass unnecessary processing of regions that do not contribute meaningfully to the targeted design goal. This approach significantly reduces the degrees of freedom — but does so in a physics-aware way that ensures the preserved degrees of freedom are focused on the most functionally impactful aspects of the antenna design. In effect, the grid representation, even before any constraints are applied, provides a substantial improvement in computational efficiency; and when constraints such as those of Fig. 20E are introduced, this improvement is further amplified.

[0280] By capturing and structuring spatial variability through a grid format, and optionally layering on physics-aware constraints that delimit variation to performance-relevant zones, system 1000 enables large-scale candidate generation, evaluation, and optimization to be performed with high efficiency, increased accuracy, and reduced hardware burden. The underlying spatial grid acts as both a representational simplification and a physics-aware filtering mechanism, making it a central element in the improved operation of computer systems and optimization processes as enabled by embodiments herein.

[0281] Embodiments herein improve the functioning of a computer system and other technology by providing a physics-aware, resource-efficient platform for digitally generating, evaluating, and optimizing electromagnetic antenna structures. This platform utilizes spatially structured, height-aware, and fabrication-compatible representations that are systematically processed via artificial intelligence models and electromagnetic simulation software. One key physics-aware construct enabling this improvement is the definition of a digital antenna structure modeling framework in which each antenna layer of a stacked antenna structure is represented as one or more pixelized spatial regions, such as an M*N grid of discrete, physically meaningful units. Each of these spatial regions can assume a binary state — “occupied” or “unoccupied” — corresponding to the presence or absence of conductive material. This format captures electromagnetic structure in a machine-consumable form that maps directly to real-world fabrication and physical behavior. By structuring antenna geometry in this way, system 1000enables scalable, accurate, and efficient simulation and optimization that minimizes extraneous processing and maximizes design relevance.

[0282] This structured spatial grid model enables substantial economization of computing resources. Each spatial region can be represented in system memory using binary matrices, and optionally enhanced with an associated height dimension H that specifies the vertical extent of deposited conductive material. This height-aware, physics-grounded modeling supports three- dimensional geometry definition and optimization within a simplified digital substrate. The resulting format allows computer systems to manipulate physical structure programmatically, without needing to engage in costly mesh generation or geometric parsing routines. This discrete model aligns cleanly with electromagnetic simulation engines and artificial intelligence workflows, enabling high-throughput evaluation using evolutionary algorithms or supervised machine learning, including but not limited to genetic algorithms and artificial neural networks.

[0283] Embodiments herein enable physics-aware generation of candidate antenna structure configurations at blocks 1802 and 1902, wherein each configuration can include one or more antenna patch structure 110 that varies between candidates. These variations may include spatial layout, vertical height, material occupancy, and interlayer connectivity. Critically, system 1000 supports application of spatial constraints that further optimize computational efficiency: specific spatial regions of a patch structure can be pre-set to a fixed “unoccupied” state in dependence on user input data received via user interface 1202. This selectively restricts variation to only those spatial zones that are relevant to performance, fabrication, or layout intent, reducing the total number of permutations the system must simulate. These constraints can be dynamically adjusted based on inputs such as design frequency band selection, physical geometry preferences, or target performance thresholds.

[0284] To this end, embodiments herein incorporate physics-aware constraint logic that links electromagnetic performance considerations directly to the computational design space. Rather than relying on blind geometric search, system 1000 employs heuristics derived from frequencydependent field behavior, impedance continuity requirements, and effective aperture constraints to precondition the solution space prior to simulation. For example, upon selection of a target design frequency band — such as via area 1205 of user interface 1202 — the system can applyspatial heuristics that constrain candidate configurations to a geometric footprint empirically or analytically known to exhibit favorable behavior within that band. As illustrated in Fig. 20E, this may include defining a first subset of spatial regions 2051 as variable design regions, where “occupied” and “unoccupied” states are permitted to vary, and a second subset 2052 as invariable exclusion regions that are fixed in the “unoccupied” state. This selective activation of the computational geometry — based on embedded physics heuristics — reduces the need for exhaustive evaluation of non-viable topologies, significantly decreasing simulation runtime and focusing resources on high-fidelity candidate configurations.

[0285] In further refinement, system 1000 can augment this physics-aware constraint logic by defining a third class of spatial regions that are permanently fixed in the “occupied” state across all candidate configurations. These pre-occupied spatial regions may correspond to areas critical for preserving electromagnetic continuity, structural grounding, or RF signal integrity, and are often derived from topology-aware field simulations or predefined connectivity rules. For example, in conjunction with a frequency band selection, system 1000 can automatically designate a collection of contiguous regions as always “occupied” to ensure that radiating elements remain physically and electrically linked to the conductive structures defining transmission line 101 and to transmitter / receiver 200. These regions may include spatial pixels that lie in direct contact with the feed network, or are wireline-connected to such regions, forming a constrained conductive backbone. By embedding such structural invariants into the configuration space, system 1000 further narrows the search domain to physically meaningful solutions while safeguarding against electrically invalid or non-functional geometries. This multi-tiered constraint strategy — spanning variable, fixed-unoccupied, and fixed-occupied regions — embodies a hierarchical, physics-aware design methodology that accelerates convergence and enhances performance predictability.

[0286] The spatial structuring approach itself — and not merely its constraints — serves as a core physics-aware mechanism for economizing computing resources. By framing physical design variability in the form of structured spatial grids, system 1000 captures relevant electromagnetic configuration possibilities in a format that is inherently discrete, computable, and aligned with physical causality. These grid-based representations serve not just as inputs for simulation, but as a foundation for fast, organized, and resource-conscious exploration. Constraints like those inFig. 20E further enhance this efficiency by limiting exploration to relevant regions. The result is a system in which exploration of antenna design space is both constrained and expanded in a purposeful way — constrained to exclude non-productive configurations, and expanded through high-resolution, physics-aware variation of the spatial regions that matter most to performance.

[0287] Embodiments herein also improve the functioning of the computer system by enabling full 3D design optimization through strategic modeling of vertical dimensionality. Each spatial region can include an associated height H, enabling the modeling of multilayer conductive features. Candidate designs generated at blocks 1802 and 1902 may include regions of height H, 2H, 3H, or higher multiples, which correspond to structures such as stacked elements, vertical parasitic features, or complex radiating structures. These vertical configurations can be evaluated and selected through optimization processes that understand and exploit near-field effects, vertical phasing, and resonant behavior across stacked layers. Embodiments herein recognize that such height-aware modeling is crucial to enabling triple-band or wideband antenna systems that fit within the confined aperture limits typical of real-world 5G and 6G applications.

[0288] System 1000 integrates this modeling framework with physics-based simulation engines such as MATLAB Antenna Toolbox, Ansys HFSS, CST Studio Suite, and others. Each candidate configuration can be automatically converted into simulation input by mapping spatial occupancy and height states to corresponding material properties and geometric definitions. The resulting simulation outputs — including S-parameters, impedance, gain, polarization, beamwidth, and radiation pattern — are recorded as structured datasets in memory 3102, and evaluated using fitness functions or applied as training labels for supervised learning. Embodiments herein support closed-loop optimization workflows in which electromagnetic results guide the generation of future candidate structures, achieving convergence toward desirable antenna designs with minimized computational waste.

[0289] Embodiments herein are further configured to autonomously evolve antenna configurations using physics-aware artificial intelligence methods such as genetic algorithms. These methods operate on binary vector encodings derived from the spatial grid model, applying crossover, mutation, and elitism across populations. Fitness is calculated not based on arbitrary or abstract heuristics, but using simulation-based metrics grounded in real-world physics, such asreturn loss, gain, and MSE error versus target behavior. The presence of domain-specific, physics-aware fitness functions ensures that computational effort is directed toward viable and fabricable solutions, eliminating simulation of unphysical or irrelevant configurations. Optimization pipelines converge faster, require fewer iterations, and result in output geometries that can be physically implemented.

[0290] Embodiments herein further improve the integration between simulation systems and downstream fabrication technologies. The structured spatial format used to represent antenna configurations is inherently compatible with real-world manufacturing, including additive manufacturing (e.g., 3D printing), subtractive lithography, and PCB fabrication. Height values, material types, and layout boundaries can be directly extracted from digital configuration data and translated to fabrication parameters. No additional data conversion or manual translation is required. The system's ability to retain fabrication-aligned formats throughout the digital design process results in higher reliability, faster time-to-deployment, and elimination of errors due to model mismatches.

[0291] Embodiments herein also enhance the usability and interactivity of the computer system by enabling intuitive input and configuration control via user interface 1202. Users can input not only design frequency band and performance metrics, but also spatial constraints, array geometries, and other physical intent. In response, the system applies targeted, physics-aware constraints to the digital design space. For example, input of a 3.5 GHz band might automatically activate a mid-thickness cross geometry such as that shown in Figs. 22A-22D. Alternatively, the system can infer optimal spatial constraints based on user-defined objectives such as beam directivity, sidelobe suppression, RCS minimization, or multiband separation. These capabilities allow users to interact with the system in terms of desired physical effects, while the system dynamically enforces constraints that focus computation on the most promising and physically valid design zones.

[0292] In sum, embodiments herein improve the functioning of computer systems and simulation environments by grounding every stage of the design pipeline — from geometry definition to Al optimization to simulation output — in structured, physics-aware modeling that prioritizes computational efficiency, physical plausibility, and manufacturability. The integration of discretespatial region encoding, intelligent constraint logic, and full-stack simulation capability enables a system that not only accelerates design discovery but aligns every digital operation with real-world physical behavior.

[0293] Embodiments herein include functionality for transforming digitally represented antenna structure configurations into fabrication-ready output data that defines a physical antenna structure suitable for deployment. Upon completion of the design optimization process, system 1000 can output a finalized design antenna structure that includes all information necessary for manufacturing, such as pixel-level material placement, height values for vertical stacking, interlayer connectivity, and geometric layout of active and parasitic elements.

[0294] This transformation is performed by processing circuitry 310 executing physics-aware generation and refinement routines, including those described in connection with optimization block 1802 and supervised learning block 1902. The input data to such routines can include design frequency band, spatial constraints, target performance characteristics, and manufacturing preferences. The output defines a specific antenna patch structure configuration, expressed using digital structures that map directly to physical fabrication instructions. These digital structures include binary matrices and / or multi-dimensional vectors that identify spatial region positions (e g., M*N pixels), height dimensions (e.g., H, 2H, etc.), and material presence (e.g., occupied or unoccupied conductive states).

[0295] The resulting data can define the placement of conductive material, layer-to-layer vias, ground plane patterns, and other structural features, such that no additional design interpretation or conversion is necessary for fabrication. For example, the output can include data suitable for direct control of additive manufacturing systems such as 3D conductive inkjet printers, photolithography systems, or etching tools. Because the digital design output remains consistent with fabrication parameters throughout the design process, system 1000 enables seamless transition from digital optimization to physical instantiation.

[0296] In one embodiment, system 1000 performs generation of a plurality of candidate antenna patch structures 110, evaluates those structures against electromagnetic simulation metrics, and selects a final design structure for output. The selected configuration may be rendered in rendering area 1208 of user interface 1202 and further transmitted to fabrication device 3112.For instance, fabrication device 31 12 can be configured to fabricate the optimized patch pattern on a dielectric or magnetic substrate using the digitally specified grid layout and material presence data.

[0297] Embodiments herein include structured, physics-aware modeling techniques that enhance manufacturability by maintaining compatibility with fabrication constraints throughout the design flow. These techniques include spatial region modeling using grids of pixels, incorporation of vertical height attributes, and restriction of state variability within constrained regions such as those depicted in Fig. 20E. As described, spatial region states may be selectively fixed or variable depending on design inputs, such as design frequency band, to ensure that only fabrication-compatible and performance-relevant geometries are considered during candidate generation.

[0298] By maintaining a fabrication-oriented representation of candidate structures throughout the optimization process, embodiments herein ensure that the final output corresponds to a defined, real-world physical geometry. The transformation from design inputs and physics-aware simulations to a manufacturable layout enables production of antenna systems with reduced design-to-fabrication latency, improved repeatability, and minimized post-processing.

[0299] In this way, system 1000 provides a digital-to-physical transformation pipeline in which high-level design requirements are computationally mapped to antenna structures that are fully defined in physical terms, including lateral dimensions, vertical layering, and electromagnetic behavior. The resulting structure can be output as a fabrication dataset without requiring downstream conversion, thus aligning the design system closely with practical implementation workflows.

[0300] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprise” (and any form of comprise, such as “comprises” and “comprising”), “have” (and any form of have, such as “has” and “having”), “include” (and any form of include, such as “includes” and “including”), “contain” (and any form contain, such as “contains” and “containing”), and any other grammatical variantthereof, are open ended linking verbs. As a result, a method or article that “comprises”, “has”, “includes” or “contains” one or more steps or elements possesses those one or more steps or elements but is not limited to possessing only those one or more steps or elements. Likewise, a step of a method or an element of an article that “comprises”, “has”, “includes” or “contains” one or more features possesses those one or more features but is not limited to possessing only those one or more features.

[0301] Terms like “obtainable” or “definable” and “obtained” or “defined” are used interchangeably. This, for example, means that, unless the context clearly dictates otherwise, the term “obtained” does not mean to indicate that, for example, an embodiment must be obtained by, for example, the sequence of steps following the term “obtained” though such a limited understanding is always included by the terms “obtained” or “defined” as a preferred embodiment.

[0302] It should be appreciated that all combinations of the foregoing concepts and additional concepts discussed in greater detail below (provided such concepts are not mutually inconsistent) are contemplated as being part of the subject matter disclosed herein. In particular, all combinations of claims subject matter appearing at the end of this disclosure are contemplated as being part of the subject matter disclosed herein. It should also be appreciated that terminology explicitly employed herein that also may appear in any disclosure incorporated by reference should be accorded a meaning most consistent with the particular concepts disclosed herein.

[0303] This written description uses examples to disclose the subject matter, and also to enable any person skilled in the art to practice the subject matter, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the subject matter is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

[0304] It is to be understood that the above description is intended to be illustrative, and not restrictive. For example, the above-described examples (and / or aspects thereof) may be used incombination with each other. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the various examples without departing from their scope. While the dimensions and types of materials described herein are intended to define the parameters of the various examples, they are by no means limiting and are merely exemplary. Many other examples will be apparent to those of skill in the art upon reviewing the above description. The scope of the various examples should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. In the appended claims, the terms “including” and “in which” are used as the plain English equivalents of the respective terms “comprising” and “wherein.” Moreover, in the following claims, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements on their objects. Forms of term “based on” herein encompass relationships where an element is partially based on as well as relationships where an element is entirely based on. Forms of the term “defined” encompass relationships where an element is partially defined as well as relationships where an element is entirely defined. Further, the limitations of the following claims are not written in means plus function format and are not intended to be interpreted based on 35 U.S.C. § 112(f) unless and until such claim limitations expressly use the phrase “means for” followed by a statement of function void of further structure. It is to be understood that not necessarily all such objects or advantages described above may be achieved in accordance with any particular example. Thus, for example, those skilled in the art will recognize that the systems and techniques described herein may be embodied or carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.

[0305] The terms “substantially”, “approximately”, “about”, “relatively”, or other such similar terms that may be used throughout this disclosure, including the claims, are used to describe and account for small fluctuations, such as due to variations in processing, from a reference or parameter. Such small fluctuations include a zero fluctuation from the reference or parameter as well. For example, they can refer to less than or equal to ± 10%, such as less than or equal to ± 5%, such as less than or equal to ± 2%, such as less than or equal to ± 1%, such as less than or equal to ± 0.5%, such as less than or equal to ± 0.2%, such as less than or equal to ± 0.1%, such as less than or equal to ± 0.05%. If used herein, the terms “substantially”, “approximately”,“about”, “relatively,” or other such similar terms may also refer to no fluctuations, that is, ± 0%. It is contemplated that numerical values, as well as other values that are recited herein can be modified by the term “about”, whether expressly stated or inherently derived by the discussion of the present disclosure. Further, any description of a range herein can encompass all subranges.

[0306] The terms “connect,” “connected,” “contact” “coupled” and / or the like are broadly defined herein to encompass a variety of divergent arrangements and assembly techniques. These arrangements and techniques include, but are not limited to (1) the direct joining of one component and another component with no intervening components therebetween (i.e., the components are in direct physical contact); and (2) the joining of one component and another component with one or more components therebetween, provided that the one component being “connected to” or “contacting” or “coupled to” the other component is somehow in operative communication (e.g., electrically, physically, optically, etc.) with the other component (notwithstanding the presence of one or more additional components therebetween). It is to be understood that some components that are in direct physical contact with one another may or may not be in electrical contact with one another. Moreover, two components that are electrically connected, electrically coupled, optically connected, optically coupled, may or may not be in direct physical contact, and one or more other components may be positioned therebetween.

[0307] While the subject matter has been described in detail in connection with only a limited number of examples, it should be readily understood that the subject matter is not limited to such disclosed examples. Rather, the subject matter can be modified to incorporate any number of variations, alterations, substitutions, or equivalent arrangements not heretofore described, but which are commensurate with the spirit and scope of the subject matter. Additionally, while various examples of the subject matter have been described, it is to be understood that aspects of the disclosure may include only some of the described examples. Also, while some examples are described as having a certain number of elements it will be understood that the subject matter can be practiced with less than or greater than the certain number of elements. Accordingly, the subject matter is not to be seen as limited by the foregoing description but is only limited by the scope of the appended claims.

[0308] Where one or more ranges are referred to throughout this specification, each range is intended to be a shorthand format for presenting information, where the range is understood to encompass each discrete point within the range as if the same were fully set forth herein.

[0309] While several aspects and embodiments of the present disclosure have been described and depicted herein, alternative aspects and embodiments may be affected by persons having ordinary skills in the art to accomplish the same objectives. Accordingly, this disclosure and the appended claims are intended to cover all such further and alternative aspects and embodiments as fall within the true spirit and scope of the present disclosure.

Claims

What is claimed is:

1. A method comprising: disposing an antenna patch pattern over a ground plane, the antenna patch pattern spaced apart from the ground plane, wherein the antenna patch pattern has been designed with use of a process that includes: generating a plurality of candidate patch patterns; evaluating the plurality of candidate patch patterns; and selecting one of the plurality of candidate patch patterns as the antenna patch pattern in dependence on the evaluating.

2. The method of claim 1, wherein the evaluating includes employing an optimization solver.

3. The method of claim 1, wherein the evaluating includes employing an evolutionary algorithm.

4. The method of claim 1, wherein the evaluating includes employing a genetic algorithm.

5. The method of claim 1, wherein the generating includes applying one or more constraint in the generating the candidate patch patterns.

6. The method of claim 1, wherein the generating includes applying one or more constraint in the generating the candidate patch patterns, the one or more constraint including a constraint to confine the plurality of candidate patch patterns to pixelated candidate patch patterns.

7. The method of claim 1, wherein the generating includes automatically generating the plurality of candidate patch patterns.

8. The method of claim 1, wherein the generating includes automatically generating the plurality of candidate patch patterns, and wherein the automatically generating includes randomly generating candidate patch patterns of the plurality of candidate patch patterns, anddependently generating patch patterns of the plurality of candidate patch patterns in dependence on prior generated patterns of the plurality of candidate patch patterns.

9. The method of claim 1, wherein the evaluating includes applying a fitness function using a simulator to simulate performance of the plurality of candidate patch patterns.

10. The method of claim 1, wherein the evaluating includes employing a genetic algorithm, wherein the employing the genetic algorithms includes applying a fitness function, performing a mutation process, and performing a crossover process.

11. An antenna structure comprising: a ground plane; and an antenna patch pattern spaced apart from the ground plane; wherein the antenna patch pattern includes one or more active antenna element in wireline communication with a transmitter / receiver and one or more parasitic antenna element.

12. The antenna structure of claim 11, wherein the antenna structure includes a tuning antenna patch pattern disposed in stacked relation with the antenna patch pattern, the tuning antenna patch pattern consisting of one or more parasitic antenna element.

13. An antenna structure comprising: a ground plane; an active antenna patch pattern spaced apart from the ground plane, the active antenna patch having at least one active antenna element; wherein the antenna structure includes a tuning antenna patch pattern disposed in stacked relation with the active antenna patch pattern, the tuning antenna patch pattern consisting of one or more parasitic antenna element.

14. The antenna structure of claim 13, wherein at least one active antenna element is wireline connected to a feedline of the antenna structure.

15. The antenna structure of claim 13, wherein the tuning antenna patch pattern is separated from the active antenna patch pattern by one or more dielectric layer.

16. The antenna structure of claim 13, wherein the tuning antenna patch pattern and the active antenna patch pattern are planar in construction.

17. An antenna structure comprising: a ground plane; and an antenna patch structure spaced apart from the ground plane; wherein the antenna patch structure includes an antenna patch pattern comprising: one or more active antenna element in wireline communication with a transmitter / receiver and one or more parasitic antenna element.

18. The antenna structure of claim 17, wherein the antenna structure includes a tuning antenna patch structure disposed in stacked relation to the antenna patch structure, the tuning antenna patch structure consisting of one or more parasitic antenna element.

19. An antenna structure comprising: a ground plane; an active antenna patch structure spaced apart from the ground plane, the active antenna patch structure having at least one active antenna element; wherein the antenna structure includes a tuning antenna patch structure disposed in stacked relation with the active antenna patch structure, the tuning antenna patch structure consisting of one or more parasitic antenna element.

20. The antenna structure of claim 19, wherein the at least one active antenna element is wireline connected to a transmitter / receiver.21 . The antenna structure of claim 19, wherein the tuning antenna patch structure is separated from the active antenna patch structure by dielectric material.

22. The antenna structure of claim 19, wherein the tuning antenna patch structure is separated from the active antenna patch structure by magnetic material.

23. The antenna structure of claim 19, wherein the tuning antenna patch structure and the active antenna patch structure are planar in construction.

24. The antenna structure of claim 19, wherein the tuning antenna patch structure is absent of any active antenna element.

25. The antenna structure of claim 19, wherein the active antenna patch structure defines a plurality of spaced apart patch antennas.

26. A method for retrofitting a previously deployed antenna structure comprising: generating a plurality antenna structure configurations, wherein respective ones of the plurality of antenna structure configurations include one or more component of the previously deployed antenna structure and one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations; simulating performance of respective ones of the plurality of antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the plurality of antenna structure configurations an antenna structure simulation dataset; and processing antenna structure simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations; outputting a design antenna structure in dependence on the processing; and adding a mechanical component fabricated in accordance with the design antennastructure to the previously deployed antenna structure.

27. The method of claim 26, wherein the simulating includes altering simulation input parameter values associated to respective ones of the plurality of antenna structure configurations in dependence on a state of spatial regions defining the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations;28. The method of claim 26, wherein the processing includes employing artificial intelligence (Al) processing.

29. The method of claim 26, wherein the processing includes employing an optimization solver.

30. The method of claim 26, wherein the processing includes employing a genetic algorithm optimization solver.

31. The method of claim 26, wherein the processing includes employing a machine learning model that has been trained by supervised machine learning.

32. The method of claim 26, wherein the processing includes employing a neural network machine learning model that has been trained by supervised machine learning.

33. The method of claim 26, wherein the one or more variable antenna patch structure of the plurality of antenna structure configurations is specified to extend in a plane.

34. The method of claim 26, wherein the one or more variable antenna patch structure of the plurality of antenna structure configurations is specified to be included in an antenna layer that extends in a horizontal plane.

35. The method of claim 26, wherein the adding the mechanical component fabricated in accordance with the design antenna structure to the previously deployed antenna structure includes adding an antenna layer to the previously deployed antenna structure the antenna layer featuring an antenna patch pattern fabricated in accordance with the design antenna structure.

36. The method of claim 26, wherein the adding the mechanical component fabricated inaccordance with the design antenna structure to the previously deployed antenna structure includes adding a parasitic tuning antenna layer to the previously deployed antenna structure that features an antenna patch pattern fabricated in accordance with the design antenna structure.

37. The method of claim 26, wherein the adding the mechanical component fabricated in accordance with the design antenna structure to the previously deployed antenna structure includes adding an antenna layer to the previously deployed antenna structure that features an antenna patch pattern fabricated in accordance with the design antenna structure, and wherein the antenna layer is accommodated within an aperture defined by X-Y plane dimensions of a ground plane of the previously deployed antenna structure.

38. A method comprising: generating a plurality of antenna structure configurations, wherein respective ones of the plurality of antenna structure configurations include one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations; simulating performance of respective ones of the plurality of antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the plurality of antenna structure configurations an antenna structure simulation dataset; processing antenna structure simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations; and outputting a design antenna structure in dependence on the processing.

39. The method of claim 38, wherein the simulating includes altering simulation input parameter values associated to respective ones of the plurality of antenna structure configurations in dependence on a state of spatial regions defining the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations.

40. The method of claim 38, wherein the processing includes applying an optimization solver.

41. The method of claim 38, wherein the processing includes applying a genetic algorithm of an optimization solver.

42. The method of claim 38, wherein the processing includes training a predictive model using simulation data of the antenna structure simulation data.

43. The method of claim 38, wherein the processing includes training a neural network using simulation data of the antenna structure simulation data.

44. The method of claim 38, wherein the one or more variable antenna patch structure of the plurality of antenna structure configurations is specified to extend in a plane.

45. The method of claim 38, wherein the one or more variable antenna patch structure of the plurality of antenna structure configurations is specified to be included in an antenna layer that extends in a horizontal plane.

46. The method of claim 38, wherein the producing, from the simulating, includes recording respective ones of the antenna structure simulation datasets into data storage in association with a matrix representation of the respective ones of the plurality of antenna structure configurations, wherein the matrix representation is defined by a pattern of binary (1 or 0) values, wherein the binary values specify occupied or unoccupied states of spatial regions defining the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations.

47. A method comprising: receiving, through a user interface, user defined structural attribute data specifying one or more structural attribute of an antenna structure and target performance data specifying one or more target performance characteristic of the antenna structure; generating, in dependence on the user defined structural attribute data, a plurality of antenna structure configurations, wherein respective ones of the plurality of antenna structure configurations include one or more variable antenna patch structure that is varied between theplurality of antenna structure configurations; simulating performance of respective ones of the plurality of antenna structure configurations; producing, from the simulating, multiple antenna structure simulation datasets, wherein, as a result of the producing, there is associated to respective ones of the plurality of antenna structure configurations an antenna structure simulation dataset; and outputting a design antenna structure in dependence on the target performance data specifying one or more target performance characteristic of the antenna structure antenna structure and on simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations.

48. The method of claim 47, wherein the simulating includes altering simulation input parameter values associated to respective ones of the plurality of antenna structure configurations in dependence on a state of spatial regions defining the one or more variable antenna patch structure that is varied between the plurality of antenna structure configurations.

49. The method of claim 47, wherein the outputting includes performing the outputting with use of an optimization solver, and wherein the method includes comparing, using a fitness function, the target performance data specifying one or more target performance characteristic of the antenna structure antenna structure to the simulation data of respective ones of the antenna structure datasets.

50. The method of claim 47, wherein the outputting includes training a neural network predictive model using simulation data of respective ones of the antenna structure simulation datasets associated to the respective ones of the plurality of antenna structure configurations, and inferencing the neural network predictive model using the target performance data specifying one or more target performance characteristic.

51. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data.

52. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints include constraints on dimensions of individual spatial regions of the one or more variable antenna patch structure.

53. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints include constraints on dimensions of individual spatial regions of the one or more variable antenna patch structure, wherein the dimensions of the individual spatial regions are constrained in dependence on a design frequency band selected by the user.

54. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints include constraining the one or more variable antenna patch structure to define a specified number and layout of patch antennas.

55. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints include constraining the one or more variable antenna patch structure to define a specified number and layout of patch antennas, wherein the specified layout comprises a predefined patch antenna array geometry selected in dependence on a design frequency band selected by the user.

56. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the user defined structural attribute data includes user input specifying activation or deactivation of one or more of the applied constraints.

57. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints restrict candidate configurations to geometries in which a defined subset of spatial regions are fixed in an unoccupied state during generation of the plurality of antenna structure configurations.

58. The method of claim 47, further comprising applying one or more constraints to the generating of the plurality of antenna structure configurations, wherein the one or more constraints are applied in dependence on the user defined structural attribute data, wherein the one or more constraints restrict candidate configurations to geometries in which a defined subset of spatial regions are fixed in an unoccupied state during generation of the plurality of antenna structure configurations, wherein the defined subset of spatial regions is selected in dependence on a footprint geometry associated with a selected design frequency band.

59. The method of claim 47, wherein each of the one or more variable antenna patch structures includes a plurality of spatial regions, each spatial region having a state that is either an occupied state or an unoccupied state, wherein the occupied state indicates presence of conductive material in a corresponding physical antenna patch region, and the unoccupied state indicates absence of conductive material in the corresponding region.

60. The method of claim 47, wherein each of the one or more variable antenna patch structures includes a plurality of spatial regions, each spatial region having a state that is either an occupied state or an unoccupied state, wherein the occupied state indicates presence of conductive material in a corresponding physical antenna patch region, and the unoccupied state indicates absence of conductive material in the corresponding region, wherein generating the plurality of antenna structure configurations includes generating, for each configuration, a different spatial distribution of occupied and unoccupied states across the plurality of spatial regions of the one or more variable antenna patch structures, such that each configuration defines a differentiated antenna patch pattern.

61. The method of claim 47, wherein each of the one or more variable antenna patch structures includes a plurality of spatial regions, each spatial region having a state that is eitheran occupied state or an unoccupied state, wherein the occupied state indicates presence of conductive material in a corresponding physical antenna patch region, and the unoccupied state indicates absence of conductive material in the corresponding region., wherein the generating includes applying a geometric constraint based on a design frequency band selected via the user interface, and further includes restricting a subset of spatial regions to a fixed unoccupied state and permitting variation of state only in a remaining subset of spatial regions, wherein the restricted and variable subsets are defined by the geometric constraint.

62. The method of claim 47, wherein each of the one or more variable antenna patch structures includes a plurality of spatial regions, each spatial region having a state that is either an occupied state or an unoccupied state, wherein the occupied state indicates presence of conductive material in a corresponding physical antenna patch region, and the unoccupied state indicates absence of conductive material in the corresponding region, wherein the spatial regions are arranged in an M*N grid, and the occupied and unoccupied states of the MxN grid are constrained such that only spatial regions within a first designated area are permitted to vary and all remaining spatial regions are fixed to the unoccupied state, in accordance with a frequency- aware geometry selected for simulation efficiency.

63. The method of claim 47, wherein the one or more variable antenna patch structures are digitally represented as binary matrices stored in memory, each binary matrix corresponding to an M^N grid of spatial regions and comprising matrix elements having binary values indicating occupied or unoccupied states of the spatial regions.

64. The method of claim 47, wherein the one or more variable antenna patch structures are digitally represented as binary matrices stored in memory, each binary matrix corresponding to an MxN grid of spatial regions and comprising matrix elements having binary values indicating occupied or unoccupied states of the spatial regions, wherein generating the plurality of antenna structure configurations includes generating a plurality of binary matrices each representing a candidate configuration, wherein the binary matrices include different combinations of binary values encoding different spatial patterns of occupied and unoccupied states across the grid.

65. The method of claim 47, wherein the one or more variable antenna patch structures are digitally represented as binary matrices stored in memory, each binary matrix corresponding toan M*N grid of spatial regions and comprising matrix elements having binary values indicating occupied or unoccupied states of the spatial regions, further comprising transforming each binary matrix into a vector representation to define a candidate solution chromosome suitable for input to an artificial intelligence model, wherein the vector comprises a flattened sequence of binary values corresponding to spatial region states.

66. The method of claim 47, wherein the one or more variable antenna patch structures are digitally represented as binary matrices stored in memory, each binary matrix corresponding to an M*N grid of spatial regions and comprising matrix elements having binary values indicating occupied or unoccupied states of the spatial regions, further comprising transforming each binary matrix into a vector representation to define a candidate solution chromosome suitable for input to an artificial intelligence model, wherein the vector comprises a flattened sequence of binary values corresponding to spatial region states, wherein the artificial intelligence model comprises an optimization solver or a supervised learning model trained on prior simulation data, and wherein the model uses the vector representations to evaluate candidate antenna structure configurations by computing or predicting performance metrics corresponding to the target performance data.

67. The method of claim 47, wherein the user defined structural attribute data includes a selected design frequency band for at least one antenna layer of the antenna structure, and wherein the generating includes selecting, in dependence on the selected design frequency band, a geometry template for a variable antenna patch structure of the at least one antenna layer.

68. The method of claim 47, wherein the generating includes defining a matrix of spatial regions for the variable antenna patch structure of the antenna structure, each spatial region having an occupied or unoccupied state, and wherein the generating includes varying the occupied or unoccupied states of a subset of the spatial regions across the plurality of antenna structure configurations.

69. The method of claim 47, wherein the generating includes constraining a first subset of the spatial regions of the variable antenna patch structure to remain in a fixed unoccupied state across the plurality of antenna structure configurations, and permitting a second subset of the spatial regions to vary in state across the plurality of antenna structure configurations.

70. The method of claim 47, wherein the generating includes generating a plurality of binary matrices representing the respective states of the spatial regions of the variable antenna patch structure, and wherein each binary matrix is used as a digital representation of a respective antenna structure configuration.

71. The method of claim 47, wherein each binary matrix representing the spatial regions of the variable antenna patch structure is converted into a vector representation used as input to an optimization solver or a neural network.

72. The method of claim 47, wherein the generating includes assigning a vertical height dimension to each spatial region of the variable antenna patch structure, and including the vertical height dimension as a parameter in the plurality of antenna structure configurations.

73. The method of claim 47, further comprising defining a second subset of spatial regions that are constrained to remain in an “occupied” state across all candidate antenna structure configurations, the second subset comprising spatial regions associated with maintaining electromagnetic continuity between an antenna structure and a signal feed network.

74. The method of claim 47, wherein defining the second subset of spatial regions that remain in the “occupied” state comprises: identifying, based on frequency-specific design heuristics, a group of spatial pixels forming a contiguous conductive path; assigning a first portion of the group as spatial regions in direct physical contact with conductive elements of a transmission line; assigning a second portion of the group as spatial regions in wireline communication with a transmitter or receiver via the first portion; and fixing both portions in the “occupied” state across all candidate antenna structure configurations to preserve electromagnetic continuity during design optimization.

75. The method of claim 47, wherein the user defined structural attribute data includes data describing a previously deployed antenna structure, and wherein generating the plurality of antenna structure configurations includes incorporating one or more structural components of the previously deployed antenna structure into each of the plurality of antenna structure configurations.

76. A method comprising:receiving, through a user interface, user defined data associated with an antenna structure; generating a plurality of candidate antenna structure configurations in dependence on the user-defined data; simulating performance of the candidate antenna structure configurations; producing simulation datasets corresponding to the simulated performance of the respective candidate antenna structure configurations; and outputting a selected antenna structure configuration in dependence on the user defined data and on the simulation datasets.

77. The method of claim 76, wherein the user defined data includes a selected design frequency band, and wherein generating the plurality of candidate antenna structure configurations includes selecting, based on the selected design frequency band, a geometry template defining a spatial region layout for at least one variable antenna patch structure.

78. The method of claim 76, wherein generating the plurality of candidate antenna structure configurations includes constraining a subset of spatial regions of a variable antenna patch structure to remain in a fixed unoccupied state across all candidate configurations, the subset selected in dependence on a geometry template.

79. The method of claim 76, wherein generating the plurality of candidate antenna structure configurations includes permitting a subset of spatial regions of a variable antenna patch structure to vary in state across the candidate configurations, the subset selected in dependence on a geometry template associated with the user defined data80. The method of claim 76, wherein each candidate antenna structure configuration is digitally represented as a binary matrix comprising elements that encode an occupied or unoccupied state of corresponding spatial regions, and wherein the binary matrix is used as a digital input during evaluation of the configuration.

81. The method of claim 76, wherein the user defined data includes a selected design frequency band, and wherein generating the plurality of candidate antenna structureconfigurations comprises: digitally representing each configuration as a binary matrix defining a spatial pattern of occupied and unoccupied states across an M*N grid of spatial regions of a variable antenna patch structure; constraining a first subset of the spatial regions to remain in a fixed unoccupied state based on a geometry template associated with the selected design frequency band; constraining a second subset of the spatial regions to remain in an occupied state to maintain electromagnetic continuity between the antenna structure and a signal feed network; permitting a third subset of the spatial regions to vary in state across the configurations; transforming each binary matrix into a vector representation suitable for input to an artificial intelligence model; using the artificial intelligence model to predict or compute performance characteristics of the candidate configurations; producing simulation datasets based on the predicted or computed performance characteristics; and outputting a selected antenna structure configuration in dependence on the predicted or computed performance characteristics and on the user defined data.

82. A method comprising: disposing an antenna patch pattern over a ground plane, the antenna patch pattern spaced apart from the ground plane, wherein the antenna patch pattern has been designed by a process comprising: automatically generating a plurality of pixelated candidate patch patterns, including randomly generating at least some candidate patch patterns and generating other candidate patch patterns in dependence on previously generated patterns; applying one or more constraints during the generating to restrict the candidate patch patterns to a specified design space; evaluating the candidate patch patterns using a genetic algorithm that includes applying a fitness function, performing a mutation process, and performing a crossover process, wherein the fitness function is applied using a simulator to simulate performance of the candidate patch patterns; andselecting the antenna patch pattern from among the candidate patch patterns in dependence on the evaluating.

83. An antenna structure comprising: a ground plane; an active antenna patch pattern spaced apart from the ground plane, the active antenna patch pattern comprising at least one active antenna element wireline connected to a feedline of the antenna structure; and a tuning antenna patch pattern consisting of one or more parasitic antenna elements, the tuning antenna patch pattern disposed in stacked relation above or below the active antenna patch pattern and separated therefrom by one or more dielectric layers, wherein the active antenna patch pattern and the tuning antenna patch pattern are each planar in construction.

84. An antenna structure comprising: a ground plane; an active antenna patch structure spaced apart from the ground plane, the active antenna patch structure comprising a plurality of spaced apart patch antennas including at least one active antenna element wireline connected to a transmitter or receiver; and a tuning antenna patch structure disposed in stacked relation with the active antenna patch structure, the tuning antenna patch structure consisting of one or more parasitic antenna elements and being absent of any active antenna element, wherein the tuning antenna patch structure is separated from the active antenna patch structure by a material comprising at least one of a dielectric material or a magnetic material, and wherein both the active antenna patch structure and the tuning antenna patch structure are planar in construction.

85. A method for retrofitting a previously deployed antenna structure, comprising: generating a plurality of candidate antenna structure configurations, wherein each candidate antenna structure configuration includes one or more component of the previously deployed antenna structure and a variable antenna patch structure that varies between the candidate configurations, the variable antenna patch structure being specified as an antenna layer extending in a horizontal plane; simulating performance of the candidate antenna structure configurations, includingaltering simulation input parameters in dependence on a state of spatial regions defining the variable antenna patch structure; producing simulation datasets corresponding to the simulated performance of the respective candidate antenna structure configurations; processing the simulation datasets using an artificial intelligence model comprising a supervised machine learning model or a genetic algorithm optimization solver; outputting a selected antenna structure configuration in dependence on the processing; and adding a mechanical component fabricated in accordance with the selected antenna structure configuration to the previously deployed antenna structure, the mechanical component comprising an antenna layer or parasitic tuning antenna layer that includes an antenna patch pattern and is accommodated within an aperture defined by X-Y plane dimensions of a ground plane of the previously deployed antenna structure.

86. A method comprising: receiving, through a user interface, user defined structural attribute data specifying one or more structural attributes of an antenna structure and target performance data specifying one or more performance characteristics of the antenna structure; generating, in dependence on the user defined structural attribute data, a plurality of antenna structure configurations, wherein each antenna structure configuration includes a variable antenna patch structure comprising a plurality of spatial regions each having a state that is either an occupied state indicating presence of conductive material or an unoccupied state indicating absence of conductive material; applying one or more constraints to the generation of the antenna structure configurations, the one or more constraints including:(1) constraining a first subset of the spatial regions to remain in a fixed unoccupied state based on a geometry template selected in dependence on a design frequency band;(2) constraining a second subset of the spatial regions to remain in an occupied state to preserve electromagnetic continuity between the antenna structure and a signal feed network; and(3) permitting a third subset of the spatial regions to vary in state across the plurality of antenna structure configurations; digitally representing each antenna structure configuration as a binary matrixcorresponding to the M*N grid of spatial regions and comprising binary values indicating occupied and unoccupied states; transforming each binary matrix into a vector representation for use as input to an artificial intelligence model, the artificial intelligence model comprising an optimization solver or a supervised learning model trained on simulation data; simulating performance of the antenna structure configurations and generating, based on the simulation, a plurality of antenna structure simulation datasets respectively associated with the plurality of antenna structure configurations; computing or predicting performance metrics of the antenna structure configurations using the artificial intelligence model and the vector representations; and outputting a selected design antenna structure in dependence on the computed or predicted performance metrics and on the target performance data.

87. A method for retrofitting a previously deployed antenna structure, comprising: receiving, through a user interface, user defined structural attribute data specifying one or more structural attributes of the previously deployed antenna structure and target performance data specifying one or more desired performance characteristics for a retrofitted configuration; generating, in dependence on the user defined structural attribute data, a plurality of candidate antenna structure configurations, wherein respective ones of the candidate configurations incorporate one or more structural components of the previously deployed antenna structure and include a variable antenna patch structure that is varied between the candidate configurations; simulating performance of the candidate antenna structure configurations; producing, from the simulating, a plurality of antenna structure simulation datasets, wherein each simulation dataset is associated with a respective candidate configuration; outputting a retrofit design antenna structure in dependence on the target performance data and the antenna structure simulation datasets; and adding a mechanical component fabricated in accordance with the retrofit design antenna structure to the previously deployed antenna structure.

88. A method comprising: receiving, through a user interface, user defined data associated with an antenna structure,the user defined data including a selected design frequency band; generating a plurality of candidate antenna structure configurations in dependence on the user defined data, wherein each candidate configuration includes a variable antenna patch structure comprising a plurality of spatial regions arranged in an M*N grid, each spatial region having a state that is either an occupied state indicating presence of conductive material or an unoccupied state indicating absence of conductive material; applying a geometry template associated with the selected design frequency band to constrain generation of the candidate configurations by:(1) fixing a first subset of the spatial regions in an unoccupied state across all configurations,(2) fixing a second subset of the spatial regions in an occupied state to maintain electromagnetic continuity with a signal feed network, and(3) permitting a third subset of the spatial regions to vary in state between candidate configurations; digitally representing each candidate configuration as a binary matrix comprising elements that encode the occupied or unoccupied states of the spatial regions, and transforming each binary matrix into a vector representation; simulating performance of the candidate configurations and producing simulation datasets corresponding to the simulated performance of respective candidate configurations; evaluating the candidate configurations using an artificial intelligence model that receives the vector representations as input and is configured to predict or compute performance characteristics based on prior simulation data; and outputting a selected antenna structure configuration in dependence on the predicted or computed performance characteristics and on the user defined data.

89. A method comprising: receiving, through a user interface, user defined data associated with an antenna structure; generating a plurality of candidate antenna structure configurations in dependence on the user defined data; and outputting a selected antenna structure configuration in dependence on the user defined data.

0. The method of claim 89, further comprising simulating performance of the candidate antenna structure configurations; producing simulation datasets corresponding to the simulated performance of the respective candidate antenna structure configurations; and wherein the outputting of the selected antenna structure configuration is further in dependence on the simulation datasets.

Images