CN111697343A - High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm - Google Patents
High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm Download PDFInfo
- Publication number
- CN111697343A CN111697343A CN202010564456.6A CN202010564456A CN111697343A CN 111697343 A CN111697343 A CN 111697343A CN 202010564456 A CN202010564456 A CN 202010564456A CN 111697343 A CN111697343 A CN 111697343A
- Authority
- CN
- China
- Prior art keywords
- branch
- neural network
- artificial neural
- knot
- long
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01Q—ANTENNAS, i.e. RADIO AERIALS
- H01Q15/00—Devices for reflection, refraction, diffraction or polarisation of waves radiated from an antenna, e.g. quasi-optical devices
- H01Q15/0006—Devices acting selectively as reflecting surface, as diffracting or as refracting device, e.g. frequency filtering or angular spatial filtering devices
- H01Q15/0086—Devices acting selectively as reflecting surface, as diffracting or as refracting device, e.g. frequency filtering or angular spatial filtering devices said selective devices having materials with a synthesized negative refractive index, e.g. metamaterials or left-handed materials
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Aerials With Secondary Devices (AREA)
Abstract
The invention discloses a high-gain electromagnetic super-surface unit for an artificial neural network fitting algorithm, wherein a medium substrate is provided with a metal layer, and the metal layer has the following structure: the tree branch structure is provided with a square frame, a first long branch knot and a second long branch knot which have the same structure are respectively arranged on two opposite sides in the square frame, and the first long branch knot and the second long branch knot are square and are symmetrically arranged; the other two opposite sides in the square frame are respectively provided with a first short branch knot and a second short branch knot which have the same structure, and the first short branch knot and the second short branch knot are square and are symmetrically arranged; the distance between the first long branch and the second long branch is shorter than the distance between the first short branch and the second short branch; several dielectric substrates with metal layers are stacked and each dielectric substrate with metal layers has a certain distance between them. The electromagnetic surface is fitted by an artificial neural network algorithm, so that the error in the phase-to-size conversion process is reduced, the gain of the antenna is improved, and the side lobe of the antenna is reduced.
Description
Technical Field
The invention relates to the field of electromagnetic super-surface structures, in particular to a high-gain electromagnetic super-surface unit for an artificial neural network fitting algorithm.
Background
The novel artificial electromagnetic material is formed by a series of sub-wavelength units with specific structures in a periodic or non-periodic mode, and the physical properties of the electromagnetic material are determined by different unit structures and different spatial arrangement modes. The artificial electromagnetic surface as a two-dimensional artificial electromagnetic material can realize the control of the wave beam by regulating the amplitude, the phase, the polarization mode or the propagation mode of the electromagnetic wave, and has the advantages of simple structure, less loss and the like. With the rapid development of electronic technology in recent years and the gradual commercial use of 5G technology, people have higher and higher requirements on antenna performance, and the proposal of the artificial electromagnetic surface enriches the types of antennas on the one hand and provides a new idea for breaking through the bottleneck of the traditional antenna performance on the other hand.
The directional diagram of feed source antenna can be changed on artifical electromagnetic surface, places the antenna and constitutes a set of complete antenna system in the position apart from artifical electromagnetic surface certain distance, and antenna system can control the electromagnetic wave that shines on artifical electromagnetic surface, realizes beam optimization, focus or even polarization control. The artificial electromagnetic surface has two main modes, one is a reflection type artificial electromagnetic surface, and the other is a transmission type artificial electromagnetic surface, and although the two artificial electromagnetic surfaces have similar structures, the two artificial electromagnetic surfaces have larger difference in principle and design difficulty. The artificial electromagnetic surface mainly serves as a phase-changing function, but only one of the artificial electromagnetic surface reflects electromagnetic waves to generate a specific waveform, and the other artificial electromagnetic surface transmits the electromagnetic waves to generate a required waveform, which have been widely studied in academia. The reflection-type artificial electromagnetic surface can be used as the planar deformation of a reflector antenna, the transmission-type artificial electromagnetic surface can be used as the planar deformation of a lens antenna, the artificial electromagnetic surfaces of the two forms have advantages, firstly, for the reflection-type artificial electromagnetic surface, only the phase-shifting characteristic of a unit needs to be considered when the unit is designed, because the bottom of the surface is a metal plate which can completely reflect electromagnetic waves, the design difficulty of the unit is greatly reduced, if the frequency of the unit is not matched with the operating frequency of an antenna system, the influence on the performance of the antenna is small, and even in the worst case, the reflected waves have a similar beam shape with the original feed antenna. But a disadvantage of reflective artificial electromagnetic surfaces is that antenna performance is affected by feed shielding. The transmission type artificial electromagnetic surface fundamentally solves the problem because the electromagnetic wave forms a beam through the unit, so that the problem of feed source shielding is fundamentally solved. However, the main problem of the transmission-type artificial electromagnetic surface is that the thickness of the unit itself is thicker than that of the reflection-type unit, and on the other hand, the influence of the transmission amplitude is additionally considered in the design, because even if the phase change of the unit meets the requirement, the function of the whole antenna system is also influenced if the transmission amplitude does not meet the requirement, and if the frequency of the unit does not match the operating frequency of the antenna system, the incident electromagnetic waves are totally reflected back, so that the electromagnetic surfaces cannot pass through, which is also the reason that the transmission-type artificial electromagnetic surface unit is more difficult to design than the reflection-type artificial electromagnetic surface unit.
The fitting method at the present stage is to extract phase and size change curve data of the units, perform fitting calculation based on a traditional fitting algorithm, calculate approximate fitting expressions of the phase and the size, calculate phase distribution of each position of the artificial electromagnetic surface according to a phase compensation theory, and further solve size distribution of each unit, but the larger error between the fitting expressions and a real phase size curve causes lower beam gain generated by the finally formed artificial electromagnetic surface, higher side lobe and unsatisfactory performance of the final antenna.
Disclosure of Invention
The invention aims to provide a high-gain electromagnetic super-surface unit for an artificial neural network fitting algorithm, which reduces the error in the phase-to-size conversion process, improves the gain of an antenna and reduces the side lobe of the antenna.
In order to solve the technical problems, the invention adopts the technical scheme that:
a high-gain electromagnetic super-surface unit for an artificial neural network fitting algorithm comprises a medium substrate, wherein a metal layer is arranged on the medium substrate, and the metal layer has a structure that: the tree branch structure is provided with a square frame, wherein a first long branch knot and a second long branch knot which have the same structure are respectively arranged on two opposite sides in the square frame, and the first long branch knot and the second long branch knot are square and are symmetrically arranged; the other two opposite sides in the square frame are respectively provided with a first short branch knot and a second short branch knot which have the same structure, and the first short branch knot and the second short branch knot are square and are symmetrically arranged; the distance between the first long branch and the second long branch is shorter than the distance between the first short branch and the second short branch; several dielectric substrates with metal layers are stacked and each dielectric substrate with metal layers has a certain distance between them.
Further, the area of the first long branch or the second long branch is larger than that of the first short branch or the second short branch.
Further, the length P of the outer edge of the square frame is 13mm, and the length L of the first short branch section and the second short branch section32.5mm, width W32.15mm, distance W between opposite sides of the square frame2Is in the range of 5mm to 13mm, the first long branch and the second long branch have a length of (W)2-W)/2mm, width L2.9 mm.
Furthermore, the dielectric substrate material adopts rogers4003, the relative dielectric constant is 3.55, and the loss tangent is 0.0027.
Compared with the prior art, the invention has the beneficial effects that:
1. the electromagnetic surface unit has a unit structure composed of symmetrical C-shaped slit structures, wherein the C-shaped slit structures are formed by branches loaded inwards by rectangular slits, the phase of the unit is from-167 degrees to-503 degrees at the central frequency of 12.5GHz, the transmission amplitude of the unit is close to 1 within the size change range, and the performance of the unit can meet the requirements of a transmission type artificial electromagnetic surface on the unit.
2. After a phase compensation value is calculated according to a phase compensation theory, an artificial neural network fitting algorithm is introduced to fit the phase value in the process of converting the unit phase value into a size value, then the corresponding size value is obtained according to the phase value, the fitting error of the artificial neural network algorithm is reduced by 88% compared with that of the traditional algorithm, the transmission type artificial electromagnetic surface subjected to artificial neural network fitting is improved by 2-4dB in gain compared with that of the traditional fitting algorithm, the side lobe is reduced by 1-3dB, and the beam width is reduced by 2-4 degrees.
3. The gain of the whole transmission type artificial electromagnetic surface reaches 24.74dBi at 12.5GHz, the beam width reaches 8.4 degrees, the aperture efficiency reaches 44.4 percent, and the gain bandwidth reaches 11.2 percent.
Drawings
FIG. 1 is a top view of a metal layer structure of an electromagnetic surface of the present invention.
FIG. 2 is a representation of the relative dimensions of a metal layer of an electromagnetic surface of the present invention.
FIG. 3 is a perspective view of an electromagnetic surface according to the present invention.
FIG. 4 is a side view correlation dimensional representation of an electromagnetic surface of the present invention.
FIG. 5 is a transmission type artificial electromagnetic surface unit with W2And simulation result graph of transmission phase and transmission amplitude
FIG. 6 is a schematic diagram of an artificial neural network fitting algorithm.
FIG. 7 is a comparison of simulated values and results of a conventional fitting algorithm.
Fig. 8 is a phase profile of the wavefront that needs to be compensated at 175.5mm from the surface at 12.5Ghz feed antenna.
FIG. 9 is a diagram of wavefront per cell at 12.5GHzW2Size distribution map of (a).
FIG. 10 is a pictorial view of an artificial electromagnetic surface.
FIG. 11 is an artificial electromagnetic surface test chart.
Fig. 12 shows the simulation and test results of the E-plane at 12.5GHz for an antenna system consisting of a transmission type artificial electromagnetic surface and a feed antenna.
Fig. 13 shows the simulation and test results of the H-plane at 12.5GHz for an antenna system consisting of a transmission-type artificial electromagnetic surface and a feed antenna.
FIG. 14 is a schematic diagram of the effect of different fitting algorithms on a histogram.
FIG. 15 is a graphical illustration of the effect of different fitting algorithms on gain.
In the figure: a square frame 1; a first long branch 2; a first short branch 3; a second long branch 4; a second short branch knot 5; a dielectric substrate 6; an air layer 7.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The invention relates to a high-gain electromagnetic super-surface unit for an artificial neural network fitting algorithm, which comprises a medium substrate 6, wherein a metal layer is arranged on the medium substrate 6, and the structure of the metal layer is as follows: the tree branch structure is provided with a square frame 1, a first long branch knot 2 and a second long branch knot 4 which have the same structure are respectively arranged on two opposite sides in the square frame 1, and the first long branch knot 2 and the second long branch knot 4 are both square and are symmetrically arranged; the other opposite sides in the square frame 1 are respectively provided with a first short branch knot 3 and a second short branch knot 5 which have the same structure, and the first short branch knot 3 and the second short branch knot 5 are both square and symmetrically arranged; the distance between the first long branch 2 and the second long branch 4 is shorter than the distance between the first short branch 3 and the second short branch 5; several dielectric substrates 6 with metal layers are stacked and a certain distance is reserved between every two dielectric substrates 6 with metal layers to form an artificial electromagnetic surface.
The center frequency of the artificial electromagnetic unit is 12.5GHz, the artificial electromagnetic unit is located in a Ku wave band, the period P of the square matrix unit is 13mm, the period length is about half of the wavelength of the whole free space, grating lobes of wave beams are avoided, and the unit structure is shown in fig. 1-4. The unit is composed of a symmetrical C-shaped gap structure, the C-shaped structure is formed by branches loaded inwards by rectangular gaps, the period length of the whole unit is P-13 mm, and the length of each branch is L32.5mm, width W32.15mm, the unit is mainly controlled by W2To achieve control of the phase, W2Ranges from 5mm to 13mm, and the specific structural parameters are shown in table 1. The dielectric substrate 6 mainly adopts the rogers4003 with low dielectric coefficient to reduce the dielectric loss of electromagnetic waves, the relative dielectric constant is 3.55, the loss tangent is 0.0027, the thickness of the whole unit is 11.032mm, and the wavelength is about half of the free space wavelength.
TABLE 1 Unit construction parameters (units: mm)
It can be seen from FIG. 5 that the phase of the cell goes from-167 to-503 at a frequency of 12.5GHz, and the transmission amplitude of the cell is at W2In the range of 5mm to 13mm, close to 1, the performance of the cell can meet the requirements of a transmission type artificial electromagnetic surface on the cell.
Basic principle of phase compensation: assuming a generation direction of
The required phase distribution of the two-dimensional electromagnetic surface unit is shown as follows:
in the formula: k is a radical of0Is the electromagnetic wavelength in free space, (X)i,Yj) Is where the unit is located. The phase required for irradiating the electromagnetic wave to the front surface is added to the phase required for generating the directional beam to obtain the phase required for compensation of each unit, as shown in the following formula:
φ(xi,yj)＝-k0dij+φT(xi,yj) (2)
in the formula: phi is aT(xi,yj) The phase of each transmission cell that needs to be compensated; dijThe distance between the horn antenna and the element, the phase to be compensated for each element can be obtained by combining equations (1) and (2):
after the units are determined, the phase and the size need to be converted according to the phase compensation required by each unit, so that the phase and the size need to be fitted firstly, the phase is converted into the corresponding size according to the phase compensation values of the units at different positions, and an artificial neural network is introduced to fit the curve in the process.
The artificial neural network is used as a network which is created in the last century and forms different networks according to different connection modes by abstracting a human brain neuron network in the information processing process. Neural networks, as a model of operation, are composed of a large number of interconnected nodes, each node representing a particular output function, called the excitation function. A connection between nodes represents a weighted value, also called a weight, of the signals passing through the connection. The artificial neural network has unique advantages in large data processing, so that the fitting processing is tried to be carried out on the conversion of the phase and the size through the artificial neural network, the error in the phase-to-size conversion process is reduced, the gain of the antenna is improved, and the side lobe of the antenna is reduced.
FIG. 6 is a schematic diagram of an artificial neural network fitting algorithm, where x ═ x (x)1,x2,......,xn) For the input signal, w ═ w1,w2,......,wn)TAs a weight of connecting neurons, mkFor a weighted sum, p is a threshold, f is an activation function, ykTo output a signal, the final output result is:
the activation function is the key of the artificial neural network, the quality of the performance of the activation function has great influence on the final effect of the whole network, so the activation function plays an important role in the establishment of the whole neural network, the S-shaped function is selected as the activation function in the design, and the formula of the activation function is as follows:
the S-shaped function can realize any nonlinear mapping from input to output, the classification is more accurate, and the applicable occasions are wider than other activation functions.
Fig. 7 is a comparison graph of simulation values and results of various fitting algorithms, and it can be seen from the graph that the difference between the effect of fitting based on the neural network and the simulation values is very small and can be almost ignored, and such low error greatly improves the phase compensation effect of the electromagnetic units, and is compared with other commonly used fitting algorithms.
The magnitude of the loss is used as an evaluation basis for judging the advantages and disadvantages of each fitting algorithm, the smaller the numerical value is, the lower the representative error is, the better the fitting effect is, and it can be seen from table 2 that the traditional fitting algorithm is compared with the artificial neural network fitting algorithm, the error of the artificial neural network is only 0.58, compared with the polynomial fitting algorithm with the best fitting effect in the traditional fitting algorithm, the fitting effect is improved by 88%, through the fitting of the artificial neural network algorithm, the conversion error between the phase and the size is greatly reduced, and the precision is improved.
TABLE 2 evaluation of the comparison of the merits of each fitting algorithm
After the fitting method is determined, phase compensation values of the units are calculated by placing the feed antenna at a certain distance from the artificial electromagnetic surface, finally the feed antenna is placed at a distance of 175.5mm from the surface, after the distance is determined, a specific phase value needing compensation of each unit is calculated based on the basic theory of phase compensation, the design is carried out in order to improve the aperture efficiency of the antenna as much as possible, so that the shape of the wavefront is selected to be circular, a phase distribution diagram of each unit needing compensation at a position of 175.5mm from the surface of the feed antenna is shown in figure 8, after the phase compensation values of the units are determined, the wavefront is based on the previously introduced fitting method of the artificial neural network, and according to figure 4, the wavefront is obtained by placing the feed antenna at a certain distance from the artificial electromagnetic surface2The size distribution of (a) is shown in fig. 9.
The caliber and the thickness of the whole transmission type artificial electromagnetic surface are pi × 100mm2×11.032mm, and an electrical dimension of pi × 4.2.2 lambda0 2×0.5λ0，λ0Is a free space wavelength with a center frequency of 12.5GHz and has a focal ratio of 0.87. The physical and test images of the artificial electromagnetic surface are shown in fig. 10 and 11. The simulation and test results of the E surface and the H surface of the antenna system consisting of the transmission type artificial electromagnetic surface and the feed source antenna at 12.5GHz are shown in fig. 12 and fig. 13, the gains are 24.74dBi and 24.59dBi respectively, the 3dB beam width is about 8.4 degrees and 8.42 degrees, and a relatively obvious directional pencil beam is formed. The aperture efficiency of the electromagnetic super-surface antenna system is 44.4%, and compared with a single horn antenna, the gain is improved by nearly 86% after the super-surface is loaded. This also shows that the feed antenna gain is significantly improved with the help of the transmission type artificial electromagnetic surface. As can be seen from the figure, the directional diagram goodness of fit of the test and simulation of the E surface is higher, the gain error of the H surface test is close to 0.3dBi compared with the simulation, and the error occurrence can be related to the process and the test environment of the electromagnetic surface processing. The antenna side lobe is kept below-20 dB in the range of-90 degrees to 90 degrees, and the relatively low side lobe level is maintained. The cross-polarization level of the antenna is 38dB and the cross-polarization level at the null point is 35 dB.
On the basis, the influence of different fitting algorithms on the antenna performance is compared, and for different fitting algorithms, the gain of the antenna is reduced and the antenna performance is influenced due to different phase compensation errors of different fitting effect introduction units. Fig. 14 shows the influence of different fitting algorithms on the antenna gain pattern, and from the perspective of the normalized pattern, the pattern of the conventional fitting algorithm generally has no good antenna performance of the electromagnetic surface after being fitted by an artificial neural network in terms of beam width and side lobe, and the side lobe is reduced by 1-3dB through simulation. The beam width is reduced by 2-4 deg.. Fig. 15 shows the final gain effect of different fitting algorithms, and it can be seen from the figure that the gains of polynomial fitting, fourier fitting and sinusoidal fitting are both around 23.8dBi, the worst antenna gain of linear fitting is only about 21dBi, and it is apparent from the figure that the gain of the electromagnetic surface after artificial neural network fitting is generally improved by approximately 2-4dB and by about 5% compared with the gain of the conventional antenna.
Claims (4)
1. The high-gain electromagnetic super-surface unit for the artificial neural network fitting algorithm is characterized by comprising a medium substrate (6), wherein a metal layer is arranged on the medium substrate (6), and the metal layer has a structure that: the tree branch structure is provided with a square frame (1), a first long branch section (2) and a second long branch section (4) which are identical in structure are respectively arranged on two opposite sides in the square frame (1), and the first long branch section (2) and the second long branch section (4) are both square and symmetrically arranged; the other two opposite sides in the square frame (1) are respectively provided with a first short branch knot (3) and a second short branch knot (5) which have the same structure, and the first short branch knot (3) and the second short branch knot (5) are both square and are symmetrically arranged; the distance between the first long branch (2) and the second long branch (4) is shorter than the distance between the first short branch (3) and the second short branch (5); a plurality of dielectric substrates (6) with metal layers are stacked, and a certain distance is reserved between every two dielectric substrates (6) with metal layers.
2. The high-gain electromagnetic super-surface unit for artificial neural network fitting algorithm according to claim 1, wherein the area of the first long branch (2) or the second long branch (4) is larger than the area of the first short branch (3) or the second short branch (5).
3. The high-gain electromagnetic super-surface unit for artificial neural network fitting algorithm according to claim 1, wherein the length P of the outer edge of the square frame (1) is 13mm, and the length L of the first stub (3) and the second stub (5) is L32.5mm, width W32.15mm, the distance W between the opposite sides of the square frame (1)2Is in the range of 5mm to 13mm, the first long branch (2) and the second long branch (4) having a length (W)2-W)/2mm, width L2.9 mm.
4. The high-gain electromagnetic super-surface unit for the artificial neural network fitting algorithm according to claim 1, wherein the dielectric substrate (6) is made of rogers4003, the relative dielectric constant is 3.55, and the loss tangent is 0.0027.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010564456.6A CN111697343A (en) | 2020-06-19 | 2020-06-19 | High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010564456.6A CN111697343A (en) | 2020-06-19 | 2020-06-19 | High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111697343A true CN111697343A (en) | 2020-09-22 |
Family
ID=72481981
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010564456.6A Pending CN111697343A (en) | 2020-06-19 | 2020-06-19 | High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111697343A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012126256A1 (en) * | 2011-03-18 | 2012-09-27 | 深圳光启高等理工研究院 | Impedance matching component and hybrid wave-absorbing material |
| CN102760965A (en) * | 2012-07-03 | 2012-10-31 | 深圳光启创新技术有限公司 | Large-angle wave-transmitting metamaterial, antenna housing thereof and antenna system |
| CN208608360U (en) * | 2018-05-08 | 2019-03-15 | 东南大学 | A kind of reflection and transmission integrated electromagnetic coding Meta Materials of multifunctional unit |
| CN209418771U (en) * | 2019-02-15 | 2019-09-20 | 南京邮电大学 | A kind of broadband electromagnetical Meta Materials wave absorbing device based on multilayered structure |
| CN110416738A (en) * | 2019-08-05 | 2019-11-05 | 深圳光启尖端技术有限责任公司 | A kind of suction wave transparent metamaterial structure and aircraft |
| CN110535486A (en) * | 2019-08-07 | 2019-12-03 | 东南大学 | The direct processing formula transceiver of radiofrequency signal based on super surface neural network |
-
2020
- 2020-06-19 CN CN202010564456.6A patent/CN111697343A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012126256A1 (en) * | 2011-03-18 | 2012-09-27 | 深圳光启高等理工研究院 | Impedance matching component and hybrid wave-absorbing material |
| CN102760965A (en) * | 2012-07-03 | 2012-10-31 | 深圳光启创新技术有限公司 | Large-angle wave-transmitting metamaterial, antenna housing thereof and antenna system |
| CN208608360U (en) * | 2018-05-08 | 2019-03-15 | 东南大学 | A kind of reflection and transmission integrated electromagnetic coding Meta Materials of multifunctional unit |
| CN209418771U (en) * | 2019-02-15 | 2019-09-20 | 南京邮电大学 | A kind of broadband electromagnetical Meta Materials wave absorbing device based on multilayered structure |
| CN110416738A (en) * | 2019-08-05 | 2019-11-05 | 深圳光启尖端技术有限责任公司 | A kind of suction wave transparent metamaterial structure and aircraft |
| CN110535486A (en) * | 2019-08-07 | 2019-12-03 | 东南大学 | The direct processing formula transceiver of radiofrequency signal based on super surface neural network |
Non-Patent Citations (1)
| Title |
|---|
| 崔铁军，刘峻峰: "《太赫兹信息超材料与超表面》", 《雷达学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110380222B (en) | Huygens super-surface unit, transmission array antenna and unit phase control method | |
| Feng et al. | Octave bandwidth transmitarrays with a flat gain | |
| Sang et al. | High-gain UWB Vivaldi antenna loaded with reconfigurable 3-D phase adjusting unit lens | |
| Karimipour et al. | Holographic-inspired multibeam reflectarray with linear polarization | |
| US20110025432A1 (en) | Phase element for introducing a phase shift pattern into an electromagnetic wave | |
| CN108183327B (en) | Antenna housing for expanding deflection angle of phase array antenna | |
| Karimipour et al. | Demonstration of broadband reflectarray using unit cells with spline-shaped geometry | |
| Derafshi et al. | A new high aperture efficiency transmitarray antenna based on Huygens metasurfaces | |
| Yang et al. | Geodesic half-maxwell fish-eye-lens antenna | |
| Wu et al. | Reflectarray antenna design with arbitrary incident and reflection beam angle | |
| Saka et al. | Pattern optimization of a reflector antenna with planar-array feeds and cluster feeds | |
| Salo et al. | Holograms for shaping radio-wave fields | |
| CN111697343A (en) | High-gain electromagnetic super-surface unit for artificial neural network fitting algorithm | |
| Artiga et al. | Holographic design of arbitrary multi-beam leaky-wave antennas | |
| CN103594791A (en) | Metamaterial plate, reflector antenna system and electromagnetic wave reflection regulating method | |
| CN115810892B (en) | Millimeter wave all-metal high-gain folding reflective array antenna | |
| Shabbir et al. | A compact single layer reflectarray antenna based on circular delay-lines for X-band applications | |
| Ramalingam et al. | Analysis and design of checkerboard leaky-wave antennas with low radar cross section | |
| CN112952396B (en) | Reflective array antenna of slotted square ring unit based on embedded loading concave arm | |
| Ataloglou et al. | Efficient aperture illumination and beamforming with Huygens’ metasurfaces exciting surface waves | |
| Elsharkawy et al. | Single-and double-beam reflectarrays for Ka band communication | |
| CA2712165A1 (en) | A phase element for introducing a phase shift pattern into an electromagnetic wave | |
| Nesil et al. | Analysis and design of X-band Reflectarray antenna using 3-D EM-based Artificial Neural Network model | |
| Oğuzer et al. | E‐polarized beam scattering by an open cylindrical PEC strip having an arbitrary “conical‐section” profile | |
| Afzal | Near-field phase transformation for radiation performance enhancement and beam steering of resonant cavity antennas |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200922 |
|
| RJ01 | Rejection of invention patent application after publication |