Measurement of a substance attribute using a metasurface

Abstract

According to implementations of the present disclosure, a solution for measuring a substance attribute using a metasurface is provided. According to the solution, a metasurface is designed to be located between an emitter and a measured object or between a receiver and the measured object. Wireless signals are emitted to the measured object and wireless signals from the measured object reflected or transmitted via a metasurface are received. A value of a substance attribute of the measured object is estimated by analyzing the received wireless signals. The metasurface is optimized for measurements of substance attributes. For example, the structure of the metasurface is determined by maximizing the sensitivity of the received wireless signals to a change in a substance attribute.

Description

MEASUREMENT OF A SUBSTANCE ATTRIBUTE USING A METASURFACE BACKGROUND

[0001] In some cases, it may be desirable to measure attributes of certain objects harmlessly and non-invasively, such as the concentration of a particular substance in an object. For example, in the medical field, continuous blood glucose monitoring is critical to diabetic patients because this helps patients understand their own condition and prevent hypoglycemia or hyperglycemic events. For another example, in environmental pollution and governance, it may be desirable to monitor the concentration of contaminants in a liquid, such as water, without disrupting the liquid. Therefore, there is a need for a non-invasive, user-friendly and low-cost substance attribute measurement technology. SUMMARY

[0002] According to an implementation of the present disclosure, there is provided a solution for substance attribute measurement using a meta surface. In this solution, the meta surface is attached directly or indirectly to a measured object, for example to the surface of a container that includes the measured object or to the human skin. A wireless probing signal is emitted by an emitter towards the measured object attached with a meta surface, and a response signal from the measured object reflected or transmitted via the metasurface is received by a receiver. A value of a substance attribute of the measured object, for example, the concentration of a target substance in the measured object, is estimated by analyzing the received response signal. The structure of the metasurface is optimized for substance attribute measurement. The structure of the metasurface may be determined by maximizing the sensitivity of a signal characteristic of the response signal to a change in the value of the substance attribute. The solution for substance attribute measurement according to the present disclosure is a non-invasive and user-friendly measurement technology. Furthermore, by optimizing the design of the metasurface for the substance attribute measurement, the substance attribute measurement with high accuracy can be implemented.

[0003] This section is provided to introduce a selection of an object in a simplified form, which will be further described below in the Detailed Description. This section is not intended to identify key or major features of the claimed subject matter, nor is it intended to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF DRAWINGS

[0004] FIG. 1 illustrates a schematic block diagram of an example measurement architecture according to some implementations of the present disclosure;

[0005] FIG. 2A illustrates a flowchart of an example optimization process for ametasurface according to some implementations of the present disclosure;

[0006] FIG. 2B illustrates a flowchart of an example measurement process using a metasurface according to some implementations of the present disclosure;

[0007] FIG. 3A illustrates a schematic diagram of an example measurement environment for liquid concentration aided by a metasurface according to some implementations of the present disclosure;

[0008] FIG. 3B illustrates a schematic diagram of an equivalent circuit corresponding to the measurement environment shown in FIG. 3A, according to some implementations of the present disclosure;

[0009] FIG. 4A illustrates a schematic diagram of an example structure of a metasurface unit constituting a metasurface according to some implementations of the present disclosure;

[0010] FIG. 4B illustrates an example of an equivalent LC circuit of a metasurface according to some implementations of the present disclosure;

[0011] FIG. 5A illustrates a schematic diagram of an equivalent circuit corresponding to a measurement environment according to some implementations of the present disclosure;

[0012] FIG. 5B illustrates a schematic diagram of an example process of optimizing a passive metasurface according to some implementations of the present disclosure;

[0013] FIG. 6 illustrates impedance of a pattern with different geometric parameters over a particular frequency range according to some implementations of the present disclosure;

[0014] FIG. 7A shows curves of amplitude versus frequency for response characteristics over different glucose concentrations according to some implementations of the present disclosure;

[0015] FIG. 7B shows curves of phase versus frequency for response characteristics over different glucose concentrations according to some implementations of the present disclosure;

[0016] FIG. 8 illustrates a schematic diagram of an example finite element model for a measurement environment according to some implementations of the present disclosure;

[0017] FIG. 9 shows a flowchart of a process of concentration measurement according to some implementations of the present disclosure;

[0018] FIG. 10 illustrates a flowchart of a process of designing a metasurface according to some implementations of the present disclosure; and

[0019] FIG. 11 illustrates a schematic block diagram of an electronic device capable of implementing a plurality of implementations of the present disclosure. DETAILED DESCRIPTION

[0020] The present disclosure will now be discussed with reference to several exampleimplementations. It should be understood that these implementations are discussed to enable one of ordinary skill in the art to better understand and thus implement the present disclosure without suggesting any limitation on the scope of the present disclosure.

[0021] As used herein, the terms “comprises” and variations thereof are to be interpreted to an open-ended term meaning “including but not limited to”. The term “based on” is to be interpreted as “based at least in part on”. The terms “one implementation” and “an implementation” are to be interpreted as “at least one implementation”. The term “another implementation” is to be interpreted as “at least one other implementation”. The terms “first,” “second,” and the like may refer to different or identical objects. Other explicit and implicit definitions may also be included below.

[0022] It should be noted that the title of any section / subsection provided herein is not limiting. Various implementations are described throughout the present disclosure and any type of implementation may be comprised in any section / subsection. Furthermore, the implementations described in any section / subsection may be combined in any manner with any other implementation described in the same section / subsection and / or different sections / subsections.

[0023] Herein, unless stated explicitly, performing a step “in response to A” does not indicate that the step is performed immediately after “A” occurs and one or more intervening steps may be included.

[0024] As used herein, the term “model” may learn from training data associations between respective inputs and outputs such that it may generate a corresponding output for a given input after training is completed. The generation of the model may be based on the machine learning technology. Deep learning (DL) is a machine learning algorithm that processes an input and provides a corresponding output by using a multi-layer processing unit. The neural network model is an example of a deep learning-based model. As used herein, a “model” may also be referred to as a “machine learning model,” a “learning model,” a “machine learning network,” or a “learning network,” which terms are used interchangeably herein.

[0025] As mentioned above, in some cases it is desirable to measure a substance attribute, such as substance concentration, in certain objects. As an example of glucose measurement, a common glucose monitoring technology uses an electrochemical method to convert the glucose concentration into an electrical signal for measurement. However, such an electrochemical method is invasive, such as finger needling, thereby limiting the frequency and range for monitoring. At present, there is proposed some non-invasive blood glucose monitoring solutions, including infrared spectrum-related technical solutions and dielectricspectrum-related technical solutions.

[0026] The working principle of the infrared spectrum-related technical solution is that glucose molecules absorb a spectrum of a specific frequency due to vibration and rotation. In such a solution, blood glucose is measured by emitting laser to the subcutaneous tissue and analyzing the spectrum of the capillary. In some solutions using the ultrasonic and photoacoustic technology, the glucose concentration is derived from pressure waves and ultrasonic signals generated by thermal expansion when the glucose molecules absorb the infrared spectrum. A disadvantage of such infrared-related solution is that it is difficult to reduce cost and improve the integration level of the laser transceiver.

[0027] The dielectric spectrum-related technical solution uses changes in dielectric constant caused by the glucose concentration. In such a solution, the blood glucose concentration is estimated by analyzing the propagation characteristics (e.g., attenuation, phase changes) of the electromagnetic signal. The dielectric spectral-related technical solution has certain limitations, especially for medical scenarios. For example, a large thickness of the liquid sample is required for significantly attenuate or phase changes in such solution, which is difficult to implement for scenarios such as blood glucose monitoring. For another example, a container with a significant surface area is typically used in such a solution to ensure that most electromagnetic waves pass through the liquid. However, blood vessels are thin and fragile, and thus it is difficult to ensure that most of the signals pass through the blood vessel, resulting in serious multipath interference of this solution.

[0028] To this end, according to implementations of the present disclosure, a solution for substance attribute measurement using a metasurface is provided. In this solution, the metasurface is attached directly or indirectly to the measured object. A wireless probing signal is emitted by the emitter towards the measured object attached with the metasurface, and a response signal from the measured object emitted or transmitted via the metasurface is received by a receiver. A value of a substance attribute of the measured object, for example, the concentration of a target substance in the measured object, is estimated by analyzing the received response signal. The structure of the metasurface is optimized for substance attribute measurement. The structure of the metasurface may be determined by maximizing the sensitivity of the signal characteristic of the response signal to the change in the value of the substance attribute.

[0029] According to implementations of the present disclosure, the metasurface is attached directly or indirectly to the measured object without immersing in the measured object. Thus, this is a non-invasive and user-friendly measurement solution. Furthermore, the used metasurface is optimized for substance attribute measurement so that a measurement of highaccuracy may be implemented.

[0030] Example implementations of the present disclosure will be described in detail below with reference to the accompanying drawings. Example Architecture

[0031] FIG. 1 illustrates a schematic block diagram of an example measurement architecture 100 according to some implementations of the present disclosure. In the measurement architecture 100, the measurement device 101 is used to measure a target substance attribute of the measured object 130, such as a dielectric constant, a composition of the target substance, a concentration of the target substance, and the like. Generally, the measurement device 101 comprises a metasurface 110, an emitter 121, and a receiver 122.

[0032] The metasurface 110 is adapted to be directly or indirectly attached to the measured object 130, which is also referred to as a target object. In some implementations, the metasurface 110 may be a passive metasurface. In this way, the production cost of the metasurface and the power consumption of the measurement device 101 may be reduced. The structure (e.g., geometry, thickness, material, etc.) of metasurface 110 may be optimized for the measurement of the substance attribute. Example implementations of designing a metasurface structure are described in detail below.

[0033] The emitter 121 is configured to emit a wireless signal, also referred to as a probing signal, to the measured object 130. The emission orientation of the wireless signal enables it to incident on the metasurface 110. The receiver 122 is configured to receive the wireless signal, also referred to as a response signal, from the measured object 130 that propagated (e.g., reflected or transmitted) via the metasurface 110. It should be noted that while FIG. 1 shows the receiver 122 receiving the wireless signal reflected via the metasurface 110, this is merely exemplary and not intended to any limitations. In implementations of the present disclosure, the wireless signal used by the measurement device 101 may be a signal of any suitable frequency band, including but not limited to an ultra-wideband (UWB) signal, a wireless fidelity (WiFi) signal. In some implementations, the wireless signal may be of a UWB frequency band, such as 1-8 GHz. In such a frequency band, the electromagnetic wave loss caused by the measured object is small. In some implementations, the wireless signal may be a WiFi signal. By utilizing the WiFi signal, the solution for substance attribute measurement in the implementation of the present disclosure may be implemented using in a daily user equipment, such as a smart phone, a smart watch, or the like.

[0034] In some implementations, the measurement device 101 may further comprises a controller 125. The controller 125 is coupled to at least the receiver 122 to obtain the response signal, e.g., the reflected signal or the transmitted signal, received by the receiver122. The controller 125 may be configured to estimate a value of a substance attribute, such as the concentration of a target substance in the measured object 130, by analyzing the received response signal. In some implementations, the controller 125 may estimate the value of the substance attribute of the measured object 130 by analyzing a frequency response characteristic of the received response signal. The frequency response characteristic may comprise an S-parameter. For example, in a case where the reflected signal is received, the value of the substance attribute of the measured object 130 may be estimated by analyzing the parameter ^^^^. For another example, in a case where the transmitted signal is received, the value of the substance attribute of the measured object 130 may be estimated by analyzing the parameter ^^ଶ^.

[0035] In some implementations, the controller 125 may further be coupled to the emitter 121 to control the probing signal emitted by the emitter 121, such as the frequency band, the amplitude, the phase, and the like of the probing signal. The controller 125 may be dedicated for the measurement device 101 for substance attribute estimation, or may be a general- purpose controller. The controller 125 may be implemented in any unit having capability of processing, such as a microcontroller, a general-purpose processing unit, or the like. Implementations of the present disclosure are not limited in this regard.

[0036] In some implementations, the measurement device 101 may be implemented as a stand-alone device. In some implementations, the measurement device 101 may be at least partly comprised in any suitable electronic device, such as a wearable device.

[0037] The metasurface 110 may be attached directly or indirectly to the measured object 130 including the target substance. The measured object 130 may be of any form, such as a liquid. In some implementations, the metasurface 110 may not be directly attached to the measured object 130, but may be coupled with the measured object 130 by a container that contains the measured object 130. For example, the metasurface 110 may be affixed to or near the surface of the container. The container may be considered as a set of environmental objects associated with the measured object 130.

[0038] The measured object 130 and the target substance attribute may be of any suitable type depending on the particular measurement scenario. In some implementations, the estimated substance attribute may comprise an attribute of a component of blood. In some implementations, the component of blood may be glucose in blood, and accordingly the estimated value of the substance attribute may be a blood glucose concentration. As an example, in a blood glucose measurement scenario, the metasurface 110 may be attached to a body surface of a measured person, for example, may be placed at the wrist. The measured object 130 may be blood of a measured person, and the target substance may be glucose. Insuch a scenario, the environmental objects associated with the measured object 130 may comprise skin, fat, muscle, bone, and the like. Alternatively, or additionally, in some implementations, the component of blood may be salt or triglyceride in blood, or the like. It should be understood that the glucose, salt, and triglyceride listed herein are merely examples of components of blood. Implementations of the present disclosure may be applied to any other components of blood and are not limited to the examples listed herein.

[0039] In some implementations, the estimated substance attribute may comprise an attribute, e.g., the concentration, of the salt in a salt solution. As an example, in the wound healing measurement scenario, the wound healing rate may be monitored by measuring the salt concentration at the wound to remind the injured to update the gauze. In such a scenario, the metasurface 110 may be attached near a wound of the injured, for example on a gauze. The measured object 130 may be a liquid at the wound, and the target substance may be salt ions, and the environmental objects associated with the measured object 130 may comprise a gauze, the skin, and the like.

[0040] In some implementations, the estimated substance attribute may comprise an attribute of a water contaminant, for example the composition, concentration, and the like. As an example, in a pollution monitoring scenario, a liquid (e.g., water) to be measured may be placed in a container or pipe. The measured object 130 is a liquid in a container or a pipe, and the target substance may be a contaminant that may be present in the liquid.

[0041] It should be understood that the measurement scenarios described above are exemplary only and are not intended to any limitations. The measurement solutions according to implementations of the present disclosure may be applied to any suitable measurement scenarios. Furthermore, example implementations of the present disclosure will be described mainly with an example of blood glucose measurement scenario in the descriptions below. It should be understood, however, that this is merely exemplary and that the principles of the present disclosure described with reference to the blood glucose measurement scenario may be applicable to any suitable measurement scenarios, not limited to the blood glucose measurement. Furthermore, example implementations of the present disclosure will be described primarily with concentrations as estimated substance attributes. However, this is merely exemplary and not intended to any limitations.

[0042] Furthermore, it should be understood that the structure and function of each element in the measurement architecture 100 is described for exemplary purposes only, without suggesting any limitation on the scope of the present disclosure. Optimization for design of metasurface

[0043] In some implementations, the metasurface 110 may be optimized in order to enableaccurate substance attribute measurement. Example implementations of an optimized design of the metasurface 110 is described below with reference to FIG. 2A. FIG. 2A illustrates a flowchart of an example optimization process 200A for a metasurface according to some implementations of the present disclosure. The optimization process 200A may be performed by any suitable electronic device.

[0044] In general, the optimization process 200A may comprise a signal characteristic derivation stage 210, a target electrical characteristic derivation stage 220, and a metasurface structure derivation stage 230. In the signal characteristic derivation stage 210, a signal characteristic, for example an amplitude or phase, of the response signal reflected or transmitted from the measured object 130 via the metasurface 110 is derived by modeling the measurement environment comprising the metasurface 110 and the measured object 130. In the target electrical characteristic derivation stage 220, the target electrical characteristic, for example the target impedance, of the metasurface 110 is determined by maximizing the sensitivity of the signal characteristic of the response signal to a change in the substance attribute (e.g., the concentration change of the target substance) in the measured object 130. In the metasurface structure derivation stage 230, the structure, for example geometry, thickness, material, and the like, of the metasurface 110 is determined based on the target electrical characteristic. The target electrical characteristic may be considered as a design target of the metasurface structure in the metasurface structure derivation stage 230.

[0045] Example implementations of these stages are described below. The blood glucose measurement will be described mainly as an example in the description below, but this is merely exemplary and not intended to any limitations. Signal Characteristic Derivation

[0046] In some implementations, an electrical coefficient of the measured object 130 is modeled at block 211 as shown in FIG. 2A. The modeled electrical coefficient varies with frequency and are therefore also referred to as frequency-related electrical coefficient. The electrical coefficient of the measured object 130 over frequencies at different values of the substance attribute (e.g., different concentration, different compositions) may be determined by modeling the electrical coefficient. That is, such relationship between the frequency- related electrical coefficient and the substance attribute may be derived.

[0047] The considered electrical coefficient may be, for example, the dielectric constant of the measured object 130. The dielectric constant, also referred to as the permittivity, of a substance is an inherent property that is influenced by the principal composition of the substance. In wireless sensing, the dielectric constant affects the transmission reflection characteristic of wireless signals.

[0048] The electromagnetic waves experience reflection and transmission when encountering a measured object (e.g., liquid). The dielectric constant of the measured object changes the characteristics (e.g., the amplitude and phase) of the reflected signal and transmitted signal. Thus, characterizing the dielectric constant for different values of the substance attribute (e.g., different concentrations of the target substance) may be used for substance attribute measurement.

[0049] The dielectric constant is typically expressed in terms of the relative permittivity ^^^. The relative permittivity is the ratio of the absolute dielectric constant and the vacuum dielectric constant, and may be expressed as a complex function of frequency: , where the ^^ is the angular frequency, represents the storage (which is the real the relativedielectric constant), represents the loss factor (imaginary part) of the energy loss due to absorption, conduction, and relaxation.

[0050] The dielectric constant of the measured object 130 may be modeled using any suitable model. As an example, the Cole-Cole model effectively captures the frequency- related dielectric constant, which enables a limited number of measurement values at a particular frequency to fit the relative dielectric constant. For example, the relative dielectric constant may be expressed by: (1)Where highfrequency, represents the dielectric constant in the free space, ^^ is the relaxation time in seconds,^^ is the distribution parameter describing the symmetry broadening of the relaxation loss peak, represents the ion conductivity. The relaxation function is used to describe the loss peak and drop of the ^^′.

[0051] Taking blood glucose monitoring as an example, in order to obtain the frequency- related dielectric constant of glucose at different concentrations, dielectric constant values (including real and imaginary parts) at some glucose concentrations may be measured in any suitable manner. These dielectric constant values may be used to fit parameters of such asthe Cole-Cole model shown in equation (1). In an example, a unipolar Cole-Cole model (where, ^^= 1, ^^ = 0, simplified as a Debye model) may be employed to fit a dielectric constant model for different glucose concentrations (e.g., 0 mg / dL, 100 mg / dL, 200 mg / dL, 300 mg / dL, 400 mg / dL, and 500 mg / dL). Parameters of the dielectric constant model of glucose at different concentrations may be obtained by the fitting, thereby deriving the frequency-related complex dielectric constant of glucose at different concentrations. For example, the frequency-related complex dielectric constant of glucose at each concentration may be represented by a function derived by fitting.

[0052] The dielectric constant modeling of the measured object is described above. Next, how the change in the dielectric constant affects the reflectivity and transmittance of electromagnetic waves at different concentrations is analyzed. The probing signal from the emitter experiences reflection and transmission at the interface of different materials depending on the dielectric constant and the thickness of the measured object. In some implementations, the measured object 130 may be modeled in an equivalent circuit as a circuit element having electrical characteristics, such as a circuit element having a certain impedance. As an example, the measured object 130 may be simulated as a transmission line involving three parameters: ^^, ^^, and ^^ related to the measured object. ^^ is the thickness of the measured object, ^^ is characteristic complex impedance of the measured object, and ^^ is the complex propagation constant of the measured object. Both ^^ and ^^ may be represented by the relative dielectric constant as follows: (2)where ^^ is the angularimpedance in the free space, is the propagation constant in the free , and c is the speed of light. In theof a non-magnetic material, such as for aglucose measurement scenario, the of the measured object may be set to 1.

[0053] By calculating the transmission matrix, parameters such as reflectivity or transmittance may be derived from the equivalent circuit to represent the frequency response characteristics of the signal. An example of a transmission matrix is an ABCD matrix, whichis a 2 × 2 matrix for describing characteristics of a linear dual-port network in a circuit. Elements A, B, C, and D in the ABCD matrix may be calculated based on specific components and connections to describe transmission characteristics.

[0054] In an example, a dual-port network with the transmission line circuit may be modeled as the following ABCD matrix: (3) where the elements of and current and the output

[0055] In order to represent the response signal (reflected or transmitted signal) received at the receiver, the ABCD parameters may then be converted to S-parameters to derive ^^^^parameters (i.e., the reflectivity) and ^^ଶ^parameters (i.e., the transmittance): (4)

[0056] The ^^^^curve and / or ^^ଶ^ofamplitude orversus frequency, in the case of different thicknesses and different values of the substance attribute of the measured object may derived by using equation (4). Taking the blood glucose measurement scenario as an example, if the thickness of the liquid is 30 cm, the change in the amplitude of the ^^^^is about 0.001 dB and the change in the phase is about 0.01 degrees for every100 mg / dL of the glucose concentration. Such changes are easily overwhelmed by ambient noises. If the thickness of the liquid is 10 cm, the change in the amplitude is about 0.001 dB and the change in the phase is about 0.02 degrees for every change of 100 mg / dL of the glucose concentration. Similarly, the ^^ଶ^parameters show a commensurate variation with the ^^^^parameters. It follows that if the thickness of the solution is small, such as 3 mm in the blood vessel, it is almost impossible to infer the concentration of glucose from parameters such as ^^^^or ^^ଶ^parameters.

[0057] Reference is made back to FIG. 2A. At block 212, the measurement environment may be modeled using an equivalent circuit. That is, an equivalent circuit representing the various components of the measurement environment is generated. Components of the measurement environment may include various layers in the metasurface 110, the measuredobject 130, and environmental objects associated with the measured object, such as surrounding material of the measured object. Individual components in the equivalent circuit may be represented as circuit elements having respective electrical characteristics (e.g., impedances).

[0058] As an example, at block 212, a corresponding equivalent circuit may be constructed according to a specific setting of the measurement environment. According to the specific setting of the measurement environment, the circuit elements representing respective components are connected together to form an equivalent circuit. Each component is represented as a respective impedance. As such, the reflected or transmitted signal may be represented by the impedance of the measured object 130, which in turn is related to the value of the substance attribute.

[0059] Example construction of an equivalent circuit is described below. In an equivalent circuit, a transmission line representing a measured object is similar to a single capacitive element. An LC resonant circuit may be considered to amplify the change in capacitance caused by the change in concentration. In an LC resonant circuit, a minor change in capacitance would change resonant characteristics, such as a resonant frequency point and an overall frequency response curve.

[0060] A simplified example of an equivalent circuit construction is described below with reference to FIGS. 3A and 3B by taking the liquid as an example of the measured object. FIG. 3A illustrates an example measurement environment 300A of the concentration of the liquid aided by a metasurface in accordance with some implementations of the present disclosure. The measurement environment 300A may be considered a simplified example of the measurement architecture shown in FIG. 1. In the measurement environment 300A, a metasurface 310 is placed on the surface of a liquid 330 being measured. A transceiver 320 is used to emit wireless signals to and receive wireless signals reflected from the metasurface 310.

[0061] FIG. 3B illustrates an equivalent circuit 300B corresponding to the measurement environment 300A. In the equivalent circuit 300B, an LC network 360 is used to represent the metasurface 310, while a single capacitive element 370 with a capacitance value is used to represent liquid 330. Thus, the objective of optimization is to optimize the LC network 360 so that it best matches the the liquid 330, ensuring that the entire circuit exhibits high qualityIn this way, the capacitance value of the liquid 330 may change the resonant characteristic to the greatest extent as the capacitance value experiences a slight change due to a change inconcentration.

[0062] Next, the LC circuit introduced by the metasurface may be modeled. An example is described with reference to FIGS. 4A and 4B. FIG. 4A illustrates an example structure of a metasurface unit 400A that constitutes a metasurface according to some implementations of the present disclosure. The entire metasurface may be implemented by repeating such as the metasurface unit 400A shown in FIG. 4A. As shown in FIG. 4, the metasurface may be a passive metasurface, which may comprise a metal layer 401 and a substrate layer 402. The substrate layer 402 may for example be made of paper, which can effectively reduce manufacture costs. The metal layer 401 is arranged above the substrate layer 402 and forms a pattern with a specific geometry.

[0063] In FIG.4A, a metasurface unit is shown having a finger pattern as an example. This is a metasurface design with planar capacitors and multi-finger periodicity units. As shown in FIG. 4A, in this example, the structural parameters for defining the geometry configuration of the metasurface may comprise a dimension L of the metasurface unit, a length b of the metal layer 401 along the y-axis, a length a of the finger-like along the x- axis, a space distance w between two neighboring finger-likes, and a number N of the finger- likes. It should be understood that the pattern types shown in FIGS. 4A are exemplary only and are not intended to limit the scope of the present disclosure. The metasurface optimization described in the implementations of the present disclosure may be applicable to a metasurface of any pattern.

[0064] FIG. 4B illustrates an example of an equivalent LC circuit 400B of a metasurface according to some implementations of the present disclosure. The equivalent LC circuit 400B is described with reference to FIG. 4A. As a wireless signal (e.g., electromagnetic wave) propagates along the z-axis and interacts with the metasurface, it first encounters the metal layer 401. The metal layer 401 equals to a parallel LC circuit in circuit modeling. The model of the LC circuit is directly related to the pattern of the metal layer 401, and the pattern of the metal layer 401 may generate an equivalent capacitance (with an impedance of , where the ^^ is an angular frequency) and an inductance (^^^^^^) to determine its

[0065] For the example pattern in FIG.4A, each pair of finger-likes forms a capacitor, i.e., ^^^^1and ^^^^2, as well as a virtual parasitic inductance caused by the connection of the capacitor, i.e., ^^i. For simplicity, the i-th metal layer in the metasurface may be represented by the equivalent impedance ^^LCi, as shown in FIG. 4B. The corresponding ABCD matrix may be represented as:. (5a)

[0066] The substrate layer which is similar to the liquid modeling described above. The substrate layer 402 has its own unique properties, such as the dielectric constant and thickness. Similarly, an ABCD matrix representing the transmission line of the substrate layer 402 may be characterized using a propagation constant and a characteristic impedance by equation (3).

[0067] In the case where the metasurface 130 is a passive metasurface, the entire ABCD matrix of the passive metasurface is as follows:of the i-th metal layer, , andthe thickness, characteristicimpedance and of the i-th substrate layer, respectively. , , , and represent the A, B, C, D elements in therespectively.

[0068] In view of the above, a complete equivalent circuit corresponding to the measurement environment may be established. An example is described with reference to FIG. 5A. FIG. 5A illustrates an equivalent circuit 500 corresponding to a measurement environment according to some implementations of the present disclosure, which may be considered as a modeling of at least a part of the measurement architecture 100 shown in FIG. 1. Specifically, in this example, the metasurface 110 may contain four metal layers and a substrate layer. In the equivalent circuit 500, the LC network 510 is used to represent the metasurface 110. The four parallel impedance elements Z1, Z2, Z3and Z4represent four metal layers, respectively, and four transmission line elements represent four substrate layers, respectively. The transmission line 530 is used to represent the measured object 130. The environmental objects associated with the measured object 130 may be represented by transmission lines 541 and 542. For example, if glucose concentration in a blood vessel in an arm is measured, the transmission line 530 may represent blood, and the transmission line 541 may represent skin, fat, and muscle, while the transmission line 542 may representmuscle, bone, fat, and skin.

[0069] Next, at block 213, the signal characteristic of the response signal may be determined based on respective electrical coefficients (e.g., dielectric constants) of the plurality of components in the equivalent circuit. For example, in the case of a reflected signal, the amplitude and / or phase of ^^^^may be determined. In the case of a transmitted signal, the amplitude and / or phase of ^^ଶ^may be determined.

[0070] An example process isbelow. The equivalent circuit 500 may be expressed using a transmission matrix such as an ABCD matrix. A good characteristic of the ABCD matrix is that the ABCD matrix of the entire circuit may be derived from the product of the ABCD matrices corresponding to the individual elements. The ABCD matrix of the four metal layers in the metasurface 110 may be represented by ^^|^|, ^^|ଶ|, ^^|ଷ|, and ^^|ସ|, and the ABCD matrix of the substrate layer in metasurface 110 ^^௧^^^,^^௧^^ଶ, ^^௧^^ଷ, and ^^௧^^ସ, and the environment object in FIG. may ^^௧^^^and ABCD matrix of measured object 130 may be represented ^^^௧^.Accordingly, the entire ABCD matrix of the measurement environment may be asfollows:

[0071] Byof the measured object 130 may be derived from the relative dielectric constant parameters. Then, the entire ABCD matrix value may be calculated and combined with equation (4) to obtain the ^^^^parameter. The ^^^^parameter represents a reflective frequency response characteristic for a wireless signal (e.g., electromagnetic waves) interacting with the metasurface and the measured object. Alternatively, or additionally, the entire ABCD matrix value may be calculated and combined with equation (4) to obtain the ^^ଶ^parameter. The ^^ଶ^parameter represents a transmission frequency response characteristic for a wireless signal (e.g., electromagnetic waves) interacting with the metasurface and the measured object.

[0072] Example implementations of constructing an equivalent circuit and deriving a representation of a wireless signal based on an equivalent circuit are described above. Specifically, the S parameter of the wireless signal may be derived. A parameter ^^^^of the response signal may be derived if the response signal is a reflected signal. An ^^ଶ^parameterof the response signal may be derived if the response signal is a transmitted signal.

[0073] Furthermore, four metal layers and finger patterns are used in the description above as examples, but this is for illustrative purposes only and is not intended to limit the scope of the present disclosure. In implementations of the present disclosure, the metasurface may have any suitable number of metal layers and substrate layers, and the metasurface units have any suitable pattern.

[0074] As an example, the construction of the equivalent circuit of the measurement environment in the blood glucose measurement scenario is described below. FIG. 5B illustrates a model 501 of a measurement environment in the blood glucose measurement scenario. The measured object is blood, and the environmental object comprises tissues such as skin, fat, muscle, bone and the like. Each kind of tissues may be modeled as a transmission line and represented by a thickness l, an equivalent characteristic impedance Z, and a propagation constant g. The ABCD matrix of the measurement environment may be expressed as follows: where thes, f, m, b the skin, fat, muscle, bone, and blood, respectively. After the entire ABCD matrix value is calculated, the reflected signal ^^^^may be obtained using equation (2).Derivation of Target Electrical Characteristic

[0075] In the target electrical characteristic derivation stage 220, the target electrical characteristic, such as the target impedance, of the metasurface 110 may be determined by solving the optimization problem. The target of this optimization problem may be to maximize the sensitivity of the received response signal to a change in the substance attribute in the measured object 130, such as sensitivity to a change in the concentration of the target substance.

[0076] Some implementations are described by taking the concentration of the target substance as an example of the estimated substance attribute and the target impedance as an example of the target electrical characteristic. In some implementations, the sensitivity maybe represented as a difference in the received response signal at different concentrations of the target substance. For example, in a case where a reflected signal is received, the target of the optimization problem may be to maximize the difference in the ^^^^parameter of the wireless signal at different concentrations of the target substance. As another example, in a case where the transmission signal is received, the target of the optimization problem may be to maximize the difference of the ^^ଶ^parameter of the wireless signal at different concentrations of the target substance. In this way, the sensitivity and accuracy of substance attribute measurement may be improved to the greatest extent, thereby implementing effective substance monitoring. To this end, in some implementations, the optimized impedance of the passive metasurface may be derived by maximizing the difference in response signals at different concentrations of the target substance.

[0077] In order to solve the optimization problem, a measurement environment and various environmental parameters may be set. Next, the liquid thickness (e.g., 3 mm), the number of metal layers and substrate layers in the metasurface, and the thickness of each substrate layer (e.g., 0.3 mm) may be determined. The frequency range of the wireless signal may then be set, for example, as 5.5-6.5 GHz. In particular, the frequency range may be set according to a specific application scenario and a frequency band of a signal that may be used.

[0078] In some implementations, the optimization target may be to optimize the design of each metal layer to cause the ^^^^parameter most sensitive to a change in the concentration of the target substance. To simplify the optimization target, each metal layer may be equivalent to one impedance, where the impedance may be, for example, a pure complex number, e.g., j^^. This is because the metal layer may be considered as an ideal electrical conductor. The impedance of the metal layer changes with frequency, and thus the signal is a function of frequency, i.e., the ^^(f). Thus, the optimization target may be represented as: (7)whereMconcentration levels of the target substance, and ^^^^ is the i- th concentration value. For example, mg / dL. The is referred to the ^^^^parameter of the measured object in the case that the concentration ofthe target substance is ^^^^. The optimization target may be to maximize the difference of the ^^^^parameters between different concentrations at the target substance.

[0079] As an example, distances between the ^^^^parameters at different concentrations may be used as the optimization target. In this way, the following advantages may be achieved: (1) maximizing the distance in the frequency response to effectively maximize the sharpness of the resonant frequency (i.e., minimizing the ^^^^^^(^^^^)) and the difference of the resonant frequencies (i.e., maximizing the ); (2) the distance-based loss function having a gradient during backpropagation as compared to the ^^^^^^() and ^^^^^^^^^^^^() function, thereby supporting gradient descent optimization.

[0080] In the above optimization process, parameters of the measurement environment and various environmental objects related to the measured object need to be set. This method is applicable to some scenarios, such as measuring the liquid within a container, since the parameters of the container are known. However, for some scenarios, the parameters of the environmental objects are unknown. For example, for blood glucose measurement, the parameters of tissues such as human skin, fat, blood vessels, muscles, and bones are unknown. One way is to assume the parameters of these environmental objects, as described above.

[0081] In some implementations, to further improve the accuracy of the measurement, calibration may be performed using the active metasurface at block 221. By using the calibration, the optimized capacitance value of the active metasurface may be derived as a result. In some implementations, an active metasurface may be deployed.

[0082] In some other implementations, at block 222, the target electrical characteristic of the passive metasurface may be determined based on a result of the calibration (e.g., based on the optimized capacitance value of the active metasurface), thereby deploying the passive metasurface. An example is described below with reference to FIG. 5B.

[0083] The active metasurface for calibration may comprise a variable capacitor to adapt to the measurement environment. The variable capacitor may be implemented, for example, by a voltage controlled varactor diode. FIG. 5B illustrates an example unit 502 of the active metasurface. As shown in FIG.5B, the unit 502 comprise a substrate layer 512, a metal layer 511 (e.g., a copper layer) arranged above the substrate layer 512. The metal layer 511 has one or more capacitors 513 of a variable capacitance value. The metal layer 511 may be, for example, a metal patch to be used as an antenna. The metal layer 511 may have a gap to expose the underlying substrate layer. The variable capacitor 513 may be attached to themetal layer and arranged in the gap within the metal layer 511. The variable capacitor 513 may be implemented by a varactor diode, respectively.

[0084] In some implementations, a range of the capacitance variation of the active metasurface may be determined based on a signal characteristic of the response signal and a range of the parameter value of the environmental object associated with the measured object. A range of the parameter value may be chosen as wide as possible to cover various situations that may occur in actual measurements.

[0085] In some implementations, a range of the impedance variation of the active metasurface may be determined based on a signal characteristic of the response signal and a range of a parameter value of the environmental object. A range of the capacitance variation of the active metasurface may be derived by using finite element modeling based on the range of the impedance variation.

[0086] As an example, a larger range of the parameter value may be used for each tissue layer for the blood glucose monitoring scenario. For example, the following tissue thickness ranges may be used: skin (0.5 -2 mm), fat (1-10 mm), muscle (10 -30 mm), bone (0-20 mm). Furthermore, publicly accessible dielectric constant data may be employed, and a certain range of the dielectric constant variation may be allowed in view of the differences between individuals. Thus, an equivalent circuit 503 such as shown in FIG. 5B may be constructed. The equivalent circuit 503 may be considered as an equivalent circuit corresponding to a case where a passive supersurface in the model 501 is replaced with the active metasurface. A range of the equivalent impedance for the active metasurface may be determined by using the equivalent circuit 503. An equivalent circuit model of the active metasurface may then be established in the finite element modeling tool to determine the range the variable capacitance based on the range of the equivalent impedance variation.

[0087] During the calibration process, the active metasurface is attached to a target environmental object, such as attached to a matching piece on the arm. Next, the capacitance of the variable capacitor may be adjusted within the aforementioned range of the capacitance. During the capacitance adjustment process, the signal characteristic of the response signal may be tracked. The capacitance value resulting in the optimal resonant coupling may be determined as an optimized capacitance value of the active metasurface. The passive metasurface with an ABCD matrix matched with that of the active metasurface may be obtained by using the optimized capacitance value, so that the design of sensitive substance concentration is implemented.

[0088] For example, the variable capacitance of the active metasurface may be adjusted to enable the active metasurface to optimally resonate with the measurement environment.Upon identifying the optimal capacitance value for the variable capacitor, finite element modeling may be used to determine the equivalent impedance of the active metasurface. The subsequent objective is to design a passive metasurface with such an equivalent impedance. That is, the equivalent impedance of the active metasurface thus obtained may be used as the target impedance for the passive metasurface. Metasurface Structure Derivation

[0089] Reference is made back to FIG. 2A. In the metasurface structure derivation stage 230, the structure of the metasurface is determined based on the target electrical characteristic of the metasurface, for example, the geometry, thickness, material and the like of the metal layer. At block 231, a structure of the metasurface matching the target electrical characteristic may be determined. Some implementations of metasurface structure derivation are described below by taking impedance as an example of the electrical characteristic.

[0090] In some implementations, the frequency-related impedance function may be derived. The impedance of the metal layer is related to its pattern. As an example, reference is still made to the example pattern in FIG. 4A. For example, in the case that the period length is 6 mm (1 / 8 of the wavelength of the electromagnetic wave of about 4 GHz), the impedance may vary from the -500^^ to the -10^^ at 4GHz by adjusting the geometric parameters such as ^^, ^^, ^^, ^^ and ^^. The target of optimizing the impedance function of the metal layer may be to optimize the geometric parameters of the pattern of each metal layer, i.e., the ^^, ^^, ^^, ^^ and ^^. In some implementations, for ease of fabrication, the metasurface units of each metal layer may be set to share a geometric parameter ^^. In this case, the optimization target may be to find the optimal geometric parameters of the metasurface units to ensure that ^^(f; ^^, ^^, ^^, ^^, ^^) maximizes the sensitivity of the change in the concentration of the target substance.

[0091] In some implementations, the impedance functions of the metal patterns with different geometric parameters may be obtained and the relationship between these geometric parameters and the impedance function coefficients is established. FIG. 6 shows the impedance (imaginary part) of a pattern with different geometric parameters over a frequency range of 1-6 GHz. Each curve in FIG. 6 corresponds to a combination of the geometric parameters of ^^, ^^, ^^, ^^, ^^, that is, corresponds to a pattern. The curves shown in FIG. 6 may be obtained by finite element analysis.

[0092] These curves may be approximated by an n-order polynomial function, i.e., . For each of the ^^^^ coefficients, thebe fitted, i.e., ^^^^ = ^^^^ (^^, ^^, ^^, ^^). In thisway, impedance information about the frequency may be obtained directly based on the geometric parameters, thereby determining the optimal geometric parameter.

[0093] By modeling the impedance function with respect to the geometric parameters of the metal pattern, the optimal metasurface design may be determined for each metal layer. For example, the objective equation may be optimized and the sensitivity of the ^^^^parameter to the concentration of the target substance may be maximized by using a gradient descent optimizer. As an example, the optimization objective function may be represented as the following form: (8) where S1(^^), ^^,^^, ^^. some may to adjust hyper-parameters to obtain the most effective metasurface design within a particular frequency range. The hyperparameter may comprise, for example, a shared period length ^^ for each metasurface unit, a number of metal layers and substrate layers, and the like.

[0094] The respective ^^^^parameters of different glucose solution concentrations are shown in FIGS. 7A for illustrative purposes only. Specifically, the metasurface inthis example resonates around 6GHz where L = 6 mm with 4 metal layers and 4 substrate layers. FIG. 7A shows curves of amplitude versus frequency of the ^^^^parameter at different glucose concentrations, where curves 711, 712, 713, 714,correspond to glucose concentrations of 0 mg / dL, 100 mg / dL, 200 mg / dL, 300 mg / dL, 400 mg / dL, 500 mg / dL, respectively. As can be seen from FIG.7A, the change in glucose concentration may result in an offset of the resonance point for about 0.8 MHz frequency offset per 100 mg / dL. Although this frequency offset is small, the ^^^^amplitude parameter at each concentration is well distinguished. Increasing theat a resonant frequency of 0 mg / dL would change the ^^^^amplitude of other concentrations at that frequency by 5 dB per 100 mg / dL. FIG. 7Bcurves of phase versus frequency of the ^^^^parameter at different glucose concentrations, where curves 721, 722, 723,726 correspond to glucose concentrations of 0 mg / dL, 100 mg / dL, 200 mg / dL, 300 mg / dL, 400 mg / dL, 500 mg / dL, respectively. The change of the phase parameter is approaching 50°. This difference is greater than the change in the ^^^^parameter without the metasurface.

[0095] Exampleof optimizing the structure of the metasurface by way ofmodeling of an equivalent circuit are described above.

[0096] In some implementations, the structure of the passive metasurface may be determined based on the equivalent impedance of the active metasurface obtained at block 222. For example, a target transmission matrix of the passive metasurface may be determined based on an equivalent impedance of the active metasurface. The structure of the passive metasurface is determined by minimizing the difference between the transmission matrices for the passive metasurface with different structures and the target transmission matrix.

[0097] As an example, may be derived by establishing the of themetal layer. In combination layer, the overall ABCD matrix of the passive metasurface represented by the parameter set may be derived, where m is the number of layers of the passive metasurface. The optimization objective is to match the equivalent impedance of the passive metasurface to the calibrated equivalent impedance of the active metasurface, and thus the ABCD matrix of the passive metasurface and the ABCD matrix of the active metasurface are equal. The ABCD matrix of the active metasurface may be represented as . In this case, the objective loss function is as follows:(9) thus, thethat minimize the loss function described above. The hyperparameter may be adjusted using hyperparameter debugging techniques to implement an optimal passive metasurface design.

[0098] To compensate for the coupling effect between different metal layers, in some implementations, the target electrical characteristic (e.g., the impedance) described above may be fine-tuned. The optimized impedance may be fine-tuned by finite element modeling of the measurement environment to derive the target impedance of the metasurface. In such implementations, the signal characteristic of the response signal is determined based on the finite element model of the measurement environment. The target electrical characteristic(such as the impedance) of the metasurface is optimized by maximizing the difference between the signal characteristics of the response signal at different values of the substance attribute (e.g., different concentrations, different components) of the measured object based on the finite element model, thereby optimizing the structure of the metasurface. The example implementations are described below by taking the impedance as an example.

[0099] The equivalent circuit model may effectively represent the impedance of each element and the frequency response of the electromagnetic waves reflected by the metasurface and the measured object is simulated by using ABCD matrix. Coupling among a plurality of metal layers in a metasurface can be accurately simulated by using finite element modeling, especially if the substrate layer thickness is small and the coupling is not negligible.

[0100] FIG. 8 illustrates a schematic diagram of an example finite element model 800 for a measurement environment according to some implementations of the present disclosure. In this example, the metasurface 810, the container layer 821 in proximity of the metasurface 810, the measured object 830 that may include the target substance, and the container layer 822 are sequentially simulated from the input port 871 to the output port 872 of the wireless signal. The metasurface 810 includes 4 metal layers, which are metal layers 811, 812, 813, and 814, respectively.

[0101] To compensate for the coupling effect between different metal layers, any suitable type of finite element simulator may be used. Such a simulator can solve the optimization problem for simulating electromagnetic behavior (e.g., near field coupling) on the basis of the initial optimized metasurface design. The same optimization target (such as equation (9)) as described above may be applied in a finite element-based simulator.

[0102] Although the design of the metasurface may be optimized directly using the finite element modeling, the two-step approach proposed in implementations of the present disclosure has more advantages. For example, by using the preliminary optimization result of the equivalent circuit as the initiation of the finite element modeling optimization, the computational cost and time consumption of the finite element solution can be greatly reduced. As another example, finite element modeling does not facilitate adjustment of hyperparameters. The adjustment of the hyper-parameters, such as the number of metal layers, requires reestablishment of the finite element model, which reduces the optimization efficiency.

[0103] Therefore, in the implementation described above, an equivalent circuit model is established by using an ABCD transmission matrix theory to obtain a rough design of the metasurface. The near field coupling between metal layers may then be compensated for byusing a finite element simulator and a metasurface which is optimized, reliable and sensitive to the concentration of the target substance, is ultimately achieved.

[0104] In some implementations, a machine learning model may be used to generate a metal pattern that matches the target electrical characteristic. Such a machine learning model is trained using a plurality of training samples. Each training sample may comprise a reference pattern for the metal layer and an electrical characteristic corresponding to the reference pattern. The reference pattern may be defined by a plurality of geometric parameters, such as those described above. The machine learning model trained in this way may learn an association relationship between electrical characteristics and patterns. Correspondingly, for each metal layer, the machine learning model may generate an optimized pattern of the metal layer based on a corresponding electrical characteristic (for example, an impedance) of the metal layer, for example, determine the geometric parameter ^^, ^^, ^^, ^^ described above.

[0105] The machine learning model used herein may be implemented based on any suitable machine learning technique. For example, the machine learning model may comprise a convolutional neural network. For another example, the machine learning module may comprise a generative adversarial network (GAN) or a diffusion model. Example Measurement Process for Substance attribute

[0106] FIG. 2B illustrates a flowchart of an example measurement process 200B using a metasurface according to some implementations of the present disclosure. As an example, the process 200B is described with reference to FIG. 1.

[0107] At block 240, the metasurface 110 is attached to the measured object 130. For example, in a blood glucose monitoring scenario, at least a part of the measurement device 101 is worn to the wrist by a user. At block 250, a probing signal is emitted to the measured object 130 to which the metasurface 110 is attached. For example, the emitter 121 emits a wireless signal within a predetermined frequency band towards the measured object 130 under the control of the controller 125.

[0108] At block 260, a response signal from the measured object 130, which is reflected or transmitted via the metasurface 110, is received. For example, the receiver 122 receives a reflected or transmitted signal which is propagates from the measured object 130 via the metasurface 110.

[0109] At block 270, a value of a substance attribute in the measured object 130, such as the composition or concentration of the target substance, is estimated by analyzing the response signal. For example, in the case of a reflected signal, the ^^^^parameter may be analyzed. As another example, in the case of a transmitted^^ଶ^parameter maybe analyzed.

[0110] In some implementations, the value of the substance attribute may be estimated by analyzing the frequency response characteristic of the response signal, as shown in block 271. In some implementations, the estimated substance attribute may be a concentration of the target substance. For example, the controller 125 may determine a resonant frequency of the response signal and compare the resonant frequency of the response signal to the respective resonant frequencies of the target substance of the measured object at different concentrations. In this way, the concentration of the target substance may be determined based on the comparison. The concentration at which the resonance frequency is mostly approximate to the resonance frequency may be determined as the concentration of the target substance.

[0111] In some implementations, the estimated substance attribute may be a composition of the target substance. For example, the controller 125 may determine a resonant frequency of the response signal and compare the resonant frequency of the response signal to respective resonant frequencies of different compositions of the target substance in the measured object. As such, the composition of the target substance may be determined based on the comparison. The composition of which the resonance frequency is mostly approximate to the resonance frequency may be determined as the composition of the target substance.

[0112] In some implementations, the value of the substance attribute, such as a dielectric constant, composition, or concentration, is estimated using a machine learning model based on the response signal, as shown at block 272. Such a machine learning model may comprise, for example, a decision tree, a neural network. Example Process

[0113] FIG. 9 illustrates a flowchart of a process 900 of substance attribute measurement according to some implementations of the present disclosure. For example, the process 900 may be implemented at measurement device 110 of FIG. 1.

[0114] At block 910, the measurement device 110 emits, by an emitter, a probing signal to the measured object to which the metasurface is attached. At block 920, the measurement device 110 receives, by the receiver, a response signal from the measured object reflected or transmitted via the metasurface. At block 930, the measurement device 110 estimates a value of a substance attribute in the measured object by analyzing the response signal.

[0115] In some implementations, the metasurface attached to the measured object is an active metasurface or a passive metasurface.

[0116] In some implementations, estimating the value of the substance attribute comprise:estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal.

[0117] In some implementations, estimating the value of the substance attribute by analyzing the frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonant frequency of the response signal to respective resonant frequencies of the target substance in the measured object at different concentrations; and determining the concentration of the target substance based on the comparison.

[0118] In some implementations, estimating the value of the substance attribute by analyzing the frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonant frequency of the response signal to respective resonant frequencies of the target substance in the measured object at different compositions; and determining the composition of the target substance based on the comparison.

[0119] In some implementations, estimating the value of the substance attribute by analyzing the response signal comprises: estimating the value of the substance attribute by using a machine learning model based on the response signal.

[0120] In some implementations, the value of the substance attribute comprises an attribute value of the following substance: a composition of blood, a salt in the salt solution, or a water contaminant.

[0121] In some implementations, the composition of blood comprises at least one of: glucose in blood, salt in blood, or triglycerides in blood.

[0122] In some implementations, the metasurface attached to the measured object is an active metasurface, and the active metasurface comprises: a substrate layer; a metal layer arranged above the substrate layer, the metal layer having one or more capacitors of a variable capacitance value.

[0123] In some implementations, the metasurface attached to the measured object is a passive metasurface, and the passive metasurface comprises: a substrate layer (e.g., paper); a metal pattern arranged above the substrate layer.

[0124] FIG. 10 shows a flowchart of a process 1000 of designing a metasurface according to some implementations of the present disclosure. The process 1000 may be implemented at any suitable electronic device.

[0125] At block 1010, the electronic device determines a signal characteristic of a response signal reflected or transmitted from a measured object via a metasurface based on a measurement environment comprising the metasurface and the measured object. At block1020, the electronic device determines the electrical characteristic of the metasurface by maximizing the sensitivity of the signal characteristic of the response signal to the change in a substance attribute of the measured object. In some implementations, at block 1030, the electronic device determines a structure of the metasurface based on the electrical characteristic of the metasurface.

[0126] In some implementations, determining the signal characteristic of the response signal reflected or transmitted from the measured object via the metasurface comprises: determining electrical coefficients of the measured object over frequencies at different values of the substance attribute; generating an equivalent circuit representing a plurality of components of the measurement environment, the plurality of components comprising individual layers in the metasurface, the measured object, and an environmental object associated with the measured object, the plurality of components being represented respectively as circuit elements having respective electrical characteristics in the equivalent circuit; and determining, according to the equivalent circuit, the signal characteristic of the response signal based on respective electrical coefficients of the plurality of components in the equivalent circuit.

[0127] In some implementations, the metasurface is a passive metasurface, and determining the electrical characteristic of the metasurface comprises: determining an optimized capacitance value of an active metasurface based on a resonant characteristic of the active metasurface within a capacitance variation range; and determining the electrical characteristic of the passive metasurface based on the optimized capacitance. In some implementations, the process 1000 further comprises determining the capacitance variation range of the active metasurface based on the signal characteristic and a parameter value range of an environmental object associated with the measured object.

[0128] In some implementations, the metasurface is an active metasurface, and determining the capacitance variation range of the active metasurface comprises: determining an impedance variation range of the active metasurface based on the signal characteristic, the measured object and the parameter value range of the environmental object; and deriving, based on the impedance variation range, the capacitance variation range of the active metasurface by using finite element modeling.

[0129] In some implementations, determining the structure of the metasurface comprises: determining a target transmission matrix for the passive metasurface based on the electrical characteristic; and determining the structure of the passive metasurface by minimizing a difference between a transmission matrix of the passive metasurface with different structures and the target transmission matrix.

[0130] In some implementations, the process 1000 further comprises: determining, based on a finite element model of the measurement environment, the signal characteristic of the response signal; and optimizing, based on the finite element model, the electrical characteristic of the metasurface by maximizing a difference of the signal characteristic of the response signal among different values of the substance attribute of the measured object.

[0131] In some implementations, the metasurface comprises one or more metal layers, and determining the structure of the metasurface comprises: determining respective structures of the one or more metal layers using a machine learning model based on respective electrical characteristics of the one or more metal layers.

[0132] In some implementations, the machine learning model comprises at least one of: a convolutional neural network, a generative adversarial network, or a diffusion model.

[0133] In some implementations, the value of the substance attribute comprises at least one of: a composition of the target substance in the measured object, or a concentration of the target substance. Example Device

[0134] FIG. 11 illustrates a schematic block diagram of an electronic device capable of implementing various implementations of the present disclosure. It should be understood that the electronic device 1100 shown in FIG. 11 is merely exemplary and should not constitute any limitation on the functionality and scope of the implementations described herein the present disclosure.

[0135] As shown in FIG. 11, the electronic device 1100 comprises an electronic device 1100 in the form of a general-purpose computing device. Components of the electronic device 1100 may comprise, but are not limited to, one or more processors or processing devices 1110, a memory 1120, a storage device 1130, one or more communication units 1140, one or more input devices 1150, and one or more output devices 1160.

[0136] In some implementations, the electronic device 1100 may be implemented as a computing device, a computing system, a server, a mainframe, or the like with computing capabilities.

[0137] The processing device 1110 may be a real or virtual processor and capable of performing various processes according to programs stored in the memory 1120. In multiprocessor systems, a plurality of processing units execute computer-executable instructions in parallel to improve parallel processing capabilities of the electronic device 1100. The processing device 1110 may comprise a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor, a controller, and / or a microcontroller, and the like.

[0138] Electronic device 1100 typically comprises a plurality of computer storage media.Such media may be any available media accessible to the electronic device 1100, comprising, but not limited to, volatile and non-volatile media, removable and non-removable media. The memory 1120 may comprise volatile memory (e.g., registers, caches, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 1130 may comprise removable or non-removable media, and may comprise computer-readable media such as internal storage, flash memory drives, magnetic disks, or any other media that can be used to store information and / or data and that can be accessed within electronic device 1100.

[0139] The electronic device 1100 may further comprise additional removable / non- removable, volatile / non-volatile storage media. Although not shown in FIG. 11, a magnetic disk drive for reading or writing from a removable, nonvolatile magnetic disk and an optical disk drive for reading or writing from a removable, nonvolatile optical disk may be provided. In these cases, each drive may be connected to a bus (not shown) by one or more data media interfaces.

[0140] The communication unit 1140 implements communication with another computing device over a communication media. Additionally, the functionality of components of the electronic device 1100 may be implemented in a single computing cluster or multiple computing machines capable of communicating through a communication connection. Thus, the electronic device 1100 may operate in a networking environment using logical connections with one or more other servers, personal computers (PCs), or another general network node.

[0141] The input device 1150 may be one or more input devices, such as a mouse, a keyboard, a data import device, or the like. The output device 1160 may be one or more output devices, such as a display, a data export device, or the like. The electronic device 1100 may also communicate with one or more external devices (not shown) with the communication unit 1140 as needed, external devices such as storage devices, display devices, etc. , communicate with one or more devices that enable a user to interact with the electronic device 1100, or communicate with any device (e.g., network card, modem, etc. ) that enables the electronic device 1100 to communicate with one or more other computing devices. Such communication may be performed via an input / output (I / O) interface (not shown).

[0142] In some implementations, some or all of the various components of the electronic device 1100 may be set in the form of a cloud computing architecture in addition to being integrated on a single device. In a cloud computing architecture, these components may beremotely arranged and may work together to implement the functionality described herein the present disclosure. In some implementations, cloud computing provides computing, software, data access, and storage services without end-user knowing the physical location or configuration of systems or hardware that provide these services. In various implementations, cloud computing provides services over a wide area network, such as the Internet, using an appropriate protocol. For example, cloud computing providers provide applications over a wide area network, and they may be accessed by a web browser or any other computing component. Software or components of the cloud computing architecture and corresponding data may be stored on servers at remote locations. The computing resources in the cloud computing environment may be merged at the location of the remote data center or they may be dispersed. The cloud computing infrastructure may provide services by the shared data center, even though they appear as a single access point for the user. Thus, the components and functions described herein may be provided from a service provider at a remote location by using a cloud computing architecture. Alternatively, they may be provided from a conventional server, or they may be directly or in other manners installed on a client device.

[0143] The electronic device 1100 may be used to implement metasurface optimization in a plurality of implementations of the present disclosure. The memory 1120 may comprise one or more modules having one or more program instructions that may be accessed and run by the processing unit 1110 to implement the functions of the various implementations described herein. For example, the memory 1120 may comprise a metasurface design module 1122 for performing a design of the metasurface. As shown in FIG.11, the electronic device 1100 may obtain the input required by the metasurface design through the input device 1150, and may provide the output of the metasurface design through the output device 1160. In some implementations, the electronic device 1100 may also receive input from other devices (not shown) via the communication unit 1140. Example Implementations

[0144] Some example implementations of the present disclosure are listed below.

[0145] In one aspect, the present disclosure provides a method of wireless measurement. The method comprises: emitting, by an emitter, a probing signal to a measured object attached with a metasurface; receiving, by a receiver, a response signal from the measured object reflected or transmitted via the metasurface; and estimating, by analyzing the response signal, a value of a substance attribute of the measured object.

[0146] In some implementations, the metasurface attached to the measured object is an active metasurface or a passive metasurface.

[0147] In some implementations, estimating the value of the substance attribute comprises estimating the value of substance attribute by analyzing a frequency response characteristic of the response signal.

[0148] In some implementations, estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonant frequency of the response signal with respective resonant frequencies at different concentrations of the target substance in the measured object; and determining the concentration of the target substance based on the comparison.

[0149] In some implementations, estimating the value of substance attribute by analyzing a frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonant frequency of the response signal with respective resonant frequencies at different compositions of the target substance in the measured object; and determining the composition of the target substance based on the comparison.

[0150] In some implementations, estimating the value of the substance attribute by analyzing the response signal comprising: estimating, based on the response signal, the value of the substance attribute by using a machine learning model.

[0151] In some implementations, the value of the substance attribute comprises an attribute value of : a composition of blood, a salt in the salt solution, or a water contaminant.

[0152] In some implementations, the composition of blood comprises at least one of glucose in blood, salt in blood, or triglycerides in blood.

[0153] In some implementations, the metasurface attached to the measured object is an active metasurface, and the active metasurface comprises: a substrate layer; a metal layer arranged above the substrate layer, the metal layer having one or more capacitors of a variable capacitance.

[0154] In some implementations, the metasurface attached to the measured object is a passive metasurface, and the passive metasurface comprises: a substrate layer; a metal pattern arranged above the substrate layer.

[0155] In one aspect, the present disclosure provides a computer-implemented method. The method comprises: determining a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface based on a measurement environment comprising a metasurface and a measured object; and determining an electrical characteristic of the metasurface by maximizing a sensitivity of the signal characteristic of the response signal to a change in a substance attribute of the measured object.

[0156] In some implementations, the method further comprises: determining a structure of the metasurface based on the electrical characteristic of the metasurface.

[0157] In some implementations, determining the signal characteristic of the response signal from the measured object reflected or transmitted via the metasurface comprises: determining electrical coefficients of the measured object over frequencies at different values of the substance attribute; generating an equivalent circuit representing a plurality of components of the measurement environment, the plurality of components comprising individual layers in the metasurface, the measured object and an environmental object related to the measured object, the plurality of components being represented respectively as circuit elements with respective electrical characteristics in the equivalent circuit; and determining, according to the equivalent circuit, the signal characteristic of the response signal based on respective electrical coefficients of the plurality of components in the equivalent circuit.

[0158] In some implementations, the metasurface is a passive metasurface, and determining the electrical characteristic of the metasurface comprises: determining an optimized capacitance of an active metasurface according to a resonance characteristic of the active metasurface within a capacitance variation range; and determining, based on the optimized capacitance, the electrical characteristic of the passive metasurface. In some implementations, the method further comprises: determining the capacitance variation range of the active metasurface based on the signal characteristic and a parameter value range of an environmental object associated with the measured object.

[0159] In some implementations, determining the capacitance variation range of the active metasurface comprises: determining an impedance variation range of the active metasurface based on the signal characteristic, the measured object and the parameter value range of the environmental object; and deriving, based on the impedance variation range, the capacitance variation range of the active metasurface by using finite element modeling.

[0160] In some implementations, determining the structure of the metasurface comprises: determining, based on the electrical characteristic, a target transmission matrix for the passive metasurface; determining the structure of the passive metasurface by minimizing a difference between a transmission matrix for the passive metasurface with different structures and the target transmission matrix.

[0161] In some implementations, the method further comprises: determining, based on a finite element model of the measurement environment, the signal characteristic of the response signal; and optimizing, based on the finite element model, the electrical characteristic of the metasurface by maximizing a difference of the signal characteristic ofthe response signal among different values of the substance attribute of the measured object.

[0162] In some implementations, the metasurface comprises one or more metal layers and determining the structure of the metasurface comprises: determining, by using a machine learning model, respective structures of the one or more metal layers based on respective electrical characteristics of the one or more metal layers.

[0163] In some implementations, the machine learning model comprises at least one of: a convolutional neural network, a generative adversarial network, or a diffusion model.

[0164] In some implementations, the value of the substance attribute comprises at least one of: a composition of the target substance in the measured object, or a concentration of the target substance.

[0165] In one aspect, the present disclosure provides a device for wireless measurement. The device for wireless measurement comprises: a metasurface attached to a measured object; an emitter configured to emit a probing signal to the measured object; a receiver configured to receive a response signal from the measured object reflected or transmitted via the metasurface; and a processor configured to estimate a value of a substance attribute of the measured object by analyzing the response signal.

[0166] In some implementations, the metasurface attached to the measured object is an active metasurface or a passive metasurface.

[0167] In some implementations, the processor is further configured to: estimate, by analyzing a frequency response characteristic of the response signal, the value of the substance attribute of the measured object.

[0168] In some implementations, the processor is further configured to: determine a resonant frequency of the response signal; compare the resonant frequency of the response signal with respective resonant frequencies at different concentrations of the target substance in the measured object; and determine the concentration of the target substance based on the comparison.

[0169] In some implementations, the processor is further configured to: determine a resonant frequency of the response signal; compare the resonant frequency of the response signal with respective resonant frequencies at different compositions of the target substance in the measured object; and determine a composition of the target substance based on the comparison.

[0170] In some implementations, the processor is further configured to: estimate the value of the substance attribute by using a machine learning model based on the response signal.

[0171] In some implementations, the value of the substance attribute comprises an attribute value of: a composition of blood, a salt in the salt solution, or a water contaminant.

[0172] In some implementations, the composition of blood comprises at least one of: glucose in blood, salt in blood, or triglycerides in blood.

[0173] In some implementations, the metasurface attached to the measured object is an active metasurface, and the active metasurface comprises: a substrate layer; a metal layer arranged above the substrate layer, the metal layer having one or more capacitors of a variable capacitance.

[0174] In some implementations, the metasurface attached to the measured object is a passive metasurface, and the passive metasurface comprises: a substrate layer; a metal pattern arranged above the substrate layer.

[0175] In another aspect, the present disclosure provides an electronic device. The electronic device comprises: a processor; and a memory coupled to the processing unit and having instructions stored thereon that, when executed by the processing unit, cause the device to perform acts comprising: determining, based on a measurement environment comprising a metasurface and an measured object, a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface; determining, an electrical characteristic of the metasurface by maximizing a sensitivity of the signal characteristic of the response signal to a change in a substance attribute of the measured object.

[0176] In some implementations, determining the signal characteristic of the response signal from the measured object reflected or transmitted via the metasurface comprises: determining electrical coefficients of the measured object over frequencies at different values of the substance attribute; generating an equivalent circuit representing a plurality of components of the measurement environment, the plurality of components comprising individual layers in the metasurface, the measured object and an environmental object related to the measured object, the plurality of components being represented respectively as circuit elements with respective electrical characteristics in the equivalent circuit; and determining, according to the equivalent circuit, the signal characteristic of the response signal based on respective electrical coefficients of the plurality of components in the equivalent circuit.

[0177] In some implementations, the metasurface is a passive metasurface, and determining the electrical characteristics of the metasurface comprises: determining an optimized capacitance of an active metasurface according to a resonance characteristic of the active metasurface within a capacitance variation range; and determining, based on the optimized capacitance, the electrical characteristic of the passive metasurface.

[0178] In some implementations, determining the capacitance variation range of the activemetasurface comprises: determining an impedance variation range of the active metasurface based on the signal characteristic, the measured object and the parameter value range of the environmental object; and deriving, based on the impedance variation range, the capacitance variation range of the active metasurface by using finite element modeling.

[0179] In some implementations, determining the structure of the metasurface comprises: determining, based on the electrical characteristic, a target transmission matrix for the passive metasurface; and determining the structure of the passive metasurface by minimizing a difference between a transmission matrix for the passive metasurface with different structures and the target transmission matrix.

[0180] In some implementations, the acts further comprise: determining, based on a finite element model of the measurement environment, the signal characteristic of the response signal; and optimizing, based on the finite element model, the electrical characteristic of the metasurface by maximizing a difference of the signal characteristic of the response signal among different values of the substance attribute of the measured object.

[0181] In some implementations, the metasurface comprises one or more metal layers, and determining the structure of the metasurface comprises: determining, by using a machine learning model, respective structures of the one or more metal layers based on respective electrical characteristics of the one or more metal layers.

[0182] In some implementations, the machine learning model comprises at least one of: a convolutional neural network, a generative adversarial network, or a diffusion model.

[0183] In some implementations, the value of the substance attribute comprises at least one of: a composition of the target substance in the measured object, or a concentration of the target substance.

[0184] In yet another aspect, the present disclosure provides a computer program product tangibly stored in a computer storage media and comprising computer-executable instructions that, when executed by a device, cause the device to perform acts comprising: emitting, by an emitter, a probing signal to a measured object attached with a metasurface; receiving, by a receiver, a response signal from the measured object reflected or transmitted via the metasurface; and estimating, by analyzing the response signal, a value of a substance attribute of the measured object.

[0185] In some implementations, the metasurface attached to the measured object is an active metasurface or a passive metasurface.

[0186] In some implementations, estimating the value of the substance attribute comprises: estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal.

[0187] In some implementations, estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonant frequency of the response signal with respective resonant frequencies at different concentrations of the target substance in the measured object; and determining the concentration of the target substance based on the comparison.

[0188] In some implementations, estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal comprises: determining a resonant frequency of the response signal; comparing the resonance frequency of the response signal with respective resonance frequencies at different compositions of a target substance in the measured object; determining a composition of the target substance based on the comparing.

[0189] In some implementations, estimating the value of the substance attribute by analyzing the response signal comprises: estimating, based on the response signal, the value of the substance attribute by using a machine learning model.

[0190] In some implementations, the value of the substance attribute comprises an attribute value of: a composition of blood, a salt in the salt solution, or a water contaminant.

[0191] In some implementations, the composition of blood comprises glucose in the blood.

[0192] In some implementations, the metasurface attached to the measured object is an active metasurface, and the active metasurface comprises: a substrate layer; a metal layer arranged above the substrate layer, the metal layer having one or more capacitors of a variable capacitance.

[0193] In some implementations, the metasurface attached to the measured object is a passive metasurface, and the passive metasurface comprises: a substrate layer; a metal pattern arranged above the substrate layer.

[0194] In yet another aspect, the present disclosure provides a computer program product tangibly stored in a computer storage media and comprising computer-executable instructions that, when executed by a device, cause the device to perform acts comprising: determining, based on a measurement environment comprising a metasurface and a measured object, a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface; determining an electrical characteristic of the metasurface by maximizing a sensitivity of the signal characteristic of the response signal to a change in a substance attribute of the measured object.

[0195] In some implementations, the acts further comprise determining a structure of the metasurface based on the electrical characteristics of the metasurface.

[0196] In some implementations, determining the signal characteristic of the response signal from the measured object reflected or transmitted via the metasurface comprises: determining electrical coefficients of the measured object over frequencies at different values of the substance attribute; generating an equivalent circuit representing a plurality of components of the measurement environment, the plurality of components comprising individual layers in the metasurface, the measured object and an environmental object related to the measured object, the plurality of components being represented respectively as circuit elements with respective electrical characteristics in the equivalent circuit; determining, according to the equivalent circuit, the signal characteristic of the response signal based on respective electrical coefficients of the plurality of components in the equivalent circuit.

[0197] In some implementations, the metasurface is a passive metasurface, and determining the electrical characteristics of the metasurface comprises: determining the capacitance variation range of the active metasurface based on the signal characteristic and a parameter value range of an environmental object associated with the measured object; determining an optimized capacitance of an active metasurface according to a resonance characteristic of the active metasurface within a capacitance variation range; determining, based on the optimized capacitance, the electrical characteristic of the passive metasurface.

[0198] In some implementations, determining the capacitance variation range of the active metasurface comprises: determining an impedance variation range of the active metasurface based on the signal characteristic, the measured object and the parameter value range of the environmental object; deriving, based on the impedance variation range, the capacitance variation range of the active metasurface by using finite element modeling.

[0199] In some implementations, determining the structure of the metasurface comprises: determining, based on the electrical characteristic, a target transmission matrix for the passive metasurface; determining the structure of the passive metasurface by minimizing a difference between a transmission matrix for the passive metasurface with different structures and the target transmission matrix.

[0200] In some implementations, the method further comprises: determining, based on a finite element model of the measurement environment, the signal characteristic of the response signal; and optimizing, based on the finite element model, the electrical characteristic of the metasurface by maximizing a difference of the signal characteristic of the response signal among different values of the substance attribute of the measured object.

[0201] In some implementations, the metasurface comprises one or more metal layers, and determining the structure of the metasurface comprises: determining, by using a machinelearning model, respective structures of the one or more metal layers based on respective electrical characteristics of the one or more metal layers.

[0202] In some implementations, the machine learning model comprises at least one of: a convolutional neural network, a generative adversarial network, or a diffusion model.

[0203] In some implementations, the value of substance attribute comprises at least one of: a composition of the target substance in the measured object, or a concentration of the target substance.

[0204] In yet another aspect, the present disclosure provides a computer-readable medium with computer-executable instructions stored thereon that, when executed by a device, cause the device to perform one or more example implementations of the method of the above aspects.

[0205] The functions described above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, example types of hardware logic components that may be used comprise: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-a-chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0206] Program code for implementing the methods of the present disclosure may be written by any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program code, when executed by a processor or controller, causes the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code may be executed entirely on a machine, partly on a machine, partly on a machine as a standalone software package and partly on a remote machine or entirely on a remote machine or server.

[0207] In the context of the present disclosure, a machine-readable media may be a tangible media that may contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable mnedia may be a machine-readable signal medium or a machine-readable storage media. The machine- readable media may comprise, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may comprise electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portablecompact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0208] Further, while operations are depicted in a particular order, this should be understood to require such operations to be performed in the particular order shown or in sequential order, or that all illustrated operations should be performed to achieve the desired results. In certain environment, multitasking and parallel processing may be beneficial. Likewise, while several specific implementation details are included in the discussion above, these should not be appreciated as the limitation to the scope of the present disclosure. Certain features described in the context of separate implementations may also be implemented in combination in a single implementation. Conversely, the various features described in the context of a single implementation may also be implemented in multiple implementations separately or in any suitable sub-combination.

[0209] Although the present subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Instead, the specific features and acts described above are merely exemplary forms for implementing the claims.

Claims

CLAIMS 1. A method of wireless measurement, comprising: emitting, by an emitter, a probing signal to a measured object attached with a metasurface; receiving, by a receiver, a response signal from the measured object reflected or transmitted via the metasurface; and estimating, by analyzing the response signal, a value of a substance attribute of the measured object.

2. The method of claim 1, wherein estimating the value of the substance attribute comprises: estimating the value of the substance attribute by analyzing a frequency response characteristic of the response signal.

3. The method of claim 2, wherein estimating the value of substance attribute by analyzing a frequency response characteristic of the response signal comprises: determining a resonance frequency of the response signal; comparing the resonance frequency of the response signal with respective resonance frequencies at different compositions of a target substance in the measured object; and determining a composition of the target substance based on the comparing.

4. The method of claim 2, wherein estimating the value of the substance attribute by analyzing the response signal comprising: estimating, based on the response signal, the value of the substance attribute by using a machine learning model.

5. The method of claim 1, wherein the metasurface attached to the measured object is an active metasurface, and the active metasurface comprises: a substrate layer; and a metal layer arranged above the substrate layer, the metal layer having one or more capacitors of a variable capacitance.

6. The method of claim 1, wherein the metasurface attached to the measured object is a passive metasurface, and the passive metasurface comprises: a substrate layer; and a metal pattern arranged above the substrate layer.

7. A method of metasurface optimization, comprising: determining, based on a measurement environment comprising a metasurface and a measured object, a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface; and determining an electrical characteristic of the metasurface by maximizing a sensitivity of the signal characteristic of the response signal to a change in a substance attribute of the measuredobject.

8. The method of claim 7, further comprising: determining a structure of the metasurface based on the electrical characteristic of the metasurface.

9. The method of claim 7, wherein determining a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface comprises: determining electrical coefficients of the measured object over frequencies at different values of the substance attribute; generating an equivalent circuit representing a plurality of components of the measurement environment, the plurality of components comprising individual layers in the metasurface, the measured object and an environmental object related to the measured object, the plurality of components being represented respectively as circuit elements with respective electrical characteristics in the equivalent circuit; and determining, according to the equivalent circuit, the signal characteristic of the response signal based on respective electrical coefficients of the plurality of components in the equivalent circuit.

10. The method of the claim 7, wherein the metasurface is a passive metasurface and determining an electrical characteristic of the metasurface comprises: determining an optimized capacitance of an active metasurface according to a resonance characteristic of the active metasurface within a capacitance variation range; and determining, based on the optimized capacitance, the electrical characteristic of the passive metasurface.

11. The method of claim 10, further comprising: determining the capacitance variation range of the active metasurface based on the signal characteristic and a parameter value range of an environmental object associated with the measured object.

12. The method of claim 11, wherein determining the capacitance variation range of the active metasurface comprises: determining an impedance variation range of the active metasurface based on the signal characteristic, the measured object and the parameter value range of the environmental object; and deriving, based on the impedance variation range, the capacitance variation range of the active metasurface by using finite element modeling.

13. The method of claim 8, wherein the metasurface is a passive metasurface and determining the structure of the metasurface comprises: determining, based on the electrical characteristic, a target transmission matrix for the passivemetasurface; and determining the structure of the passive metasurface by minimizing a difference between a transmission matrix for the passive metasurface with different structures and the target transmission matrix.

14. The method of claim 7, further comprising: determining, based on a finite element model of the measurement environment, the signal characteristic of the response signal; and optimizing, based on the finite element model, the electrical characteristic of the metasurface by maximizing a difference of the signal characteristic of the response signal among different values of the substance attribute of the measured object.

15. An electrical device, comprising: a processing unit; and a memory coupled to the processing unit and having instructions stored thereon that, when executed by the processing unit, cause the device to perform acts comprising: determining, based on a measurement environment comprising a metasurface and a measured object, a signal characteristic of a response signal from the measured object reflected or transmitted via the metasurface; and determining an electrical characteristic of the metasurface by maximizing a sensitivity of the signal characteristic of the response signal to a change in a substance attribute of the measured object.

Images