Low-power-consumption-oriented reconfigurable intelligent metasurface codebook optimization method

Abstract

The invention provides a low-power-consumption-oriented reconfigurable intelligent metasurface codebook optimization method, and provides a solution for rapid iterative optimization through a field source on the basis of an equivalent surface source theory by constructing a dynamic energy consumption model of a metasurface reconfigurable element. The method comprises the following steps: firstly, establishing a dynamic energy consumption model according to physical parameters of a metasurface element, and describing the energy consumption condition of the metasurface element; secondly, converting the energy consumption model into an equivalent surface source optimization problem; then, constructing a transformation pair of an equivalent source and a radiation field; and finally, the optimal source distribution is obtained by using the rapid alternate projection of the field source, so that the dynamic optimization of the energy consumption is realized on the premise of ensuring the communication performance. The method has the advantages of high speed and high flexibility, and is suitable for optimization design of a large-scale intelligent metasurface system.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a codebook optimization method for a large-scale reconfigurable intelligent metasurface array with low power consumption. Background Art

[0002] In modern communications, radar, the Internet of Things and other fields, with the continuous expansion of system scale and the increase in demand, traditional antennas and RF systems are facing increasingly severe challenges in energy efficiency and performance optimization. In order to meet the needs of high performance, low power consumption and flexibility, the emerging technology of intelligent metasurface (Reconfigurable Intelligent Surface, RIS) has begun to attract widespread attention. By introducing a controllable phase control unit on the propagation path of electromagnetic waves, the intelligent metasurface can flexibly adjust the reflection direction, amplitude and phase of electromagnetic waves, thereby significantly improving the communication capability and energy efficiency of the system.

[0003] However, as the scale of intelligent metasurfaces increases, the power consumption of the system also increases exponentially. Especially in the application scenarios of dynamic control and real-time adjustment, how to effectively reduce the energy consumption of the system has become one of the key issues that need to be solved urgently. Existing intelligent metasurface technology mainly focuses on improving system performance, and the research on power consumption optimization is relatively limited. Especially in large-scale intelligent metasurfaces, the power consumption management problem is more complicated. On the one hand, intelligent metasurfaces need to continuously adjust the working state of their surface units to adapt to changes in the external environment; on the other hand, how to achieve dynamic optimization of energy consumption while ensuring communication performance is the focus and difficulty of current research.

[0004] Some current studies have proposed energy consumption optimization schemes based on fixed algorithms, which control the state of each metasurface unit through pre-set strategies and reduce the energy consumption of some areas through partition control. However, these methods have limited effects in large-scale application scenarios and cannot flexibly respond to complex external environments and real-time communication needs. Therefore, there is an urgent need for a new method that can achieve dynamic energy consumption optimization in the context of large-scale reconfigurable smart metasurfaces, thereby improving system performance while reducing power consumption and extending the service life of the device. Summary of the invention

[0005] The technical scheme of the present invention provides a solution that is significantly different from the prior art to the problems of excessive energy consumption and difficulty in flexible regulation in large-scale reconfigurable intelligent metasurface systems. It mainly provides a dynamic energy consumption model of a reconfigurable intelligent metasurface, which is based on surface source equivalence and introduces field source rapid iteration to achieve source distribution with minimum energy consumption, thereby achieving dynamic optimization of energy consumption while ensuring communication performance.

[0006] The specific technical solution provided by the present invention is a codebook optimization method for a reconfigurable intelligent metasurface with low power consumption, comprising the following steps:

[0007] S1. Construct an array dynamic energy consumption model based on the physical parameters of the metasurface reconfigurable elements to describe the energy consumption of the metasurface under different working states and physical parameters;

[0008] S2. Based on the equivalent surface source theory, the dynamic energy consumption model of the reconfigurable intelligent metasurface array is converted into an optimization problem of the equivalent surface source;

[0009] S3, construct transformation pairs of equivalent sources and radiation fields;

[0010] S4, based on the rapid alternating projection of the field source, the optimal source distribution is achieved, thereby ensuring the minimum energy consumption under the premise of ensuring the same radiation field.

[0011] Further, the step S1 includes: describing the physical parameters of the metasurface tunable element, establishing a mathematical model of unit energy consumption, and establishing a total energy consumption model of the array;

[0012] Furthermore, the physical state of the metasurface reconfigurable elements directly affects the power consumption and operating performance;

[0013] Optionally, the reconfigurable element of the metasurface can change the direction of the electromagnetic wave by phase control. The phase adjustment range of each element is between 0 and 2π. The phase control can be achieved by a PIN diode or a varactor diode. The accuracy of the phase adjustment affects the energy consumption of the reconfigurable element.

[0014] Optionally, the reconfigurable element of the metasurface can achieve different reflection beam adjustment effects through amplitude regulation, and the amplitude regulation can be achieved through a PIN diode or a varactor diode, and the accuracy of the amplitude regulation affects the energy consumption of the reconfigurable element;

[0015] Furthermore, the energy consumption of the metasurface unit reconfigurable element is positively correlated with the physical parameters;

[0016] Optionally, the unit energy consumption of the phase control element can be expressed as:

[0017] P i =f θ (θ i ) (1)

[0018] where θ i is the phase adjustment state of the i-th unit reconfigurable element;

[0019] Optionally, the unit energy consumption of the amplitude control element can be expressed as:

[0020] P i =f A (Ai ) (2)

[0021] Among them A i is the amplitude adjustment state of the i-th unit reconfigurable element;

[0022] Furthermore, a large-scale smart metasurface array contains N units, so the total energy consumption of the entire array is the accumulation of the energy consumption of all units:

[0023]

[0024] Furthermore, as the environment and communication requirements change, the phase and amplitude of each unit of the metasurface will change, and the array power consumption will also change dynamically. By adjusting different physical parameters in real time, dynamic energy consumption optimization under different communication requirements can be achieved;

[0025] Furthermore, step S2 describes the response of each unit on the metasurface to the incident electromagnetic wave through the equivalent surface current density and the equivalent surface magnetic flux density, thereby controlling the physical parameters of the reflected electromagnetic wave, including the phase and amplitude. The equivalent surface source directly describes the response characteristics of the metasurface to the electromagnetic wave. By optimizing the equivalent source, the energy consumption of the entire system can be indirectly optimized.

[0026] Optionally, for a phase reconfigurable metasurface, the phase θ i Corresponding to the equivalent current source J i The phase shift of each unit can be expressed as:

[0027]

[0028] Phase adjustment energy consumption P θ With θ i The adjustment state is related to the adjustment state, by adjusting the phase θ of the equivalent current source i , the energy consumption generated by phase adjustment can be optimized, and the corresponding optimization goal is:

[0029]

[0030] Optionally, for the amplitude reconfigurable metasurface, the equivalent current source J i Amplitude |J i | is proportional to the amplitude of the reflected wave, the amplitude adjustment energy consumption is related to the amplitude of the equivalent current source, and the optimization goal is:

[0031]

[0032] Further, the step S3 includes: constructing a radiation field of the equivalent source, and reconstructing the equivalent source based on the radiation field;

[0033] Furthermore, the step of constructing the radiation field of the equivalent source is:

[0034] Furthermore, assuming that the metasurface S is located in the xy plane of the Cartesian coordinate system and satisfies z = 0, considering the Kirchhoff boundary condition, the radiation field of the metasurface S at any position (x, y, z) in free space is:

[0035]

[0036] Among them, J(x i ,y i ) is (x i ,y i ) is the equivalent current source of the metasurface unit, is the three-dimensional Green's function of free space, is the distance from the observation point (x, y, z) to the hypersurface unit (x i ,y i ) distance, k is the spatial wave number;

[0037] Furthermore, the first Rayleigh-Sommerfeld diffraction integral can be further derived based on the second Green's function, and considering the Kirchhoff approximation, formula (7) can be simplified to:

[0038]

[0039] Where λ is the wavelength;

[0040] Furthermore, considering the Francois Hoffer far-field approximation, formula (8) can be simplified to:

[0041]

[0042] Furthermore, formula (9) is converted into a spatial angle domain expression:

[0043]

[0044] Among them, θ is the pitch angle of the far-field radiation field, φ is the direction angle of the far-field radiation field, and R is the distance of the far-field observation point relative to the metasurface;

[0045] Furthermore, the step of reconstructing the equivalent source based on the radiation field is:

[0046] Furthermore, consider the stationary phase method to obtain the spectrum approximation of the far field:

[0047]

[0048] Among them, F(k x ,k y ) is the spectrum of the plane wave, k x , k y are the components of the spatial wave number k on the coordinate axis respectively;

[0049] Furthermore, combined with the plane wave spectrum, the equivalent surface source can be obtained from the far-field pattern:

[0050]

[0051] Furthermore, formula (10) and formula (12) constitute a transformation pair of equivalent source and radiation field;

[0052] Further, the step S4 includes:

[0053] Step 1, initialization parameters: maximum number of iterations iter max , the current iteration number is iter, determine the physical parameter T that contributes the least to energy consumption, the proportion factor k, and the expected field distribution Minimum root mean square error σ min ;

[0054] Optionally, for the phase-controlled metasurface, the physical parameter that contributes the least to energy consumption is the phase control amount θ that produces the minimum energy consumption min ;

[0055] Optionally, for the amplitude-controlled metasurface, the physical parameter that contributes the least to energy consumption is the amplitude-controlled amount A that produces the minimum energy consumption. min ;

[0056] Furthermore, the proportion factor is the number N of supersurface units that satisfy the physical parameter quantity T. T The ratio of the total number of units

[0057] Step 2: Randomly initialize the equivalent surface source J(x i ,y i );

[0058] Step 3: Calculate the far-field radiation pattern E of the equivalent surface source according to formula (10): far (θ,φ);

[0059] Step 4: Convert the far-field radiation pattern E far The modulus of (θ, φ) is replaced by the expected field The modulus of

[0060] Step 5, using formula (12) to inversely deduce the surface equivalent surface source from the corrected far-field radiation pattern;

[0061] Step 6, according to the proportion factor k, select N×k units in the metasurface equivalent surface source whose physical parameters are close to T, and correct their physical parameters to T;

[0062] Step 7: Calculate the far-field radiation pattern E using formula (10) for the corrected surface equivalent source far(θ,φ) and calculate the RMS error from the expected far-field radiation pattern:

[0063]

[0064] Step 8, if σ>σ min , then repeat steps 4 to 7 until the minimum RMS error condition is met or the maximum number of iterations is met, the algorithm terminates, and a larger proportion factor can be updated to restart the algorithm.

[0065] Beneficial Effects

[0066] As can be seen from the above, the beneficial effects of the present invention are:

[0067] (1) In existing large-scale reconfigurable intelligent metasurface systems, energy consumption increases exponentially as the scale of the system expands. The present invention optimizes the energy consumption of the system by introducing a dynamic energy consumption model and using the concept of equivalent surface sources. Through the fast iteration algorithm of the field source, the equivalent source distribution with the minimum energy consumption can be found under different working conditions, thus avoiding unnecessary energy consumption in traditional systems.

[0068] (2) The present invention significantly simplifies the energy consumption control process of large-scale metasurfaces through an optimization model based on equivalent sources, thereby significantly improving the scalability and deployment efficiency of the system.

[0069] (3) The present invention introduces a dynamic control mechanism and combines it with an equivalent source model for rapid iterative optimization. It can adjust the configuration of the metasurface system according to changes in communication requirements and the external environment, and has strong versatility and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a flow chart of a codebook optimization method for a reconfigurable intelligent metasurface with low power consumption;

[0071] Figure 2 Schematic diagram of unit energy consumption of a PIN diode in an embodiment of the present invention;

[0072] Figure 3 Schematic diagram of unit energy consumption of a varactor diode in an embodiment of the present invention;

[0073] Figure 4 is a schematic diagram of the equivalent surface source of the metasurface;

[0074] Figure 5 It is a schematic diagram of the alternating iteration of the equivalent source and field of the metasurface;

[0075] Figure 6 It is a flow chart of energy consumption optimization algorithm. DETAILED DESCRIPTION

[0076] In order to clarify the purpose, implementation scheme and technical advantages of the present invention, the present invention is further described in detail below in combination with specific examples and with reference to the accompanying drawings. Several embodiments of the present invention are given in the accompanying drawings, but the present invention can be implemented in different forms and is not limited to the embodiments described in the text. On the contrary, these embodiments are provided to make the content disclosed in the present invention more thorough and comprehensive.

[0077] Figure 1 The present invention is a flow chart of a codebook optimization method for a reconfigurable intelligent metasurface with low power consumption, including:

[0078] Step S1: constructing an array dynamic energy consumption model according to the physical parameters of the metasurface reconfigurable elements to describe the energy consumption of the metasurface under different working states and physical parameters;

[0079] Step S2: Based on the equivalent surface source theory, the dynamic energy consumption model of the reconfigurable intelligent metasurface array is converted into an optimization problem of the equivalent surface source;

[0080] Step S3: constructing a transformation pair of equivalent source and radiation field;

[0081] Step S4: Optimal source distribution is achieved based on rapid alternating projections of field sources, thereby achieving minimum energy consumption under the premise of ensuring the same radiation field.

[0082] The step S1 includes: describing the physical parameters of the metasurface tunable element, constructing the basic assumptions of the energy consumption model, establishing a mathematical model of unit energy consumption, and establishing a total energy consumption model of the array;

[0083] The physical state of metasurface reconfigurable components directly affects power consumption and operating performance;

[0084] In some optional embodiments, the reconfigurable element of the metasurface can change the direction of the electromagnetic wave by phase control, and the phase adjustment range of each element is between 0 and 2π. The phase control can be achieved by a PIN diode or a varactor diode, and the accuracy of the phase adjustment affects the energy consumption of the reconfigurable element.

[0085] In some optional embodiments, the reconfigurable element of the metasurface can achieve different reflected beam adjustment effects through amplitude control, and the amplitude control can be achieved through a PIN diode or a varactor diode, and the accuracy of the amplitude control affects the energy consumption of the reconfigurable element;

[0086] Furthermore, the energy consumption of metasurface reconfigurable elements is positively correlated with the physical parameters;

[0087] The unit energy consumption of the phase control element can be expressed as:

[0088] P i =f θ (θ i) (1)

[0089] where θ i is the phase adjustment state of the i-th unit reconfigurable element;

[0090] The unit energy consumption of the amplitude control element can be expressed as:

[0091] P i =f A (A i ) (2)

[0092] Among them A i is the amplitude adjustment state of the i-th unit reconfigurable element;

[0093] In some optional embodiments, the control element can be a PIN diode. The energy consumption of the PIN diode is mainly conduction loss. When the PIN diode works under forward bias, the current passes through the P-type and N-type regions, and the current is conducted in the intrinsic region (I region) in a drifting manner. The resistance of the I region is large, and the drift speed of electrons and holes in this region is slow, resulting in a certain amount of power loss. The energy consumption is mainly reflected in the resistive loss, that is, the power loss when the current passes through the resistor, which is usually expressed as P=I 2 The form of R;

[0094] Under zero bias conditions, since no forward or reverse voltage is applied, the carriers in the PIN diode are not driven and cannot flow across the PN junction region. There is no significant current passing through the diode at zero bias, and the energy consumption is close to zero. Therefore, when the control element is a PIN diode, the energy consumption is Figure 2 As shown;

[0095] When the PIN diode adjusts the phase, it satisfies:

[0096]

[0097] When the PIN diode adjusts the amplitude, it satisfies:

[0098]

[0099] Among them, P ON is the energy consumption when the PIN diode is turned on;

[0100] In some optional embodiments, the regulating element may be a varactor diode. The energy consumption of the varactor diode is mainly the leakage current loss of the reverse bias. In the reverse bias state, the current is ideally almost zero and the energy consumption should also be very small. However, due to the limitations of materials and manufacturing processes, the actual varactor diode will have a small amount of reverse leakage current. When these leakage currents pass through the PN junction area, a small amount of power consumption will be generated, which is usually manifested as a tiny resistive loss, expressed as: P = V × I, where V is the reverse bias voltage and I is the reverse leakage current;

[0101] When the adjustable element is a varactor diode, the energy consumption is Figure 3 As shown, f θ (θ i ) is linearly related to the phase, f A (A i ) is linearly related to the amplitude;

[0102] Furthermore, a large-scale smart metasurface array contains N units, so the total energy consumption of the entire array is the superposition of the energy consumption of all units:

[0103]

[0104] As the environment and communication requirements change, the phase and amplitude of each unit of the metasurface will change, and the array power consumption will also change dynamically. Dynamic energy consumption optimization under different communication requirements can be achieved by adjusting different physical parameters in real time.

[0105] Furthermore, the step S2 describes the response of each unit on the metasurface to the incident electromagnetic wave by using the equivalent surface current density and the equivalent surface magnetic flux density, thereby controlling the physical parameters of the reflected electromagnetic wave, including the phase and amplitude, such as Figure 4 As shown in the figure, because the equivalent surface source directly describes the response characteristics of the metasurface to electromagnetic waves, the energy consumption of the entire system can be indirectly optimized by optimizing the equivalent source;

[0106] Optionally, for a phase reconfigurable metasurface, the phase θ i Corresponding to the equivalent current source J i The phase shift of each unit can be expressed as:

[0107]

[0108] Phase adjustment energy consumption P θ With θ i The adjustment state is related to the adjustment state, by adjusting the phase θ of the equivalent current source i , the energy consumption generated by phase adjustment can be optimized, and the corresponding optimization goal is:

[0109]

[0110] Optionally, for the amplitude reconfigurable metasurface, the equivalent current source J i Amplitude |J i | is proportional to the amplitude of the reflected wave, the amplitude adjustment energy consumption is related to the amplitude of the equivalent current source, and the optimization goal is:

[0111]

[0112] Further, the step S3 includes: constructing a radiation field of the equivalent source, and reconstructing the equivalent source based on the radiation field;

[0113] Furthermore, the step of constructing the radiation field of the equivalent source is:

[0114] Furthermore, assuming that the metasurface S is located in the xy plane of the Cartesian coordinate system and satisfies z = 0, considering the Kirchhoff boundary condition, the radiation field of the metasurface S at any position (x, y, z) in free space is:

[0115]

[0116] Among them, J(x i ,y i ) is (x i ,y i ) is the equivalent current source of the metasurface unit, is the three-dimensional Green's function of free space, is the distance from the observation point (x, y, z) to the hypersurface unit (x i ,y i ) distance, k is the spatial wave number;

[0117] Furthermore, the first Rayleigh-Sommerfeld diffraction integral can be further derived based on the second Green's function, and considering the Kirchhoff approximation, formula (9) can be simplified to:

[0118]

[0119] Where λ is the wavelength;

[0120] Furthermore, considering the Francois Hoffer far-field approximation, formula (10) can be simplified to:

[0121]

[0122] Furthermore, formula (11) is converted into a spatial angle domain expression:

[0123]

[0124] Among them, θ is the pitch angle of the far-field radiation field, φ is the direction angle of the far-field radiation field, and R is the distance of the far-field observation point relative to the metasurface;

[0125] Furthermore, the step of reconstructing the equivalent source based on the radiation field is:

[0126] Furthermore, consider the stationary phase method to obtain the spectrum approximation of the far field:

[0127]

[0128] Among them, F(k x ,k y ) is the spectrum of the plane wave, k x , k y are the components of the spatial wave number k on the coordinate axis respectively;

[0129] Furthermore, combined with the plane wave spectrum, the equivalent surface source can be obtained from the far-field pattern:

[0130]

[0131] Furthermore, formula (12) and formula (14) constitute a transformation pair of equivalent source and radiation field, such as Figure 5 As shown;

[0132] Further, the step S4 is as follows Figure 6 As shown, including:

[0133] Step 1, initialization parameters: maximum number of iterations iter max , the current iteration number is iter, determine the physical parameter T that contributes the least to energy consumption, the proportion factor k, and the expected field distribution Minimum root mean square error σ min ;

[0134] Optionally, for the phase-controlled metasurface, the physical parameter that contributes the least to energy consumption is the phase control amount θ that produces the minimum energy consumption min ;

[0135] Optionally, for the amplitude-controlled metasurface, the physical parameter that contributes the least to energy consumption is the amplitude-controlled amount A that produces the minimum energy consumption. min ;

[0136] Furthermore, the proportion factor is the number N of supersurface units that satisfy the physical parameter quantity T. T The ratio of the total number of units

[0137] Step 2: Randomly initialize the equivalent surface source J(x i ,y i );

[0138] Step 3: Calculate the far-field radiation pattern E of the equivalent surface source according to formula (12): far (θ,φ);

[0139] Step 4: Convert the far-field radiation pattern E far The modulus of (θ, φ) is replaced by the expected field The modulus of

[0140] Step 5, using formula (14) to inversely deduce the surface equivalent surface source from the corrected far-field radiation pattern;

[0141] Step 6, according to the proportion factor k, select N×k units in the metasurface equivalent surface source whose physical parameters are close to T, and correct their physical parameters to T;

[0142] Step 7: Calculate the far-field radiation pattern E using formula (12) for the corrected surface equivalent source far (θ,φ) and calculate the RMS error from the expected far-field radiation pattern:

[0143]

[0144] Step 8, if σ>σ min , then repeat steps 4 to 7 until the minimum RMS error condition is met or the maximum number of iterations is met, the algorithm terminates, and a larger proportion factor can be updated to restart the algorithm.

Claims

1. A codebook optimization method for a low-power reconfigurable intelligent metasurface, characterized in that: The steps include: S1. Construct an array dynamic energy consumption model based on the physical parameters of the metasurface reconfigurable elements to describe the energy consumption of the metasurface under different working states and physical parameters; S2. Based on the equivalent surface source theory, the dynamic energy consumption model of the reconfigurable intelligent metasurface array is converted into an optimization problem of the equivalent surface source; S3, construct transformation pairs of equivalent sources and radiation fields; S4, based on the rapid alternating projection of the field source, the optimal source distribution is achieved, thereby ensuring the minimum energy consumption under the premise of ensuring the same radiation field.

2. The method according to claim 1, characterized in that The step S1, constructing an array dynamic energy consumption model according to the physical parameters of the metasurface reconfigurable element, describing the energy consumption of the metasurface under different working states and physical parameters, includes: Describe the physical parameters of the metasurface tunable elements, establish a mathematical model of unit energy consumption, and establish a total energy consumption model of the array.

3. The method according to claim 2, characterized in that: The physical parameters describing the metasurface tunable element are expressed as follows: the physical state of the metasurface reconfigurable element directly affects the energy consumption and working performance; The mathematical model of the unit energy consumption is expressed as follows: the energy consumption of the metasurface unit reconfigurable element is positively correlated with the regulated physical parameters; The total energy consumption model of the array is expressed as follows: the total energy consumption of the array is the accumulation of the energy consumption of all units.

4. The method according to claim 1, characterized in that: The step S2, based on the equivalent surface source theory, converts the dynamic energy consumption model of the reconfigurable intelligent metasurface array into an optimization problem of the equivalent surface source, comprises: The response of each unit on the metasurface to the incident electromagnetic wave is described by the equivalent surface current density and the equivalent surface magnetic current density, thereby controlling the physical parameters of the reflected electromagnetic wave, including phase and amplitude. The equivalent surface source directly describes the response characteristics of the metasurface to the electromagnetic wave. By optimizing the equivalent source, the energy consumption of the entire system can be indirectly optimized. The control phase θ of the i-th unit i Corresponding to the equivalent current source J i The phase shift of each unit can be expressed as: Phase Adjustment Energy Consumption vs. θ i The adjustment state is related to the adjustment state, by adjusting the phase θ of the equivalent current source i , the energy consumption generated by phase adjustment can be optimized; Equivalent current source J i Amplitude |J i |It is proportional to the amplitude of the reflected wave, and the amplitude control energy consumption of the metasurface unit is related to the amplitude of the equivalent current source.

5. The method according to claim 1, characterized in that The step S3, constructing a transformation pair of equivalent source and radiation field, comprises: Construct the radiation field of the equivalent source; Reconstruct the equivalent source based on the radiation field.

6. The method according to claim 5, characterized in that The construction of the radiation field of the equivalent source includes: According to the Hough boundary condition, the radiation field of the metasurface S at any position (x, y, z) in free space is: Among them, J(x i ,y i ) is (x i ,y i ) is the equivalent current source of the metasurface unit at the position, is the Green's function of free space, is the distance from the observation point (x, y, z) to the hypersurface unit (x i ,y i ), k is the spatial wave number, and n is the normal unit vector of the metasurface; Furthermore, the first Rayleigh-Sommerfeld diffraction integral can be further derived based on the second Green's function, and considering the Kirchhoff approximation, it is converted into a spatial angle domain expression, and the free space radiation field is: Among them, θ is the pitch angle of the far-field radiation field, φ is the direction angle of the far-field radiation field, and R is the distance of the far-field observation point relative to the metasurface.

7. The method according to claim 5, characterized in that The method of reconstructing an equivalent source based on a radiation field comprises: The far-field spectrum is approximated using the stationary phase method: Among them, F(k x ,k y ) is the spectrum of the plane wave, k x , k y are the components of the spatial wave number k on the coordinate axis respectively; Furthermore, combined with the plane wave spectrum, the equivalent surface source can be obtained from the far-field pattern: Among them, E far (θ, φ) is the far-field radiation pattern.

8. The method according to claim 1, characterized in that The step S4, achieving optimal source distribution based on rapid alternating projection of the field source, thereby achieving minimum energy consumption under the premise of ensuring the same radiation field, includes: Step 1: Initialize parameters, the maximum number of iterations is iter max , the current iteration number is iter, determine the physical parameter T that contributes the least to energy consumption, the proportion factor k, and the expected field distribution Minimum root mean square error σ min ; Step 2: Randomly initialize the equivalent surface source J(x i ,y i ); Step 3: Calculate the far-field radiation pattern E of the equivalent surface source far (θ,φ); Step 4: Convert the far-field radiation pattern E far The modulus of (θ, φ) is replaced by the expected field The modulus of Step 5: Inversely deduce the equivalent surface source from the corrected far-field radiation pattern; Step 6: According to the proportion factor k, select N×k units in the metasurface equivalent surface source whose physical parameters are close to T, and correct their physical parameters to T; Step 7: Calculate the far-field radiation pattern E for the corrected surface equivalent source far (θ,φ) and calculate the RMS error from the expected far-field radiation pattern: Step 8: If σ>σ min , then repeat steps 4 to 7 until the minimum root mean square error condition is met or the maximum number of iterations is met, and the algorithm terminates.

Images