Nanosctructure scatterometry scheme determination method and system

Abstract

The present application relates to the technical field of semiconductor measurement, and particularly relates to a kind of nanostructure scattering measurement scheme determination method and system, determine target structure and its target parameter, and the candidate measurement condition of scattering measurement for target structure;According to the candidate measurement condition, forward for target structure, determine the sensitivity of the measurement signal of the band of forward to target structure;Determine the noise model in the scattering measurement process, and the corresponding extended information matrix is obtained when the forward is carried out in the candidate measurement condition in combination with sensitivity;Edge is solved to the extended information matrix, and the edge covariance matrix corresponding to the candidate measurement condition is obtained;The candidate measurement condition is sorted in combination with edge covariance matrix;According to the sorting result, determine the candidate measurement condition of scattering measurement for target structure, form the recommended measurement scheme for target structure.The present application can automatically generate scattering measurement scheme, reduce artificial experience dependence.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor measurement technology, and particularly relates to a method and system for determining a nanostructure scattering measurement scheme. Background Technology

[0002] Three-dimensional nanostructures in advanced integrated circuit manufacturing require simultaneous measurement of parameters such as critical dimensions, structural height, sidewall angles, film thickness, and process drift. While traditional optical scattering measurements offer advantages such as non-destructive nature, high speed, and online application, they are limited by wavelength, numerical aperture, and structural complexity at advanced nodes, and some structural parameters exhibit strong correlations.

[0003] Extreme ultraviolet (EUV) and soft X-ray bands, with their shorter wavelengths, can provide higher spatial frequency information for advanced process nanostructures. EUV / SXR scattering measurements can utilize 0th-order reflection signals or non-zero-order diffraction signals such as ±1st and ±2nd orders. Different wavelengths, incident angles, azimuth angles, polarization, and diffraction orders exhibit varying sensitivities to target structural parameters and are affected by instrument disturbances such as light source spectral line intensity drift, incident angle deviation, detector gain / background, and carbon contamination.

[0004] Current measurement scheme selection usually relies on human experience, single-point sensitivity observation or simple signal strength comparison, making it difficult to answer the following questions simultaneously: (1) Which wavelength-angle-order combinations can simultaneously distinguish critical size CD, height and sidewall angle; (2) Is the 0th order signal suitable for fast screening or can it support precise measurement; (3) Is it worthwhile to increase the measurement time cost for non-zero order diffraction signals; (4) After adding instrument disturbances such as light source drift, angle deviation, background and contamination, is the original optimal measurement scheme still optimal; (5) How to provide auditable uncertainty, parameter correlation and warning signs.

[0005] Therefore, there is a need for an automatic optimization method for measurement schemes that can uniformly incorporate target structure sensitivity, instrument disturbance sensitivity, measurement noise, and time cost. Summary of the Invention

[0006] In view of this, the present invention aims to provide a method and system for determining nanostructure scattering measurement schemes, which can automatically generate EUV / SXR scattering measurement schemes. It unifies the target parameter sensitivity and instrument perturbation sensitivity for measurement scheme generation, so that the instrument status can be taken into account during the selection stage of the measurement scheme. This solves the problem that the optimal measurement scheme may fail under instrument perturbation when only the target parameter is considered, thereby providing a more suitable measurement scheme.

[0007] To achieve the above objectives, the technical solution created by this invention is implemented as follows: A method for determining a nanostructure scattering measurement scheme includes: S1: Determine the target structure to be measured and its target parameters, as well as the candidate measurement conditions for scattering measurement of the target structure; S2: Perform forward modeling of the target structure according to the candidate measurement conditions in step S1 to obtain the measurement signal; determine the sensitivity of the measurement signal to the disturbances generated by the target parameters and the instruments involved in the measurement; S3: Based on the measurement signal in step S2, determine the noise model in the scattering measurement process, and combine it with the sensitivity obtained in step S2 to obtain the extended information matrix of the target structure when performing forward modeling with the candidate measurement conditions in step S1. S4: Based on the extended information matrix obtained in step S3, obtain the marginal covariance of the target parameters; sort the candidate measurement conditions according to the marginal covariance; S5: Based on the sorting results of step S4, determine the candidate measurement conditions for scattering measurement of the target structure and form a recommended measurement scheme for the target structure.

[0008] Furthermore, in step S2, the target structure is modeled using an electromagnetic forward model to obtain a measurement signal including a 0th-order diffraction signal and at least one non-zero-order diffraction signal.

[0009] Furthermore, in step S2, determining the sensitivity of the measurement signal to the target structure includes: performing differential calculation on the measurement signal with respect to the target parameters to obtain the first sensitivity of the measurement signal to the target parameters; and performing differential calculation on the measurement signal with respect to the natural disturbances caused by the instruments involved in the scattering measurement and the environment to the scattering measurement to obtain the second sensitivity of the measurement signal to the natural disturbances.

[0010] Furthermore, the noise model in step S3 is as follows: σ i 2 =(r i ×y i ) 2 +y i / N i +σ f,i 2 ; Where, σ i 2 r represents the total noise standard deviation of the i-th observation channel in the detector receiving the measurement signal. i y represents the relative noise figure of the i-th observation channel. i N represents the measurement signal received by the i-th observation channel. i σ represents the equivalent photon number or integral scale in the i-th observation channel. f,i This represents the background noise of the i-th observation channel in the detector.

[0011] Furthermore, the process of obtaining the extended information matrix in step S3 includes: integrating the first sensitivity and the second sensitivity to obtain a preliminary extended matrix; and obtaining the extended information matrix using the following formula: F=J T Σ -1 J+diag(0,P); Where F represents the extended information matrix, J represents the initial extended matrix, Σ represents the noise covariance matrix constructed based on the noise model, P represents the prior information matrix of natural disturbances, is the inverse matrix of the prior covariance of the instrument disturbance parameters, and diag represents the function that creates the diagonal of the matrix.

[0012] Furthermore, in step S4: when the measurement signal only includes the 0th-order diffraction signal, the measurement time of the candidate measurement conditions in the scattering measurement process is calculated; when the measurement signal only includes the non-zero-order diffraction signal, the candidate measurement conditions related to the geometric dimensions in the target parameters are weighted down; when the measurement signal includes both the 0th-order diffraction signal and the non-zero-order diffraction signal, the matrix geometric properties of the edge covariance matrix are calculated, and the matrix geometric properties corresponding to all candidate measurement conditions are sorted. This sorting result is the measurement uncertainty when the target structure is scattered using the candidate measurement conditions.

[0013] Furthermore, step S4 also includes: calculating the rank or minimum singular value of all marginal covariance matrices; when the rank is less than the number of target parameters, or the minimum singular value is lower than a preset threshold, the corresponding candidate measurement conditions are excluded or marked as incompletely identifiable; adding artificial perturbation to step S2, and repeating steps S2 to S4 to obtain the sorting result of the candidate measurement conditions after adding perturbation.

[0014] This invention also provides a system for determining a nanostructure scattering measurement scheme. According to the method provided by this invention, the system includes: a target structure input module for determining the target structure to be measured and its target parameters; a candidate measurement scheme generation module for determining candidate measurement conditions for scattering measurements of the target structure; a forward modeling module for performing a forward modeling of the target structure according to the candidate measurement conditions to obtain a measurement signal; a sensitivity calculation module for determining the sensitivity of the measurement signal to disturbances generated by the target parameters and the instruments involved in the measurement; an extended information matrix construction module for determining a noise model in the scattering measurement process based on the measurement signal, and combining it with the sensitivity to obtain the extended information matrix corresponding to the forward modeling under the candidate measurement conditions; a measurement scheme ranking module for marginalizing the extended information matrix to obtain the marginal covariance matrix corresponding to the candidate measurement conditions; ranking the candidate measurement conditions based on the marginal covariance matrix; and a measurement scheme output module for determining candidate measurement conditions for scattering measurements of the target structure based on the ranking results, forming a recommended measurement scheme for the target structure.

[0015] Compared with the prior art, the present invention can achieve the following beneficial effects: This invention provides a method and system for determining nanostructure scattering measurement schemes, which can automatically generate EUV / SXR scattering measurement schemes, reducing reliance on manual experience. The method and system provide unified evaluation of 0th-order and non-zero-order diffraction signals, enabling switching between rapid screening and precise measurement modes. The method and system incorporate instrument perturbation during the measurement scheme selection stage, rather than providing error explanations after inversion. The method and system can identify the difference between the optimal measurement scheme considering only target parameters and the perturbation-aware optimal measurement scheme. Finally, the method and system are scalable to different forward modeling solvers, different structure types, and different instrument models. Attached Figure Description

[0016] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments and descriptions of the invention are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 A schematic diagram of the process for determining the nanostructure scattering measurement scheme and system described in the embodiments of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0018] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this invention can be combined with each other. In the description of this invention, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0019] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] like Figure 1 As shown in the embodiments of the present invention, the method and system for determining the nanostructure scattering measurement scheme include: S1: Determine the target structure to be measured and its target parameters, as well as the candidate measurement conditions for scattering measurements of the target structure.

[0021] For example, the target structure can be a one-dimensional periodic grating, a two-dimensional periodic structure, a line / spacing structure, a BEOL-like structure, or other periodic / quasi-periodic nanostructures suitable for EUV / SXR scattering measurements; the target parameters include, but are not limited to, critical size CD, structure height H, sidewall angle SWA, film thickness, fillet radius, line edge roughness, or material parameters; the candidate measurement conditions include at least EUV / SXR wavelength, incident angle, azimuth angle, polarization, and diffraction order, wherein the diffraction order includes 0th order, ±1st order, ±2nd order, or higher.

[0022] S2: Perform forward modeling of the target structure according to the candidate measurement conditions in step S1 to obtain the measurement signal; determine the sensitivity of the measurement signal to the disturbances generated by the target parameters and the instruments involved in the measurement.

[0023] In some embodiments, an electromagnetic forward model is used to perform forward modeling on the target structure to obtain a measurement signal including a 0th-order diffraction signal and at least one non-zero-order diffraction signal. The process of performing forward modeling on the target structure can be represented by the following equation: y=G(θ,ε,γ); Where θ represents the target parameter, ε represents the perturbation parameter caused by the natural disturbances of the instruments and environment involved in the scattering measurement, γ represents the candidate measurement conditions, and G represents the model used in forward modeling. For example, natural disturbances include variations in the spectral intensity of the light source, wavelength drift and polarization error of the emitted incident light, nonlinearity of the detector receiving the measurement signal, measurement signal drift caused by sample contamination of the target structure, and scattering background in the environment where the scattering measurement takes place.

[0024] In other embodiments, the forward model may also be RCWA, S4, FEM, FDTD, boundary element method, or a trained surrogate model. In one embodiment, a combination of multiple candidate measurement conditions from step S1 is used to perform forward modeling of the target structure.

[0025] In some embodiments, determining the sensitivity of the measurement signal to the target structure in step S2 includes: The first sensitivity of the measured signal to the target parameter is obtained by differentiating the measured signal with respect to the target parameter, as shown in the following formula: j θ =dy / dθ; Where, j θ Indicates the first sensitivity; The second sensitivity of the measured signal to natural disturbance is obtained by differentiating the measured signal with respect to natural disturbance, as shown in the following formula: j ε =dy / dε; Where, j ε This indicates the second sensitivity.

[0026] In some other embodiments, the first sensitivity and the second sensitivity can be obtained by differentiation using finite difference, automatic differentiation, adjoint method or surrogate model differentiation.

[0027] S3: Based on the measurement signal in step S2, determine the noise model in the scattering measurement process, and combine it with the sensitivity obtained in step S2 to obtain the extended information matrix of the target structure when performing forward modeling under the candidate measurement conditions in step S1.

[0028] In some embodiments, the noise during the scattering measurement process includes relative noise, shot noise generated when incident photons are incident on the target structure, and the background noise of the detector used to receive the measurement signal. Therefore, the noise model in step S3 is: σ i 2 =(r i ×y i ) 2 +y i / N i +σ f,i 2 ; Where, σ i 2 r represents the total noise standard deviation of the i-th observation channel in the detector receiving the measurement signal. i y represents the relative noise figure of the i-th observation channel. i N represents the measurement signal received by the i-th observation channel. i σ represents the equivalent photon number or integral scale in the i-th observation channel. f,i This represents the background noise of the i-th observation channel in the detector.

[0029] The process of obtaining the extended information matrix includes: integrating the first sensitivity and the second sensitivity to obtain a preliminary extended matrix J=[j] in Jacobian matrix form. θ ,j ε The extended information matrix is ​​obtained through the following formula: F=J T Σ -1 J+diag(0,P); Where F represents the extended information matrix, J represents the preliminary extended matrix, Σ represents the noise covariance matrix constructed based on the noise model. Specifically, the noise covariance matrix Σ is constructed from the total noise standard deviation in the noise model. When the observation channels in the detector are independent of each other, the total noise standard deviation forms the noise covariance matrix according to the order of the observation channels. When the observation channels in the detector are related to each other, the noise covariance matrix Σ constructed from the total noise standard deviation is the complete covariance matrix of off-diagonal terms. P represents the prior information matrix of natural disturbances. The prior information matrix P is the inverse matrix of the prior covariance of the instrument disturbance parameters. diag represents the function that creates the diagonal of the matrix.

[0030] S4: Based on the extended information matrix obtained in step S3, obtain the marginal covariance of the target parameters; sort the candidate measurement conditions according to the marginal covariance.

[0031] In some embodiments, step S4 includes: When the measurement signal only includes the 0th order diffraction signal, the fast screening mode is adopted, that is, the measurement time of the candidate measurement conditions in the scattering measurement process is calculated. When the measurement signal only includes non-zero order diffraction signals, the candidate measurement conditions related to the geometric dimensions in the target parameters are weighted down, such as the candidate measurement conditions for structural height H, sidewall angle SWA, film thickness, and fillet radius in step S1 above. When the measurement signal includes both 0th-order and non-zero-order diffraction signals, a precise screening mode is employed. This involves calculating the geometric properties of the edge covariance matrix and ranking the geometric properties of all candidate measurement conditions. This ranking result represents the measurement uncertainty when performing scattering measurements on the target structure using the candidate measurement conditions. The geometric properties of the matrix can be the trace, determinant, maximum standard deviation, or weighted objective function of the edge covariance matrix. In one specific embodiment, a larger geometric property indicates a smaller measurement uncertainty.

[0032] In some other embodiments, the target parameter CRLB (Crame-Rhodes lower bound) can also be directly calculated from the edge covariance matrix. Specifically, the standard deviation of the target parameter CRLB is obtained from the square root of the diagonal elements of the edge covariance matrix.

[0033] In some embodiments, step S4 further includes: calculating the rank or minimum singular value of all marginal covariance matrices; when the rank is less than the number of target parameters, or the minimum singular value is lower than a preset threshold, the corresponding candidate measurement condition is excluded or marked as not fully identifiable; adding artificial perturbation to step S2, and repeating steps S2 to S4 to obtain the ranking result of the candidate measurement conditions after adding perturbation.

[0034] S5: Based on the sorting results of step S4, determine the candidate measurement conditions for scattering measurement of the target structure and form a recommended measurement scheme for the target structure.

[0035] In some implementations, the output recommended measurement scheme includes information contribution and early warning indicators. An audit report on the information contribution of the recommended measurement scheme is also output to improve its interpretability. Further explanation: information contribution is used to quantify and evaluate the merits of different recommended measurement schemes. In one specific embodiment, an information entropy index is used for precise measurement of information contribution to distinguish the information value of different recommended measurement schemes. Early warning indicators are used to transform the abstract recommended measurement scheme into intuitive, actionable safety signals, grading and quantifying the risks that will arise after implementing the recommended measurement scheme, and guiding relevant personnel to take appropriate recommended measurement schemes based on the quantified risk level.

[0036] This invention provides several specific implementation methods and systems for determining the proposed nanostructure scattering measurement scheme, and provides comparisons thereof.

[0037] Example 1: Using a one-dimensional silicon trapezoidal grating as the target structure, the target parameters include CD, height, and sidewall angles. Candidate measurement conditions include incident angles of 25°, 35°, 45°, 55°, and 65°, and incident photon wavelengths of 13.5 nm, 20 nm, and 29 nm. A candidate measurement condition grid is constructed using the five incident angles as columns and the three wavelengths as rows. Each cell in the grid represents a single candidate measurement condition consisting of one incident angle and one wavelength. The 0th, -1st, and +1st order reflection signals generated by the target structure under a single candidate measurement condition are calculated using forward modeling. After obtaining two sensitivities through finite difference, an extended information matrix is ​​constructed, and CRLB is calculated. The results show that a single wavelength-angle condition cell cannot simultaneously accommodate all target parameters.

[0038] Example 2: Based on Example 1, two-candidate measurement condition combinations and three-candidate measurement condition combinations are selected from the candidate measurement condition grid constructed in Example 1, i.e., two and three elements are selected from the candidate measurement condition grid, respectively. Forward modeling is used to calculate the 0th, -1st, and +1st order reflection signals generated by the target structure under the two-condition and three-condition combinations. When only the target parameters are considered, the resulting three-condition joint measurement scheme is as follows: wavelength 13.5 nm, incident angle 65°; wavelength 20.0 nm, incident angle 65°; wavelength 29.0 nm, incident angle 45°. This measurement scheme reduces the uncertainties in CD, height, and sidewall angles compared to the two-condition joint measurement scheme.

[0039] Example 3: Based on Example 2, an incident angle deviation was added as an artificial perturbation when the incident photons entered the target structure, and prior values ​​were set for the relevant perturbation parameters of both natural and artificial perturbations. After adding the perturbation, the 0th, -1st, and +1st order reflection signals generated by the target structure under two-condition and three-condition combinations were calculated using forward modeling. The optimal three-condition joint measurement scheme considering only the target parameters differs from the optimal measurement scheme for perturbation sensing. The optimal measurement schemes obtained in this case are: wavelength 13.5 nm, incident angle 35°; wavelength 29.0 nm, incident angle 45°; wavelength 29.0 nm, incident angle 55°. This indicates that instrument perturbation not only changes the uncertainty value but may also change the recommended measurement scheme itself.

[0040] This invention, based on the provided method for determining nanostructure scattering measurement schemes, also provides a system for determining nanostructure scattering measurement schemes, including a target structure input module, a candidate measurement scheme generation module, a forward modeling solution module, a sensitivity calculation module, an extended information matrix construction module, a measurement scheme ranking module, and a measurement scheme output module. The target structure input module determines the target structure to be measured and its target parameters. The candidate measurement scheme generation module determines candidate measurement conditions for scattering measurements on the target structure. The forward modeling solution module performs a forward modeling of the target structure according to the candidate measurement conditions to obtain the measurement signal. The sensitivity calculation module determines the sensitivity of the measurement signal to disturbances generated by the target parameters and the instruments involved in the measurement. The extended information matrix construction module determines the noise model in the scattering measurement process based on the measurement signal, and, combined with the sensitivity, obtains the extended information matrix corresponding to the forward modeling under the candidate measurement conditions. The measurement scheme ranking module performs marginalization on the extended information matrix to obtain the marginal covariance matrix corresponding to the candidate measurement conditions; and ranks the candidate measurement conditions based on the marginal covariance matrix. The measurement scheme output module is used to determine candidate measurement conditions for scattering measurements of the target structure based on the sorting results, and to form a recommended measurement scheme for the target structure.

[0041] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0042] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining a nanostructure scattering measurement scheme, characterized in that, include: S1: Determine the target structure to be measured and its target parameters, as well as the candidate measurement conditions for scattering measurement of the target structure; S2: Perform forward modeling of the target structure according to the candidate measurement conditions in step S1 to obtain the measurement signal; determine the sensitivity of the measurement signal to the disturbances generated by the target parameters and the instruments involved in the measurement; S3: Based on the measurement signal in step S2, determine the noise model in the scattering measurement process, and combine it with the sensitivity obtained in step S2 to obtain the extended information matrix of the target structure when performing forward modeling with the candidate measurement conditions in step S1. S4: Based on the extended information matrix obtained in step S3, obtain the marginal covariance of the target parameters; sort the candidate measurement conditions according to the marginal covariance; S5: Based on the sorting results of step S4, determine the candidate measurement conditions for scattering measurement of the target structure and form a recommended measurement scheme for the target structure.

2. The method for determining the nanostructure scattering measurement scheme according to claim 1, characterized in that, In step S2, the target structure is modeled using an electromagnetic forward model to obtain a measurement signal that includes a 0th-order diffraction signal and at least one non-zero-order diffraction signal.

3. The method for determining the nanostructure scattering measurement scheme according to claim 1, characterized in that, Determining the sensitivity of the measurement signal to the target structure in step S2 includes: The first sensitivity of the measurement signal to the target parameter is obtained by differentiating the measurement signal with respect to the target parameter. The second sensitivity of the measurement signal to natural disturbances caused by the instruments involved in the scattering measurement and the environment is obtained by differentiating the measurement signal with respect to the natural disturbances.

4. The method for determining the nanostructure scattering measurement scheme according to claim 3, characterized in that, The noise model in step S3 is: s i 2 =(r i ×y i ) 2 +y i / N i +s f,i 2 ; Where, σ i 2 r represents the total noise standard deviation of the i-th observation channel in the detector receiving the measurement signal. i y represents the relative noise figure of the i-th observation channel. i N represents the measurement signal received by the i-th observation channel. i σ represents the equivalent photon number or integral scale in the i-th observation channel. f,i This represents the background noise of the i-th observation channel in the detector.

5. The method for determining the nanostructure scattering measurement scheme according to claim 3, characterized in that, The process of obtaining the extended information matrix in step S3 includes: Integrating the first and second sensitivities yields a preliminary extended matrix; The extended information matrix is ​​obtained using the following formula: F=J T Σ -1 J+diag(0,P): Where F represents the extended information matrix, J represents the initial extended matrix, Σ represents the noise covariance matrix constructed based on the noise model, P represents the prior information matrix of natural disturbances, is the inverse matrix of the prior covariance of the instrument disturbance parameters, and diag represents the function that creates the diagonal of the matrix.

6. The method for determining the nanostructure scattering measurement scheme according to claim 1, characterized in that, In step S4: When the measurement signal only includes the 0th order diffraction signal, calculate the measurement time of the candidate measurement conditions in the scattering measurement process; When the measurement signal only includes non-zero order diffraction signals, the candidate measurement conditions related to the geometric dimensions in the target parameters are weighted down. When the measurement signal includes both 0th-order diffraction signals and non-zero-order diffraction signals, the geometric properties of the matrix corresponding to the calculated edge covariance matrix are used to sort all candidate measurement conditions. This sorting result is the measurement uncertainty when performing scattering measurements on the target structure under the candidate measurement conditions.

7. The method for determining the nanostructure scattering measurement scheme according to claim 6, characterized in that, Step S4 further includes: calculating the rank or minimum singular value of all marginal covariance matrices; when the rank is less than the number of target parameters, or the minimum singular value is lower than a preset threshold, the corresponding candidate measurement conditions are excluded or marked as incompletely identifiable; adding artificial perturbation to step S2, and repeating steps S2 to S4 to obtain the ranking result of the candidate measurement conditions after adding perturbation.

8. A system for determining a nanostructure scattering measurement scheme, characterized in that, The method for determining the nanostructure scattering measurement scheme according to any one of claims 1 to 7 includes: The target structure input module is used to determine the target structure to be measured and its target parameters; The candidate measurement scheme generation module is used to determine the candidate measurement conditions for scattering measurements of the target structure. The forward modeling module is used to perform forward modeling of the target structure according to candidate measurement conditions to obtain measurement signals; The sensitivity calculation module is used to determine the sensitivity of the measurement signal to the perturbations generated by the target parameters and the instruments involved in the measurement. The extended information matrix construction module is used to determine the noise model in the scattering measurement process based on the measurement signal, and combine it with the sensitivity to obtain the extended information matrix corresponding to the forward modeling under the candidate measurement conditions. The measurement scheme ranking module is used to perform edge-based solution on the extended information matrix to obtain the edge covariance matrix corresponding to the candidate measurement conditions; and to rank the candidate measurement conditions based on the edge covariance matrix. The measurement scheme output module is used to determine candidate measurement conditions for scattering measurements of the target structure based on the sorting results, and to form a recommended measurement scheme for the target structure.

Images