Method for converting precision of layout data

Abstract

The application provides a layout data precision conversion method, comprising the following steps: providing original layout data and reading initial DBU from the original layout data; performing division calculation on the initial DBU and a set target DBU to obtain a division calculation result; judging whether the division calculation result has a floating point error; if the division calculation result has the floating point error, eliminating the floating point error; the division calculation result after eliminating the floating point error is converted layout data; if the division calculation result does not have the floating point error, the division calculation result is the converted layout data. The application converts layout data with lower precision into layout data with higher precision, and improves the conversion efficiency.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and in particular to a method for converting layout data precision. Background Technology

[0002] Optical Proximity Correction (OPC) uses specific calculation methods to modify the shape and size of patterns on a photomask to varying degrees. For example, it can alter the topology of a transparent area on the photomask or add corresponding fine auxiliary patterns to ensure that the actual pattern projected onto the photoresist is as consistent as possible with the design pattern. OPC plays a crucial role in semiconductor manufacturing processes. The current mainstream layout data storage formats are GDSII and OASIS. By applying OPC to the layout data, the pattern projected onto the photoresist can be kept as distortion-free as possible compared to the original pattern on the photomask, thereby improving the yield of integrated circuit designs.

[0003] As technology advances, the accuracy requirements for GDSII format layout data are becoming increasingly higher. Therefore, it is necessary to convert low-precision GDSII format layout data into higher-precision layout data.

[0004] In existing technologies, GDSII format layout data is imported into the layout tool using an external tool, and a unit is set to convert the low-precision GDSII format layout data to high-precision GDSII format layout data. One cell of layout data is converted at a time, and the converted layout data of all cells are then concatenated before being exported from the layout tool. However, in existing technologies, the conversion efficiency is very low when importing multiple data sets. Summary of the Invention

[0005] The purpose of this invention is to provide a method for converting layout data precision, which can convert layout data with lower precision into layout data with higher precision, and achieves higher conversion efficiency.

[0006] To achieve the above objectives, the present invention provides a method for converting the precision of layout data, comprising:

[0007] Provide raw map data and read the initial DBU from the raw map data;

[0008] The initial DBU and the set target DBU are divided to obtain the division result.

[0009] Determine whether the division result has a floating-point error;

[0010] If there is a floating-point error, the floating-point error is eliminated, and the division calculation result after eliminating the floating-point error is the converted layout data;

[0011] If there is no floating-point error, the result of the division calculation is the converted layout data.

[0012] Optionally, in the method for converting the precision of the layout data, both the original layout data and the converted layout data are layout data in GDSII format.

[0013] Optionally, in the method for converting the precision of the layout data, before providing the original layout data and reading the initial DBU from the original layout data, the method further includes:

[0014] Provide a set of original map data, which includes multiple sets of original map data.

[0015] Optionally, in the aforementioned method for converting the precision of the layout data, each piece of original layout data undergoes DBU conversion, and after the conversion is completed, all the converted layout data form a set of converted layout data.

[0016] Optionally, in the method for converting the precision of the map data, the original map data set is stored according to a quadtree data structure, and each node on the quadtree stores one piece of original map data.

[0017] Optionally, in the method for converting the precision of the map data, the original map data includes data of polygonal graphics.

[0018] Optionally, in the method for converting the precision of the map data, the polygon data includes the position coordinates of each point of the polygon.

[0019] Optionally, in the method for converting the precision of the layout data, the method for determining whether the division calculation result has a floating-point error includes: determining that the quadrilateral shape contained in the layout data has an angle relative to the initially copied quadrilateral shape.

[0020] Optionally, in the method for converting the precision of the layout data, the method for eliminating the floating-point error includes:

[0021] Find the shapes within a certain range of the given quadrilateral shape;

[0022] Merge the quadrilateral and a certain range of figures within it.

[0023] Optionally, in the aforementioned method for converting the precision of the layout data, the 1nm value of the quadrilateral pattern is searched.

[0024] The quadrilateral shape and the shapes within 1 nm of the quadrilateral shape are merged.

[0025] In the layout data precision conversion method provided by this invention, the method includes: providing original layout data and reading an initial DBU from the original layout data; performing a division calculation between the initial DBU and a set target DBU to obtain a division calculation result; determining whether the division calculation result has a floating-point error; if a floating-point error exists, eliminating the floating-point error, and the division calculation result after eliminating the floating-point error is the converted layout data; if no floating-point error exists, the division calculation result is the converted layout data. This invention converts lower-precision layout data into higher-precision layout data and improves conversion efficiency. Attached Figure Description

[0026] Figure 1 This is a flowchart of a method for converting layout data precision according to an embodiment of the present invention. Detailed Implementation

[0027] The specific embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. The advantages and features of the present invention will become clearer from the following description. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0028] In the following text, the terms “first,” “second,” etc., are used to distinguish between similar elements and are not necessarily used to describe a specific order or chronological sequence. It should be understood that these terms, as used herein, may be replaced where appropriate. Similarly, if the methods described herein comprise a series of steps, and the order of these steps presented herein is not necessarily the only possible order in which they can be performed, and some described steps may be omitted and / or other steps not described herein may be added to the method.

[0029] Please refer to Figure 1 This invention provides a method for converting the precision of layout data, including:

[0030] S11: Provides raw map data and reads the initial DBU from the raw map data;

[0031] S12: Perform a division calculation between the initial DBU and the set target DBU to obtain the division result;

[0032] S13: Determine whether the division result has floating-point error;

[0033] S14: If there is floating-point error, the floating-point error is eliminated. The result of the division calculation after eliminating the floating-point error is the converted layout data.

[0034] S15: If there is no floating-point error, the result of the division calculation is the converted layout data.

[0035] In this embodiment of the invention, both the original layout data and the converted layout data are in GDSII format. This embodiment primarily focuses on GDSII format layout data; therefore, the data before and after conversion is also in GDSII format. The conversion only affects the precision, not the format. Precision conversion can transform low-precision GDSII format layout data that is unsuitable for current production or design into high-precision GDSII format layout data, making it usable in design or production. Simultaneously, it can also unify the precision of GDSII format layout data of different precisions, facilitating production and management.

[0036] In this embodiment of the invention, before providing the original map data and reading the initial DBU from the original map data, the method further includes: providing a set of original map data, which includes multiple sets of original map data.

[0037] In this embodiment of the invention, each original layout data undergoes DBU conversion. After the conversion is complete, all the converted layout data form a converted layout data set. One original layout data is the original layout data of one cell. A wafer includes multiple cells, so there are multiple cells of original layout data. Therefore, during precision conversion, the original layout data is processed one by one, and the resulting layout data is also one by one. Finally, after processing all the cell layout data, they are stitched together to form the layout data set to be used by OPC.

[0038] In this embodiment of the invention, the original layout data set is stored according to a quadtree data structure, with each node in the quadtree storing one original layout data. During storage, the entire complete layout is divided into four cells, each cell's name representing a branch of the second level of the quadtree. Then, each cell is further divided into four sub-cells, and the second-level branches of the quadtree are further divided into four branches, each represented by a branch. This process continues, allowing the desired cell, or graphic, to be found by looking down layer by layer from the top of the quadtree. Therefore, when performing precision transformation, the initial layout data of all cells is transformed one by one according to the quadtree data structure.

[0039] In this embodiment of the invention, the original map data includes data of polygonal graphics. Almost all map graphics are composed of polygonal graphics; therefore, the data of a polygon can be stored by storing the positions of its edges or points.

[0040] In this embodiment of the invention, the polygon data includes the position coordinates of each point of the polygon. The layout data of a cell mainly stores the position coordinates of each vertex of the polygon. By using the position of each vertex, not only can the side length of the polygon be calculated, but the polygon can also be located.

[0041] In this embodiment of the invention, the method for determining whether the division calculation result has floating-point error includes: determining whether the quadrilateral graphic contained in the layout data has an angle relative to the initially copied quadrilateral graphic. The layout data is divided into multiple layers, with data stacked one layer at a time. If the graphic of the second layer of data is the same as the graphic of the first layer of data, the first layer of data can be directly copied. However, if the graphic of the second layer of data has an angle relative to the graphic of the first layer of data, direct data copying may result in floating-point error during precision transformation. Data copying is accomplished through matrix transformation, and the formula for matrix transformation includes:

[0042]

[0043] Where P is a quadrilateral, θ is the angle of the quadrilateral, Δx is the relative offset on the X-axis between the quadrilateral and the initial transformed shape, and Δy is the relative offset on the Y-axis between the quadrilateral and the initial transformed shape. P is the polygon copied from the first layer of data graphics. The constants in this formula are fixed, and the copied polygon can be obtained using this matrix transformation. The relative offset can be calculated using the position coordinates of the two polygons. The X-axis and Y-axis are perpendicular to each other and are commonly used coordinate systems. The value of θ ranges from 0° to 360°, such as 90°, 180°, etc.

[0044] In this embodiment of the invention, the method for eliminating floating-point errors includes: searching for graphics within a certain range of the quadrilateral graphic; merging the quadrilateral graphic and graphics within a certain range of the quadrilateral graphic. If there are no floating-point errors, the initial layout data is directly converted into the target layout data. The specific conversion method is to directly obtain the result by multiplying by a certain factor, which is the result of dividing the target DBU by the initial DBU. That is, when the copied polygon has no angle with the original polygon, i.e., no matrix transformation is performed during the conversion process, this factor has no floating-point error. However, when there are floating-point errors, it is necessary to eliminate the floating-point errors. First, search for graphics within a certain range of the cell (polygon) that needs to be precision transformed. This certain range is set manually, and the range refers to the range around the 360° of the polygon, which can be calculated using coordinates. In this embodiment of the invention, find graphics within 1nm of the quadrilateral graphic; merge the quadrilateral graphic and graphics within 1nm of the quadrilateral graphic. In other embodiments of the invention, graphics within other numerical ranges can also be searched, such as 2nm. If the found graphic is one that can be removed, it is merged with the polygon. The merged result is a single piece of map data.

[0045] In summary, the layout data precision conversion method provided in this embodiment of the invention includes: providing original layout data and reading an initial DBU from the original layout data; performing a division calculation between the initial DBU and a set target DBU to obtain the division result; determining whether the division result has floating-point error; if it has floating-point error, eliminating the floating-point error, and the division result after eliminating the floating-point error is the converted layout data; if it does not have floating-point error, the division result is the converted layout data. This invention converts lower-precision layout data into higher-precision layout data and improves conversion efficiency.

[0046] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A method for converting the precision of map data, characterized in that, include: Provide raw map data and read the initial DBU from the raw map data, wherein the raw map data includes data of polygon graphics; The initial DBU and the set target DBU are divided to obtain the division result; Determine whether the division result has a floating-point error; If there is a floating-point error, the floating-point error is eliminated, and the division calculation result after eliminating the floating-point error is the converted layout data; If there is no floating-point error, the result of the division calculation is the converted layout data; The method for determining whether the division calculation result has floating-point error includes: determining whether the quadrilateral shape contained in the layout data has an angle relative to the initially copied quadrilateral shape; The method for eliminating the floating-point error includes: finding graphics within a certain range of the quadrilateral graphic; merging the quadrilateral graphic and the graphics within a certain range of the quadrilateral graphic, wherein the value of the certain range is a set value, and the range is the area around the 360° of the polygon graphic.

2. The method for converting the precision of layout data as described in claim 1, characterized in that, Both the original and converted layout data are layout data in GDSII format.

3. The method for converting the precision of layout data as described in claim 1, characterized in that, Before providing the original map data and reading the initial DBU from the original map data, the process also includes: Provide a set of original map data, which includes multiple sets of original map data.

4. The method for converting the precision of layout data as described in claim 3, characterized in that, Each of the original layout data is transformed using DBU. Once the transformation is complete, all the transformed layout data form a set of transformed layout data.

5. The method for converting the precision of layout data as described in claim 1, characterized in that, The original map data set is stored in a quadtree data structure, and each node in the quadtree stores one piece of original map data.

6. The method for converting the precision of layout data as described in claim 1, characterized in that, The polygon data includes the position coordinates of each point of the polygon.

7. The method for converting the precision of layout data as described in claim 1, characterized in that, Find the 1nm pattern of this quadrilateral pattern; The quadrilateral shape and the shapes within 1 nm of the quadrilateral shape are merged.

Images