Interferometer measurement system signal pre-processing method based on digital correlation method
By processing laser interferometer signals using the digital correlation method, the timing error and phase measurement ambiguity problems in multi-channel signal synchronization processing are resolved, high-precision phase measurement and delay error compensation are achieved, and precision measurement of high-end equipment is supported.
Patent Information
- Application Number
- CN202510956831.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-11
- Publication Date
- 2025-09-26
AI Technical Summary
In high-end equipment such as microelectronic lithography machines, traditional laser interferometer signal processing systems have difficulty in synchronously processing multi-channel signals, resulting in large timing errors and low phase measurement accuracy. In addition, traditional phase measurement methods cannot distinguish between positive and negative phase differences and quadrants, affecting measurement accuracy.
A signal pre-processing method based on digital correlation method is adopted. Through analog-to-digital conversion of reference signal and measurement signal, dual orthogonal phase locking, digital correlation method phase difference measurement and time compensation module, the phase difference between two signals is measured and time delay error compensation is performed to solve the phase multi-value problem.
It achieves high-precision phase measurement, reduces timing errors, improves the synchronization and measurement accuracy of the signal processing system, and supports the precision measurement needs of high-end equipment.
Smart Images
Figure CN120702607A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of photoelectric signal processing, and designs a signal pre-processing method for an interferometer measurement system based on a digital correlation method. Background Art
[0002] Ultra-precise, high-speed laser interferometers embedded in high-end equipment such as microelectronics lithography machines have become a prerequisite and a crucial guarantee for achieving the ultimate in precision and efficiency. In integrated circuit manufacturing, laser interferometers are responsible for high-speed, ultra-precise measurement of the spatial position of mask and workpiece stages within lithography machines.
[0003] For example, current mainstream lithography machines typically integrate ultra-precision, high-speed laser interferometers with 6 to 22 axes or more to measure the 6-degree-of-freedom position and posture of the high-speed moving mask and wafer stages. This multi-axis measurement demands even higher requirements for the simultaneous multi-channel signal acquisition and processing of signal processing boards, necessitating the coordinated implementation of multiple digital chip processing boards.
[0004] During signal processing, the reference signal and the measurement signal need to be collected and processed separately through different channels. This will affect the phase measurement accuracy of the interference signal, which is a key factor in determining the accuracy of heterodyne laser interferometry. Therefore, it is necessary for the signal processing system to synchronously process multi-channel signals to ensure that the engineering results are closer to the theoretical results.
[0005] Multi-channel consistency compensation ensures synchronous processing. Differences in analog and digital components, signal processing algorithms, and optical paths can introduce delays and phase shifts to the measurement data, resulting in an inaccurate representation of the stage's current position. This error, known as timing error, depends on the stage's speed and can become significant if the speed is not constant during dynamic calibration.
[0006] Knowing the phase difference or time delay of the signals between channels is a necessary condition for compensating the time error. At the same time, the traditional correlation method based on a single reference signal can only obtain the cosine value of the phase difference ( ), due to the even symmetry of the cosine function ( = ) and periodicity ( = ), it is impossible to distinguish the positive and negative phase difference and the quadrant, resulting in phase measurement ambiguity and affecting the accuracy of subsequent physical quantity calculations. Summary of the Invention
[0007] In order to solve the above problems, this paper proposes a method based on digital correlation to measure the phase difference of two signals in the interferometer measurement system, which further reduces the noise of the signal during the measurement process and avoids the phase difference. The multi-value problem of the interferometer measurement system is solved to provide a prerequisite for the next step of delay error compensation.
[0008] The method for solving the above problems of the present invention is achieved as follows:
[0009] The method comprises a reference signal 1, a measurement signal 2, an analog-to-digital conversion module 3, a dual-orthogonal phase-locked module 4, a digital correlation method phase difference measurement module 5, and a time compensation module 6. In the method, the reference signal 1 and the measurement signal 2 are processed by the analog-to-digital conversion module 3 and the dual-orthogonal phase-locked module 4 to output two pairs of orthogonal low-frequency signals with the same frequency but different phases. The phase difference between the two pairs of signals is measured in the digital correlation method phase difference measurement module 5 and is provided to the time compensation module 6 to provide a basis for subsequent time error compensation.
[0010] The measurement signal and the reference signal pass through the analog-to-digital conversion module, and after mixing and filtering the two orthogonal internal signals generated by the dual orthogonal phase-locked module and the DDS, two pairs of orthogonal low-frequency signals with the same frequency but different phases are obtained. 、 、 、 :
[0011]
[0012] make , , Equation 1 and Equation 2 can be rewritten as:
[0013]
[0014] 、 After passing through the 1 / 4 cycle delay module, we can get 、 :
[0015]
[0016] For the same frequency signal and 、 and 、 and 、 and Perform cross-correlation operations using discrete-time signal processing methods:
[0017]
[0018]
[0019]
[0020]
[0021] 、 、 and 、 、 They are 、 、 and 、 、 It is easy to know that are the delays between two pairs of orthogonal signals. There is a correlation, so when →0 o'clock, and It only depends on the phase difference between the two signals.
[0022]
[0023] Respectively 、 Perform autocorrelation calculation:
[0024]
[0025]
[0026] In summary, we can get:
[0027]
[0028] Similarly, we can get:
[0029]
[0030]
[0031]
[0032] Last use 、 、 、 , perform inverse tangent operation in the phase solution module to solve the phase difference and . BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 It is a signal pre-processing method for interferometer measurement system based on digital correlation method.
[0034] Figure 2 The invention relates to an embodiment of a signal pre-processing method for an interferometer measurement system based on a digital correlation method. DETAILED DESCRIPTION
[0035] The present invention will be further described below with reference to specific embodiments.
[0036] Figure 2 An embodiment of a signal preprocessing method for an interferometer measurement system based on digital correlation is presented. The optical signal output by a dual-frequency laser interferometer 13 is directly coupled via optical fiber to a photoelectric conversion module 14 on a signal processing board. A photodetector responds to the light intensity signal and converts it into a corresponding current signal at the same frequency. A transimpedance amplifier then converts the current signal into a voltage signal. A signal conditioning module 15 compensates for weak light intensity signals to prevent sampling failures caused by signal amplitudes too small. It also balances the voltage signal strengths between different channels. An analog-to-digital conversion module 16 converts analog electrical signals to digital signals, enabling real-time digital signal processing in subsequent computational modules.
[0037] The measurement signal and reference signal converted into digital signals are mixed and filtered by the dual orthogonal phase-locked module 17 and the two orthogonal internal signals generated by the DDS to obtain two pairs of low-frequency signals. and 、 and 、 and 、 and Their characteristic is that they have the same frequency but different phases. Therefore, they can be cross-correlated.
[0038] There are two ways to deal with this. The first is to solve the phase difference offline. The data of the two pairs of signals are uploaded to the host computer and processed using MATLAB or other signal data processing software.
[0039] 18 pairs of delay modules 、 Perform 1 / 4 cycle delay processing to obtain 、 The autocorrelation operation module 19 uses a discrete time signal processing method to process the signal 、 、 、 、 、 、 、 Perform autocorrelation operation; cross-correlation operation modules 20, 21, 22, 23 perform autocorrelation operation on the same frequency signal and 、 and 、 and 、 and Perform cross-correlation operations.
[0040] The calculation results are then applied to equations (18) to (21). The inverse tangent function is then used to calculate the final phase difference in the phase calculation module 20 and pass it to the time compensation module 21. This method can be used if the signal frequency is low, the phase changes slowly, and real-time tracking is not necessary, but if measurement accuracy is a priority. This method offers strong algorithm flexibility, high computational accuracy, and convenient data analysis. However, its disadvantages are its lack of real-time performance and strong hardware dependency.
[0041] The second method uses an FPGA for online real-time measurement. The principles are the same, but a synchronization control unit is required to ensure that related operations are completed within the same time window to ensure consistent processing results. Cross-correlation operations can be implemented using multipliers, dividers, and accumulators. The quarter-cycle delay module can be implemented using a specific delay filter, and the inverse tangent function can be implemented using a Cordic unit. This method allows real-time measurement and is more suitable for scenarios with high-frequency and rapidly changing signals. However, its disadvantages are the increased development difficulty, truncation errors introduced during correlation operations, and the limited fixed-point calculation method, resulting in lower accuracy than the first method.
Claims
1. A signal pre-processing method for an interferometer measurement system based on a digital correlation method, comprising a reference signal 1, a measurement signal 2, an analog-to-digital conversion module 3, a dual-orthogonal phase-locked module 4, a delay module 5, an autocorrelation operation 6, a cross-correlation operation 7, a cross-correlation operation 8, a cross-correlation operation 9, a cross-correlation operation 10, a phase resolution module 11, and a time compensation module 12. The delay module 5, the autocorrelation operation 6, the cross-correlation operation 7, the cross-correlation operation 8, the cross-correlation operation 9, the cross-correlation operation 10, and the phase resolution module 11 together constitute a digital correlation method phase difference measurement module. In the method, the reference signal 1 and the measurement signal 2 are processed by the analog-to-digital conversion module 3 and the dual-orthogonal phase-locked module 4 to output two pairs of orthogonal low-frequency signals with the same frequency but different phases. The two pairs of signals are measured in the digital correlation method phase difference measurement module 5 to determine their phase difference, and the phase difference is given to the time compensation module 6 to provide a basis for subsequent time error compensation.
2. The digital correlation method phase difference measurement module of the interferometer measurement system signal pre-processing method based on the digital correlation method according to claim 1 is characterized by: After the reference signal 1 and the measurement signal 2 pass through the analog-to-digital conversion module 3, they are mixed and filtered with the two orthogonal internal signals generated by the dual-orthogonal phase-locked module 4 and the DDS, resulting in two pairs of orthogonal low-frequency signals with the same frequency but different phases. The two pairs of signals can be expressed as:
3. 、 After passing through the delay module 5, it can be expressed as:
4. The characteristic of the autocorrelation operation module 6 is that it uses discrete time signal processing method to process the signal 、 、 、 、 、 、 、 Perform autocorrelation operations.
5. Cross-correlation operation 7, cross-correlation operation 8, cross-correlation operation 9, and cross-correlation operation 10 are characterized by using discrete time signal processing methods to process the same frequency signal. and 、 and 、 and 、 and Perform cross-correlation operations.
6. The characteristic of the phase solution module is that the calculated values of the autocorrelation operation module 6 and the cross-correlation operation module 7, the cross-correlation operation module 8, the cross-correlation operation module 9, and the cross-correlation operation module 10 are used to obtain the final phase difference value between the different channel signals using the inverse tangent operation.