A hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition

Abstract

This invention proposes a hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition. The method includes the following steps: Step 1. Phase-weighted superposition based on S-transform; Step 2. Linear superposition based on data filtering. This method can effectively improve the signal-to-noise ratio of weak signals.

Description

Technical Field

[0001] This invention belongs to the field of seismic wave signal processing, and particularly relates to a hybrid signal enhancement algorithm based on S-transform phase-weighted superposition and linear superposition. Background Technology

[0002] Seismic exploration requires processing a large volume of seismic wave signals. Among these, many are weak signals with low signal-to-noise ratios (SNR) due to factors such as instrument reception quality, noise pollution, and the inherently weak energy of the seismic source. These weak signals are essential data for imaging subsurface structures, and a low SNR severely hinders the proper extraction of seismic wave signals, thus impacting the overall exploration results. Therefore, it is necessary to enhance these weak signals. Summary of the Invention

[0003] To address this issue, this invention proposes a hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition to improve the signal-to-noise ratio of weak signals.

[0004] A hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition, the method comprising the following steps:

[0005] Step 1. Phase-weighted superposition based on S-transform;

[0006] Step 2. Linear overlay based on data filtering.

[0007] Furthermore, the implementation method for step 1 is as follows:

[0008] For a single u(t) from N time-amplitude domain seismic wave signal records, the S-transform method is used to transform it to the time-frequency domain. The transformed signal is represented as:

[0009]

[0010] Where S(τ, f) represents the time-frequency domain signal after S-transformation, ω(τ-t, f) represents a Gaussian window with an intermediate time of τ and a window width of ω.

[0011]

[0012] Where k is a constant parameter variable;

[0013] The weighted coefficients for a given time τ and frequency f are expressed as follows:

[0014]

[0015] Where N represents the total number of earthquake records, and S is the result of phase-weighted superposition. pws (τ,f) is expressed as

[0016] S pws (τ,f)=c(τ,f)S ls (τ,f)

[0017] Among them, S ls (τ,f) represents the linear superposition of N seismic records after S-transformation. Then, the superimposed time-frequency domain data is inversely transformed to the time-amplitude domain to obtain...

[0018] Furthermore, the specific implementation method of step 2 is as follows:

[0019] Linearly superimpose each existing earthquake record onto... If the signal-to-noise ratio is improved, it is retained; in this case, the superimposed seismic data is represented as follows: If the signal-to-noise ratio does not improve, the data is discarded. After multiple iterations of stacking, the final stacked seismic data is represented as follows: M represents the total number of linear superpositions.

[0020] This method significantly improves the signal-to-noise ratio (SNR) for weak signals. Here, SNR is defined as the ratio of the mean absolute values ​​of the amplitudes in the signal window and the noise window. Attached Figure Description

[0021] Figure 1 This is a flowchart of the method of the present invention;

[0022] Figure 2 This is a comparison chart of the effects of different methods. Detailed Implementation

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] This invention proposes a hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition. The method comprises two steps: 1. S-transform-based phase-weighted superposition; 2. Linear superposition based on data filtering.

[0025] Step 1, Phase-weighted superposition based on S-transform. For a single u(t) from N time-amplitude domain seismic wave signal records, the S-transform method is used to transform it to the time-frequency domain. The transformed signal is represented as:

[0026]

[0027] Where S(τ, f) represents the time-frequency domain signal after S-transformation, ω(τ-t, f) represents a Gaussian window with an intermediate time of τ and a window width of ω.

[0028]

[0029] Here, k is a constant parameter variable that controls the resolution of the Gaussian window.

[0030] The weighted coefficients for a given time τ and frequency f are expressed as follows:

[0031]

[0032] Where N represents the total number of earthquake records. The result of phase-weighted superposition is S. pws (τ,f) can be expressed as:

[0033] S pws (τ,f)=c(τ,f)S ls (τ,f)

[0034] Among them, S ls (τ,f) represents the linear superposition of N seismic records after S-transformation. Then, the superimposed time-frequency domain data is inversely transformed to the time-amplitude domain to obtain...

[0035] Step 2. Linear stacking based on data filtering. Linearly stack each existing seismic record onto... If the signal-to-noise ratio is improved, it is retained; in this case, the superimposed seismic data is represented as follows: If the signal-to-noise ratio does not improve, the data point is discarded. After multiple iterations of overlay, the final overlay of seismic data is represented as follows: M represents the total number of linear superpositions. This concludes the entire algorithm process. The flowchart is as follows: Figure 1 As shown.

[0036] After testing, this method showed a significant improvement in signal-to-noise ratio, such as... Figure 2 The figure shows, from top to bottom, a single data point from the actual dataset: the unstacked data, the data after linear stacking, the data after S-transform weighted stacking, and the data after stacking using this method. The signal-to-noise ratios (SNRs) are 1.2, 3.7, 21.2, and 23.1, respectively. Here, SNR is defined as the ratio of the mean absolute values ​​of the amplitudes in the signal window to the mean values ​​in the noise window.

[0037] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A hybrid signal enhancement method based on S-transform phase-weighted superposition and linear superposition, characterized in that, The method includes the following steps: Step 1. Phase-weighted superposition based on S-transform; Step 2. Linear overlay based on data filtering; The implementation method for step 1 is as follows: for A time-amplitude domain seismic wave signal in a seismic record The S-transform is used to transform it to the time-frequency domain. The transformed signal is represented as follows: in, This represents the time-frequency domain signal after S-transformation. This represents a Gaussian window with a median time of... The time window width is ; in, It is a constant parameter variable; Given time and frequency The weighting coefficients are expressed as follows: in, This represents the total number of earthquake records, the result of phase-weighted superposition. Expressed as: in, express The seismic records were linearly stacked after S-transformation, and then the stacked time-frequency domain data were inversely transformed to the time-amplitude domain to obtain... .

2. The method according to claim 1, characterized in that, The specific implementation method of step 2 is as follows: Linearly superimpose each existing seismic record onto... If the signal-to-noise ratio is improved, it is retained; in this case, the superimposed seismic data is represented as follows. If the signal-to-noise ratio does not improve, the data is discarded. After multiple iterations of stacking, the final stacked seismic data is represented as follows: , This represents the total number of linear superpositions.

Images