Frequency-Varying Seismic Filtering for Crosstalk Attenuation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Seismic data from simultaneous swept frequency vibratory sources is challenged by severe crosstalk noise, which complicates static correction and image quality, especially in simultaneous sources acquisition where clear first-breaks are necessary for accurate processing.

Innovation Solution

A computer-implemented method and system that applies frequency-varying filtering to seismic data by transforming it into the frequency-space domain, normalizing amplitudes, and using a variable-frequency mean filter to attenuate cross-talk in cross-spread azimuth-offset gathers, thereby improving signal preservation and reducing noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If simultaneous swept frequency vibratory sources are used for seismic acquisition, then source productivity and acquisition efficiency are improved, but severe crosstalk noise is generated that complicates processing and reduces image quality

Engineering Contradiction:

Improvesource productivityVSAvoidcrosstalk noise

Core Design Contradiction:

ProductivityVSObject-generated harmful factors

Solution Approach 1:

The patent segments the frequency spectrum into multiple discrete frequency bins and processes each bin independently through histogram analysis. This segmentation allows the method to handle crosstalk noise in the frequency domain by analyzing amplitude distributions at each frequency separately, thereby resolving the contradiction between maintaining high source productivity and reducing crosstalk interference.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces histogram analysis as an intermediary process between data acquisition and final processing. By computing histograms of trace amplitudes at each frequency bin and using statistical measures (median, mean, standard deviation) as intermediaries, the method effectively separates signal from crosstalk noise without compromising the benefits of simultaneous source acquisition.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Object-generated harmful factors

If blended acquisition with randomized source timings is used, then cross-talk noise can be attenuated in different domains, but additional noise removal processing is required beyond standard stacking

Engineering Contradiction:

Improvecross-talk noise attenuationVSAvoidprocessing complexity

Core Design Contradiction:

Object-generated harmful factorsVSDevice complexity

Solution Approach 1:

The patent replaces complex time-domain deblending algorithms with a simpler frequency-domain histogram analysis approach. By transforming the problem from the time domain to the frequency domain and using statistical analysis of amplitude distributions, the method achieves effective noise attenuation with reduced processing complexity and without requiring iterative deblending procedures.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the processing parameters from time-domain waveforms to frequency-domain spectral components. By analyzing the data in the frequency domain and computing statistical parameters (median, mean, standard deviation) for each frequency bin, the method simplifies the noise attenuation process while maintaining effectiveness, thereby reducing overall processing complexity.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If standard stacking is applied to blended seismic data without further noise removal, then random energy is suppressed, but crosstalk noise remains and affects first-break picking accuracy

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidfirst-break picking accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies preliminary frequency-domain filtering and histogram-based noise characterization before the critical first-break picking stage. By pre-processing the data to remove crosstalk noise through frequency-bin analysis and statistical filtering, the method ensures that subsequent first-break picking operates on cleaned data, thereby maintaining both processing efficiency and picking accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces frequency-domain histogram analysis as an intermediary processing step between standard stacking and first-break picking. This intermediary process characterizes and removes crosstalk noise by analyzing amplitude distributions at each frequency bin, ensuring that the data presented to first-break picking algorithms is free from contaminating noise while maintaining processing efficiency.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS9588241B2Frequency-varying filtering of simultaneous source seismic data

Publication Date: 2017.03.07 SAUDI ARABIAN OIL CO

AI summary

Seismic data acquired by independent simultaneous sweeping (ISS®) techniques are processed is to attenuate random uncompressed cross-talk signals and improve the resolution of the pre-stack migrated time image. A frequency-varying mean filter is applied on cross-spread offset-azimuth gathers of the data. The frequency-space domain filter may vary its window size according to the characteristics of the cross-talk.