Seismic Signal Noise Attenuation via Adaptive Subband Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Seismic data acquisition in marine environments faces challenges in separating low frequency seismic signals from noise, as noise is stronger and seismic signals are weaker at lower frequencies, while at higher frequencies, noise is weaker and signals are stronger, making it difficult to effectively remove noise and preserve the seismic signal.
Innovation Solution
The technique involves decomposing seismic signals into subbands using successive stages with adaptive noise attenuation and variable length spatial filtering, which selectively attenuates noise and reconstructs the signal, allowing for better noise removal and preservation of seismic information across different frequency ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If adaptive noise attenuation is applied to the entire signal at once, then noise removal is attempted, but the low frequency seismic signals are lost along with the noise
Solution Approach 1:
The signal is decomposed into multiple frequency subbands using wavelet transform, allowing selective processing of different frequency ranges. This segmentation enables noise attenuation to be applied differently to each subband, preserving low frequency seismic signals while removing noise from other frequency ranges.
Solution Approach 2:
Different noise attenuation strategies are applied to different frequency subbands based on their specific characteristics. Low frequency subbands use methods that preserve seismic signals, while other subbands use more aggressive noise attenuation, optimizing the balance between noise removal and signal preservation in each local frequency region.
2Object-affected harmful factors
If strong noise attenuation is applied, then noise is reduced, but the weaker low frequency seismic signals are also attenuated
Solution Approach 1:
By dividing the signal into frequency subbands, the system can apply different attenuation strengths to each band. Low frequency subbands containing weak seismic signals receive gentler processing, while other subbands with stronger noise components receive more aggressive attenuation, maintaining overall signal reliability.
Solution Approach 2:
The noise attenuation parameters are adjusted based on the frequency characteristics of each subband. Low frequency subbands use parameters optimized for signal preservation, while other subbands use parameters optimized for noise removal, allowing strong overall noise attenuation without sacrificing weak seismic signals.
3Reliability
If frequency-based decomposition is used to preserve low frequency signals, then signal preservation improves, but processing complexity increases
Solution Approach 1:
The wavelet transform decomposes the signal into a manageable number of frequency subbands, making the complex task of selective signal preservation feasible. Each subband can be processed independently with optimized algorithms, reducing the overall computational complexity compared to processing the entire signal uniformly.
Solution Approach 2:
The processing system dynamically adjusts the number and characteristics of subbands based on the specific signal characteristics and processing requirements. This allows optimization of processing complexity while maintaining effective signal preservation, adapting the decomposition strategy to the particular seismic data being processed.
Data Source
AI summary
A technique includes decomposing a signal that is derived from a seismic acquisition into a plurality of signals such that each signal is associated with a different frequency band. For each signal of the plurality of signals, the technique includes performing the following: decomposing the signal into subbands in successive stages, where the subbands are associated with at least different frequency ranges of the signal; selectively applying adaptive noise attenuation in between the successive stages such that the stages decompose noise-attenuated subbands; and reconstructing the signal from the subbands resulting from the decomposition. The technique includes combining the reconstructed signals.