Signal Separation via Temporal Synchronization and ICA
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current signal processing technologies face challenges in separating target signals from unwanted noise due to asynchronous propagation delays between transducers, which prevents effective noise reduction and selective amplification of target signals in applications like audio recordings, remote sensing, and biomedical devices.
Innovation Solution
The method involves synchronizing input signals by detecting noise segments, calculating time delays, and applying independent component analysis to separate target signals from noise, reducing the asychronization effect and improving noise separation performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple transducers are used to measure signals from physically separated sources, then the distances between transducers and sources provide necessary information for signal separation, but propagation delays cause asychronization that prevents effective separation
Solution Approach 1:
The patent applies preliminary action by synchronizing the input signals before processing them for separation. The synchronization step compensates for propagation delays by aligning the temporal characteristics of signals from multiple transducers, ensuring that asychronization does not prevent subsequent effective signal separation. This preliminary timing adjustment enables the system to utilize distance information while eliminating its negative effects.
2Reliability
If traditional filtering algorithms are used to separate target signals from noise, then the processing can be performed in real time, but filtering is ineffective when noise and target signals share similar frequency characteristics
Solution Approach 1:
The patent applies parameter changes by transitioning from frequency-domain filtering to time-domain synchronization and separation. Instead of relying on frequency characteristics that are shared by both signal and noise, the method synchronizes signals based on temporal alignment and uses independent component analysis to separate sources. This parameter transformation from frequency to time domain resolves the ineffectiveness of traditional filtering while maintaining real-time processing capability.
3Measurement precision
If signals are synchronized to compensate for propagation delays, then asychronization effect is reduced and noise separation performance is improved, but additional processing steps are required
Solution Approach 1:
The patent applies merging by combining the synchronization and separation operations into an integrated processing pipeline. The synchronization step is seamlessly integrated with the independent component analysis separation process, where the time alignment information obtained during synchronization directly informs the separation algorithm. This merging reduces the overall processing complexity compared to performing synchronization and separation as completely independent stages.
Data Source
AI summary
Methods, systems and storage medium for separating a target signal from noise are disclosed. A method comprises providing a plurality of input signals, each of the plurality of input signals comprising the target signal; synchronizing the plurality of input signals; and separating the plurality of synchronized input signals into the target signal and the noise.