Source Signal Separation Using XCSPE for Real-Time Peak Isolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional signal separation techniques require prohibitive processing power and are often incapable of real or near real-time identification and isolation of signal sources, limiting their effectiveness in separating and enhancing individual signal components.

Innovation Solution

The method employs a source-agnostic approach using Complex Spectral Phase Evolution (CSPE) and Singlet Transform Process to achieve high-resolution signal processing, enabling accurate detection and separation of oscillator peaks from a signal, which are then grouped and reconstituted into coherent groups for enhanced signal separation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If maximum likelihood estimation (MLE) is used for source signal separation, then separation accuracy can be improved, but the system requires accurate knowledge of sensor positions and orientations which are difficult to obtain in practice

Engineering Contradiction:
Improvesource signal separation accuracyVSAvoidsensor positioning and orientation measurement system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and eliminates the requirement for accurate sensor position and orientation knowledge from the MLE process. By formulating an objective function that operates directly on the observed mixed signals without requiring explicit sensor geometry parameters, the method removes this complex measurement requirement while maintaining separation accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter space of the separation algorithm by replacing physical sensor parameters (positions and orientations) with statistical parameters derived from the signal data itself. This transformation allows the system to operate in a parameter-free regime regarding sensor geometry, achieving the same separation goal through different mathematical parameters.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If conventional MLE methods are used, then theoretical separation performance can be achieved, but path propagation effects between sources and sensors are not accounted for leading to degraded practical performance

Engineering Contradiction:
Improveseparation performance under ideal conditionsVSAvoidpath propagation effects
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary anti-action by pre-compensating for path propagation effects in the objective function formulation. By incorporating models of acoustic propagation, electromagnetic wave propagation, or other relevant physics into the separation criterion, the method counteracts the degrading effects before they corrupt the separation results.

Inventive Principle:
Principle #9Preliminary anti-action

Solution Approach 2:

The patent converts the harmful path propagation effects into beneficial information by using the known propagation characteristics to inform the separation process. Rather than treating propagation effects as mere disturbances, the method utilizes them as additional constraints or prior knowledge that actually improves separation accuracy when properly incorporated into the objective function.

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

3Measurement precision

If accurate sensor position and orientation measurements are obtained using additional sensors, then separation accuracy improves, but the number of sensors and system complexity increase

Engineering Contradiction:
Improvesensor geometric parameter accuracyVSAvoidnumber of sensors
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts the need for additional sensing hardware by reformulating the separation problem to eliminate dependence on precise geometric measurements. The method achieves the same informational goal using only the signal data already captured by the primary sensor array, without requiring extra sensors for calibration or geometry measurement.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent enables the sensor array to serve itself by using the mixed signal data to implicitly capture and exploit the geometric relationships between sensors and sources. The system performs its own geometric characterization through the statistical properties of the observed signals, eliminating the need for external measurement systems or additional sensors.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3129795B1Methods and systems for improved measurement, entity and parameter estimation, and path propagation effect measurement and mitigation in source signal separation
Publication Date: 2026.04.29 XMOS INC
  • EP3129795B1 patent drawingFigure 1
  • EP3129795B1 patent drawingFigure 2
  • EP3129795B1 patent drawingFigure 3

AI summary

A method of processing a signal includes taking a signal recorded by a plurality of signal recorders, applying at least one super-resolution technique to the signal to produce an oscillator peak representation of the signal comprising a plurality of frequency components for a plurality of oscillator peaks, computing at least one Cross Channel Complex Spectral Phase Evolution (XCSPE) attribute for the signal to produce a measure of a spatial evolution of the plurality of oscillator peaks between the signal, identifying a known predicted XCSPE curve (PXC) trace corresponding to the frequency components and at least one XCSPE attribute of the plurality of oscillator peaks and utilizing the identified PXC trace to determine a spatial attribute corresponding to an origin of the signal.