Tensor Signal Detection Using Trace Invariants Without Data Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing signal detection methods for multi-channel images and hyperspectral data represented as tensors of order greater than or equal to 3 suffer from performance loss due to data reduction to matrices, leading to ineffective noise handling.

Innovation Solution

A method involving a sensor that acquires raw signals as tensors, calculates an invariance value based on trace invariants, compares it with reference values, and processes the signal if the signal-to-noise ratio is non-zero, without converting tensors to matrices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data reduction is applied to convert tensors to matrices or vectors, then signal detection methods can be applied, but performance is lost

Engineering Contradiction:
Improveapplicability of signal detection methodsVSAvoidsignal detection performance
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent changes the mathematical parameters by using trace invariants of tensors directly, rather than converting to matrices. This allows signal detection methods to operate on tensor data in its native form, preserving the full data structure and avoiding performance loss associated with data reduction.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical process of data reduction (converting tensors to matrices/vectors) with a direct mathematical approach using trace invariants. This substitution eliminates the need for dimensionality reduction while maintaining compatibility with existing signal detection algorithms.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Ease of operation

If tensors are converted to matrices or vectors, then existing signal detection methods can process the data, but noise handling becomes ineffective

Engineering Contradiction:
Improvecompatibility with signal detection methodsVSAvoidnoise handling capability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent changes the representation parameter from matrix/vector form to tensor form with trace invariants. This allows existing signal detection methods to operate directly on tensor data without conversion, maintaining both ease of operation and effective noise handling capabilities.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent creates a universal approach where trace invariants enable signal detection methods to handle tensor data directly, making the system multi-functional without requiring separate processing paths for different data types. This maintains compatibility with existing methods while improving noise handling.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If data reduction is performed to apply signal detection methods, then processing can be performed, but data integrity is compromised

Engineering Contradiction:
Improveprocessing capabilityVSAvoiddata integrity
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent substitutes the data reduction mechanism with a direct tensor-based approach using trace invariants. This replacement maintains data integrity by avoiding the information loss inherent in converting tensors to matrices or vectors, while still enabling processing through established signal detection methods.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4579578A1Signal detection in tensor data
Publication Date: 2025.07.02 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
  • EP4579578A1 patent drawingFigure 1
  • EP4579578A1 patent drawingFigure 2~3A
  • EP4579578A1 patent drawingFigure 3B

AI summary

The present description relates to a method for detecting a useful signal, the method comprising: - acquiring a raw signal, by a sensor; - providing the raw signal, to a processing device, the signal being represented by a tensor of order d greater than or equal to 3; - calculating an invariance value associated with the tensor, the invariance value being calculated on the basis of at least one trace invariant for tensors of order d; - comparing the invariance value associated with the tensor with a first reference value; - based on the comparison, providing, by the processing device, an estimate of the signal-to-noise ratio of the raw signal; and - if the estimated signal-to-noise ratio is different from 0, providing the tensor to a circuit configured to process the raw signal.