AI Direction Finding Signal Count Estimation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing direction-finding measurement applications struggle to accurately estimate the number of received signals, especially in scenarios with multiple emitters, due to limitations in current methods that require covariance matrices and eigenvalue decompositions, which are computationally expensive and not suitable for time-critical applications.

Innovation Solution

A computer-implemented method and device that utilizes a pre-trained artificial intelligence algorithm to estimate the number of received signals in direction-finding measurement applications, without the need for covariance matrices or eigenvalue decompositions, by processing input data from direction-finding antennas.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional methods using covariance matrices and eigenvalue decomposition are used to estimate the number of received signals, then measurement precision may be improved, but device complexity and computational cost increase significantly

Engineering Contradiction:
Improvenumber of received signalsVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the traditional mechanical/mathematical computation system (covariance matrix calculation and eigenvalue decomposition) with an artificial intelligence-based system. The AI algorithm processes the input data directly to estimate the number of received signals, eliminating the need for complex matrix operations and significantly reducing computational complexity while maintaining measurement precision.

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

2Measurement precision

If traditional methods using covariance matrices and eigenvalue decomposition are used to estimate the number of received signals, then measurement precision may be improved, but processing time increases

Engineering Contradiction:
Improvenumber of received signalsVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-training the artificial intelligence algorithm offline before actual deployment. During runtime, the pre-trained AI model can rapidly estimate the number of received signals without performing time-consuming eigenvalue decomposition. This separates the computationally intensive training phase from the time-critical measurement phase, significantly reducing processing time while maintaining precision.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If traditional methods using covariance matrices are used, then reliable results can be obtained, but the method is not suitable for devices with limited computational power

Engineering Contradiction:
Improveresult reliabilityVSAvoiddevice compatibility
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal solution by developing an AI-based estimation method that can be deployed across different device types with varying computational capabilities. The pre-trained AI model can run on devices with limited computational power, making the solution adaptable and versatile across multiple platforms while maintaining reliable results, unlike traditional methods that require significant computational resources.

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

Data Source

PatentEP4502638A1Computer implemented method, computer program product, and measurement application device for direction finding
Publication Date: 2025.02.05 ROHDE & SCHWARZ GMBH & CO KG
  • EP4502638A1 patent drawingFigure 1
  • EP4502638A1 patent drawingFigure 2
  • EP4502638A1 patent drawingFigure 3

AI summary

The present disclosure provides a computer implemented method for estimating the number of received signals in a direction-finding measurement application, the method comprising receiving input data acquired with at least one direction-finding antenna, providing the input data to a pre-trained artificial intelligence algorithm that is trained to determine the number of received signals based on the input data, and outputting the number of received signals determined by the pre-trained artificial intelligence algorithm. Further, the present disclosure provides a respective computer program product, and a respective measurement application device.