Cyclostationary Signal Detection Without Prior Knowledge

Solution Overview

Problem

Existing methods for detecting telecommunications signals on a frequency band require a priori knowledge of the signal or noise levels, limiting their effectiveness in determining signal presence without such information.

Innovation Solution

A method involving the calculation of an energy vector representing autocorrelation function components at various shift times, followed by correlation element calculation and statistical analysis to determine a statistical indicator, which is compared to a threshold to detect cyclostationary telecommunications signals without prior knowledge.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If radiometric (energy-based) detection is used to detect signals by measuring energy above noise threshold, then signal detection capability is improved, but a priori knowledge of noise level is required which limits adaptability

Engineering Contradiction:

Improvesignal detection capabilityVSAvoida priori knowledge requirement

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The detection system performs self-calibration by automatically estimating the noise power spectral density from the received signal itself, without requiring external noise level information. The system uses the measured signal to adaptively determine detection thresholds, making the detection process self-sufficient and eliminating the need for a priori noise knowledge.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system changes the parameter being measured from simple total energy to the power spectral density across multiple frequency bins. By analyzing the spectral distribution of signal energy rather than just total energy, the system can distinguish between noise and signal more effectively without requiring prior noise level information, thus improving detection capability while reducing knowledge requirements.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If cyclostationary signal detection methods are used to reveal cyclic frequencies, then signal presence can be determined, but only one cyclic frequency can be detected at a time and a priori knowledge of cyclic frequencies is required

Engineering Contradiction:

Improvesignal presence determinationVSAvoiddetection of multiple cyclic frequencies

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The detection process is segmented into multiple independent frequency bin analyses. Instead of analyzing the entire spectrum as a single entity, the system divides the power spectral density into multiple frequency bins and performs detection on each bin independently. This segmentation allows the system to detect multiple cyclic frequencies simultaneously without requiring prior knowledge of their specific values.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The detection algorithm is designed to be universal by not requiring specific cyclic frequency values as input parameters. The system can detect any cyclostationary signal regardless of its specific cyclic frequency, making the detector multi-functional and adaptable to different signal types without reconfiguration or prior knowledge.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If existing signal detection methods are used, then detection can be performed, but a priori knowledge of signal or noise characteristics is required which reduces ease of operation

Engineering Contradiction:

Improvedetection accuracyVSAvoida priori knowledge requirement

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The detection system performs self-calibration by automatically estimating the noise power spectral density from the received signal itself, without requiring external noise level information. The system uses the measured signal to adaptively determine detection thresholds, making the detection process self-sufficient and eliminating the need for a priori noise knowledge.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system changes the parameter being measured from simple total energy to the power spectral density across multiple frequency bins. By analyzing the spectral distribution of signal energy rather than just total energy, the system can distinguish between noise and signal more effectively without requiring prior noise level information, thus improving detection capability while reducing knowledge requirements.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS8280311B2Method of determining the presence of a telecommunications signal on a frequency band

Publication Date: 2012.10.02 ORANGE SA

AI summary

The invention relates to a method of determining the presence of a telecommunications signal on a frequency band, said signal being assumed cyclostationary, comprising steps of:determining an energy vector {circumflex over (T)} (14) comprising m components respectively representative of energy values of an autocorrelation function of the signal received on said frequency band for m shift times,calculating correlation elements (15) between the m components of the energy vector {circumflex over (T)},performing a statistical calculation (16) on the correlation elements calculated so as to determine a statistical indicator λ,comparing (17) the statistical indicator obtained λ with a predetermined threshold with the aim of determining the presence of a telecommunications signal on said frequency band.