Single Sampling Frequency Modulation Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current technologies for recognizing modulation types of wireless signals using artificial neural networks are inefficient due to the need for multiple sampling frequencies, which slows down signal classification in radio wave monitoring, especially when trying to classify signals with different frequency bandwidths.

Innovation Solution

A method and apparatus that determine a single sampling frequency for classifying modulation types by calculating the variance and mean of bandwidth values, allowing the classifier to learn using I/Q data sampled at this frequency, enabling quick classification of modulation types in a multi-frequency environment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple sampling frequencies are used to restore wireless signals with different frequency bandwidths, then the accuracy of signal restoration is improved, but the speed of obtaining I/Q data is slowed down

Engineering Contradiction:

Improvesignal restoration accuracyVSAvoidsignal classification speed

Core Design Contradiction:

Measurement precisionVSSpeed

Solution Approach 1:

The patent changes the parameter of sampling frequency from multiple variable frequencies to a single standardized frequency. By standardizing the sampling frequency to a fixed value, the system eliminates the need for frequency switching while maintaining sufficient classification accuracy through feature-based analysis rather than waveform restoration.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If multiple sampling frequencies are switched to classify modulation types of signals with different bandwidths, then the classification accuracy is improved, but the operation complexity increases

Engineering Contradiction:

Improvemodulation type classification accuracyVSAvoidreceiver operation complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the receiver universal by using a single sampling frequency that can handle multiple modulation types and different bandwidth signals. The feature-based classification approach allows the same hardware configuration to classify various modulation types without requiring frequency switching or reconfiguration.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Productivity

If feature-based classification is used instead of waveform restoration, then the classification speed is improved, but the research maturity is insufficient

Engineering Contradiction:

Improvesignal classification productivityVSAvoidclassification reliability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent performs preliminary feature extraction from the I/Q data before classification. By extracting relevant features from the sampled data and using these features as input for the classification algorithm, the system achieves both speed and reliability - the features capture essential signal characteristics while enabling fast processing.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11817977B2Method of determining single sampling frequency of classification target signal in order to predict modulation type of classification target signal, and method and apparatus for predicting modulation type by using classification target signal sampled with single sampling frequency

Publication Date: 2023.11.14 ELECTRONICS & TELECOMM RES INST

AI summary

In the present invention, in classifying modulation types of a plurality of modulation signals by using a classifier (an artificial neural network model based on machine learning), the classifier may classify the modulation types of the modulation signals by using pieces of I/Q data, sampled with one sampling frequency, as input data, and thus, may quickly classify the modulation signals.