This paper presents an overview of feature-based (FB) methods developed for Automatic classification of digital modulations. Only the most well-known features and classifiers are considered, categorized, and defined. The features include instantaneous time domain (ITD) parameters, Fourier transform (FT), wavelet transform (WT), higher order moments (HOM) to name a few. The classifiers are artificial neural networks (ANN), support vector machines (SVMs), and decision tree (DT). We also highlight the advantages and disadvantages of each technique in classifying a certain modulation scheme. The objective of this work is to assist newcomers to the field to choose suitable algorithms for intended applications. Furthermore, this work is expected to help in determining the limitations associated with the available FB automatic modulation classification (AMC) methods.

An overview of feature-based methods for digital modulation classification

This paper presents a new method based on bootstrap technique for modulation recognition under the low SNR and multi-path conditions. The bootstrap method is performed when only few samples are available at the receiver. A new neural network is also proposed to select effective features and classifies signals. Extensive simulations show that the new method has preferable performance in low SNR and multi-path environment.

A New Method for Modulation Classification Based on Bootstrap Technique

This paper was proposed a modified blind OFDM system parameters estimation method. The method was based on the periodicity of mobile autocorrelation function, and it estimated the OFDM symbol duration by using of the cyclic frequency test. Moreover, it limited the searching range of cyclic frequency which reduced the computational complexity. Simulation results show that the proposed time-parameter estimation algorithms could obtain good performance in low SNR and multi-path channel condition, which are also proved robust to phase offset and frequency offset.

A modified blind OFDM systems parameters estimation method

Currently, deep learning methods with stacking small size convolutional filters are widely used for automatic modulation classification (AMC). In this report, we find some experienced improvements by using large kernel size for convolutional deep convolution neural network based AMC, which is more efficient in extracting multi-scale features of the raw signal I/Q sequence data. Also, Squeeze-and-Excitation (SE) mechanisms can significantly help AMC networks to focus on the more important features of the signal. As a result, we propose a multi-scale feature network with large kernel size and SE mechanism (SE-MSFN) in this paper. SE-MSFN achieves state-of-the-art classification performance on the public well-known RADIOML 2018.01A dataset, with average classification accuracy of 64.50%, surpassing CLDNN by 1.42%, maximum classification accuracy of 98.5%, and an average classification accuracy of 85.53% in the lower SNR range OdB to 10dB, surpassing CLDNN by 2.85%. In addition, we also verified that ensemble learning can help further improve classification performance. We hope this report can provide some references for developers and researchers in practical scenes.

Deep Multi-Scale Representation Learning with Attention for Automatic Modulation Classification

The multicarrier OFDM technology has been chosen by many recent networking standards as preferred modulation scheme at physical layer as it offers high robustness against multipath effects. Automatic modulation recognition of OFDM signal has been thus intensive research area in cognitive radios. Several algorithms have been proposed in past that carry out effective detection, parameter estimations, and automatic recognition of OFDM signals as part of radio sensing techniques. In this paper, we proposed neural network-based classification of WiMAX IEEE 802.19 physical layer OFDM signal which does not require any a priori information or depends on cooperative embedded information from transmitter. The proposed algorithm classifies WiMAX IEEE 802.16d OFDM signal in a heterogeneous network environment having other digital modulation signals. The proposed features are robust to channel noise and multipath fading effects on wireless channel.

Automatic Classification of WiMAX Physical Layer OFDM Signals Using Neural Network

This paper introduces a novel method of modulation classification for digitally modulated signals in the presence of additive white Gaussian noise (AWGN). This method does not need any prior knowledge of the signals, such as SNR (Signal-to-Noise Ratio), symbol rate and carrier frequency. Four kinds of features are extracted to achieve a tree-based classification approach, and three radial basis function (RBF) neural networks are employed in the classifier. The computer simulation is carried out and the results are presented.

A Novel Method of Modulation Classification for Digital Signals

This paper presents a novel method of symbol duration estimation for orthogonal frequency division multiplexing (OFDM) signals. This method exploits the cyclostationarity of OFDM signals and estimates the symbol duration by maximizing a sum of modulus squares of cyclic correlation estimates in the cyclic domain. The computer simulation has been carried out and good performance has been obtained when the SNR is above 4 dB

Lei Huo

Papers

A Novel Method of Modulation Classification for Digital Signals

Cyclic Correlation Based Symbol Duration Estimation for OFDM Signals