Deep Learning-Based Automatic Modulation Classification With Blind OFDM Parameter Estimation

Abstract

Automatic modulation classification (AMC) is an essential factor in dynamic spectrum access to fulfill the spectrum demand of 5G wireless communications for achieving high data rate and low latency. Many deep learning (DL)-based AMC methods have achieved improved accuracy performance for classifying analog modulation schemes, single-carrier-based modulation schemes, and multi-carrier signals using several DL architectures such as the convolutional neural network (CNN) and long-short term memory (LSTM). However, most conventional DL-based AMC methods have confused the orthogonal frequency multiplexing division (OFDM)-based signals with different OFDM useful symbol lengths. To resolve the issue, we propose a CNN model operating on the fast Fourier transformation window banks (FWB) to extract the useful symbol length in OFDM, which represent the identification of each OFDMbased wireless communication technology. After extracting the OFDM useful symbol length, we propose a DL-based AMC system combined with FWB and in-phase and quadrature phase (IQ) signals to classify the OFDM symbol length and single-carrier modulation schemes simultaneously. Furthermore, we explore the constraints of the FWB parameters according to the length and the FFT size of the OFDM signal to achieve good classification accuracy through the experiment. We constructed a dataset by generating OFDM signals of different lengths while changing the FFT size in a fixed bandwidth and by selecting only quadrature amplitude modulation (QAM) schemes from RadioML2016.10a. Experimental results show the improved performance of the classification accuracy by about 30% over conventional classifiers in additive white Gaussian noise, synchronization impairments, and fading environments.

A Survey of Blind Modulation Classification Techniques for OFDM Signals

“…In [ 90 ], a CNN-based MC is studied in order to classify SC and OFDM systems with varying symbol lengths. The majority of older DL-based MC algorithms misinterpreted OFDM-based signals with varying OFDM usable symbol lengths.…”

“…However, this technique has good classification accuracy for the normalized carrier frequency offset of range . In [ 90 ], a CNN-based MC is studied to classify modulation format for OFDM systems in the presence of CFO. FFT window banks (FWB) are utilized as input to the CNN model to estimate the length of an OFDM symbol.…”

“…Various MC algorithms for the OFDM systems were carried out in [ 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 ]. The algorithms for multiple-input multiple-output and OFDM (MIMO-OFDM) systems based on deep neural network (DNN) and Gibbs sampling are investigated in [ 44 ].…”

Robust Automatic Modulation Classification in Low Signal to Noise Ratio

“…The FB-AMC makes feature extraction and classification. In order to extract features, expert systems carefully extract various manual features, such as wavelet-based features [9], instantaneous features [10], and statistical features [3], [11].…”

Detection of OFDM modulations based on the characterization in the phase diagram domain

“…In Park et al (2021), the authors propose a deep learning-based automatic modulation classification system to classify higherorder OFDM modulations, 64 OFDM to 512 OFDM, at an SNR value of 20 dB. The results obtained are quite low in a classic deep learning manner, as the authors also showed, obtaining an accuracy below 30%.…”