Real-Time Radio Technology and Modulation Classification via an LSTM Auto-Encoder

Ke, Ziqi; Vikalo, Haris

doi:10.1109/twc.2021.3095855

Cited by 55 publications

(19 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another LSTM-based model proposed in [23], which incorporates fully connected layers and a temporal attention mechanism, is proved to be robust to uncertain noise conditions. An LSTM denoising auto-encoder is designed in [32] to automatically infer modulation schemes using a compact RNN architecture readily implemented on a low-cost computational platform while exceeding state-of-the-art accuracy in [56]. Benefitting from the respective advantages of RNN and CNN, researchers have further proposed hybrid models combining RNN and CNN to further improve the AMR performance, as shown below.…”

Section: Rnn-based Modelsmentioning

confidence: 99%

Deep learning based automatic modulation recognition: Models, datasets, and challenges

Zhang

Luo

et al. 2022

Digital Signal Processing

View full text Add to dashboard Cite

Section: Rnn-based Modelsmentioning

confidence: 99%

Deep learning based automatic modulation recognition: Models, datasets, and challenges

Zhang

Luo

et al. 2022

Digital Signal Processing

View full text Add to dashboard Cite

“…We used the RadioML 2016.10A dataset to compare and verify the algorithms mentioned in the reference ALRT (Hong and Ho, 2003), GLRT (Panagiotou et al, 2000), HLRT (Hong and Ho, 2003), CNN (O'Shea et al, 2016), MT-CNN (Qiao et al, 2022), SPWVD -CNN (Hou et al, 2021), LSTM-AE (Ke and Vikalo, 2022), LSTM (Rajendran et al, 2018) and SOTA (Zhang et al, 2021). The precision curves of various algorithms are shown in Figure 5.…”

Section: Model Testingmentioning

confidence: 99%

“…In the 0 dB SNR environment, the proposed multi-task CNN method outperforms the traditional CNN method by 20%. Ke and Vikalo (2022) designed a learning framework for LSTM denoising encoder, which can automatically extract stable robustness features from noisy signals according to amplitude and phase, and use the learned robustness features for modulation classification. This model is structurally compact, easy to implement on low-cost embedded platforms, and can effectively classify received wireless signals.…”

Section: Introductionmentioning

confidence: 99%

Multimodal attention-based deep learning for automatic modulation classification

Han

Yang

2022

Front. Energy Res.

View full text Add to dashboard Cite

Wireless Internet of Things (IoT) is widely accepted in data collection and transmission of power system, with the prerequisite that the base station of wireless IoT be compatible with a variety of digital modulation types to meet data transmission requirements of terminals with different modulation modes. As a key technology in wireless IoT communication, Automatic Modulation Classification (AMC) manages resource shortage and improves spectrum utilization efficiency. And for better accuracy and efficiency in the classification of wireless signal modulation, Deep learning (DL) is frequently exploited. It is found in real cases that the signal-to-noise ratio (SNR) of wireless signals received by base station remains low due to complex electromagnetic interference from power equipment, increasing difficulties for accurate AMC. Therefore, inspired by attention mechanism of multi-layer perceptron (MLP), AMC-MLP is introduced herein as a novel AMC method for low SNR signals. Firstly, the sampled I/Q data is converted to constellation diagram, smoothed pseudo Wigner-Ville distribution (SPWVD), and contour diagram of the spectral correlation function (SCF). Secondly, convolution auto-encoder (Conv-AE) is used to denoise and extract image feature vectors. Finally, MLP is employed to fuse multimodal features to classify signals. AMC-MLP model utilizes the characterization advantages of feature images in different modulation modes and boosts the classification accuracy of low SNR signals. Results of simulations on RadioML 2016.10A public dataset prove as well that AMC-MLP provides significantly better classification accuracy of signals in low SNR range than that of other latest deep-learning AMC methods.

show abstract

“…As shown in Figure 1(a), the long short-term memory (LSTM)-based denoising autoencoder (DAE) proposed by Ke and Vikalo (2021) can recognize the gaussian minimum shift keying (GMSK) signal simulated in the HKDD_AMC36 data set (Zheng et al, 2023) with a maximum recognition accuracy of 100% (sequence length is 1024), while for the real At the same time, the influence of noise will also cause the recognition performance of the model to decrease. The two pictures in Figure 1(b), respectively, visualize the IQ waveforms of the two GMSK signals.…”

Section: Introductionmentioning

confidence: 99%

TLN-LSTM: an automatic modulation recognition classifier based on a two-layer nested structure of LSTM network for extremely long signal sequences

Qian,

Tu,

Hou

et al. 2024

IJWIS

View full text Add to dashboard Cite

Purpose Automatic modulation recognition (AMR) is a challenging problem in intelligent communication systems and has wide application prospects. At present, although many AMR methods based on deep learning have been proposed, the methods proposed by these works cannot be directly applied to the actual wireless communication scenario, because there are usually two kinds of dilemmas when recognizing the real modulated signal, namely, long sequence and noise. This paper aims to effectively process in-phase quadrature (IQ) sequences of very long signals interfered by noise. Design/methodology/approach This paper proposes a general model for a modulation classifier based on a two-layer nested structure of long short-term memory (LSTM) networks, called a two-layer nested structure (TLN)-LSTM, which exploits the time sensitivity of LSTM and the ability of the nested network structure to extract more features, and can achieve effective processing of ultra-long signal IQ sequences collected from real wireless communication scenarios that are interfered by noise. Findings Experimental results show that our proposed model has higher recognition accuracy for five types of modulation signals, including amplitude modulation, frequency modulation, gaussian minimum shift keying, quadrature phase shift keying and differential quadrature phase shift keying, collected from real wireless communication scenarios. The overall classification accuracy of the proposed model for these signals can reach 73.11%, compared with 40.84% for the baseline model. Moreover, this model can also achieve high classification performance for analog signals with the same modulation method in the public data set HKDD_AMC36. Originality/value At present, although many AMR methods based on deep learning have been proposed, these works are based on the model’s classification results of various modulated signals in the AMR public data set to evaluate the signal recognition performance of the proposed method rather than collecting real modulated signals for identification in actual wireless communication scenarios. The methods proposed in these works cannot be directly applied to actual wireless communication scenarios. Therefore, this paper proposes a new AMR method, dedicated to the effective processing of the collected ultra-long signal IQ sequences that are interfered by noise.

show abstract

Real-Time Radio Technology and Modulation Classification via an LSTM Auto-Encoder

Cited by 55 publications

References 48 publications

Deep learning based automatic modulation recognition: Models, datasets, and challenges

Deep learning based automatic modulation recognition: Models, datasets, and challenges

Multimodal attention-based deep learning for automatic modulation classification

TLN-LSTM: an automatic modulation recognition classifier based on a two-layer nested structure of LSTM network for extremely long signal sequences

Contact Info

Product

Resources

About