Abstract
In recent trends, automatic modulation recognition (AMR) has become a primary candidate in cognitive radio (CR). Modulation recognition using deep learning (DL) drawn attention of users in recent times due to its capability of performing in blind scenarios. In this article, convolutional neural network (CNN) model has been preferred for modulation classification purpose. This model learns from the features of modulation schemes present in the training data. CNN classifier operates in all adverse channel conditions. MPSK and MQAM modulation schemes are chosen in this paper. Total 7 modulation classes is trained in CNN with different training rates at different SNR. In this paper, proposed approach is compared with traditional models. Simulation results shows that proposed model performs well at different SNR.
Access provided by Autonomous University of Puebla. Download conference paper PDF
Similar content being viewed by others
Keywords
1 Introduction
Future wireless applications demand continuous growth in certain aspects like mobility and quality of service. In telecommunication field, several challenges still need to be covered like providing encrypted wireless communication, accurate channel estimation and upgrading the QoS of the wireless methodologies used [1]. AMR can be an ultimate solution to the challenges which are stated above. AMR is used as a primary module in cognitive radio, which identifies the modulation class in presence of noise [2]. AMR is having a large research scope in wireless communication. It is used for spectrum management and interference rejection in civilian applications. It is used for jamming and demodulation of the signal in battlefield scenarios. It has been extended to various fields like radar [3, 4] and medical. Al-Qatab et al. proposed a framework that incorporates feature selection and acoustic features that will be helpful in speech classification, depending on the levels of deterioration [5].
In the last few years, various algorithms are proposed for AMR in the literature [6]. In general, signal parameter detection methods are usually classified into likelihood ratio-based (LB) approach and feature-based (FB) approaches [7]. LB approach is formulated as a hypothesis problem [8]. It is computed by setting a specific threshold value; later, it will compare the likelihood ratio with that threshold value [9]. Zhu et al. selected the modulation type based on probability distribution that increases the loglikelihood function [10]. In LB approach, some prior probability information is required, and it is selected at the expense of high computational complexity. FB approaches have dominance in practical implementations due to their low complexity. FB approach makes decisions based on extracted features. For feature extraction characteristics like cyclic statistics [11], wavelet transform [12] and high-order cumulants [13, 14] are used. ML algorithms like ANN [11], KNN [15] and SVM [16] have been widely used for classification. But these methods were developed such that it can be operated in limited environments. Feature extraction and classification both are independent processes. Due to their mismatch, the performance will be degraded for the above approaches.
With the recent advances in wireless technologies, a new paradigm based on DL is adopted for AMR. In [17,18,19,20], high performance is achieved, which decreases human participation. DL mainly focuses on the network models, which includes CNN and recurrent neural network (RNN) for AMR. DL has numerous applications in natural language understanding [21], speech recognition and image recognition. DL has been considered as important tool in many areas such as IoT detection [22, 23], MIMO detection [24, 25], radio control [26,27,28] and channel estimation [29, 30].
Recently, CNNs have created a big impact in computer vision [31,32,33]. With this success, it is now playing a lead role in AMR [34]. Initial design model of CNN was developed by O’Shea et al. for classification [35], which gave better performances compared to previous approaches, and it is more robust for different kind of signals. Authors at [36] developed a model cascading two CNN in their work. Out of these two, first CNN utilized in-phase and quadrature (IQ) components and the second one made use of the constellation diagrams. Long-term dependency is better handled by RNN. Karpathy et al. [37] combined RNN and CNN for image classification problem, in which CNN is used for feature extraction, and RNN is used for modeling temporal dependencies.
This paper is presented as follows. Section 2 provides a brief introduction of proposed model. Section 3 gives an overview of experimental setup. Section 4 shows the results and comparison tables with traditional approaches. Concluding remarks are presented in Sect. 5.
2 Proposed Model
CNN model is preferred in this article for modulation recognition. Almost all CNN models comprise various layers like convolutional layer, batch normalization layer, rectification layer (ReLU), max pooling layer, fully connected layer, Softmax layer and output layer. Convolution layer is utilized for extracting features from input data, and these are processed to next stage. Excluding input layer, remaining layers needs an activation function. It is provided by ReLU and parametric ReLU. Pooling layer is mainly used to improve the performance. It is also used to reduce the dimensions of feature to overcome overfitting. In CNN model, fully connected layer will be used as a classifier. For better classification, fully connected layer makes use of results from previous layers. Softmax layer utilizes Softmax activation function for mapping the output data. It is preferred for multi-classification problem.
The proposed CNN model comprises image input layer, convolution 2D layer, batch normalization layer, ReLU layer and max pooling 2D layer. The following layers are repeated in 6 stages. In the last stage, average pooling 2D layer is used rather than max pooling 2D layer. Finally, fully connected layer is utilized followed by Softmax layer and classification output layer. The proposed CNN model is shown in Fig. 1.
3 Experimental Setup
For modulation recognition, 7 modulation classes are considered, i.e., 64QAM, 16QAM, CPFSK, GFSK, QPSK, BPSK and 8PSK. For each modulation class, several frames are generated for effective training, where 80% of the features are used for training, and 10% of them is utilized for testing phase. In the network training stage, 10% of features are used for validation. Parameters chosen for waveform generation are samples per frame is 1024 samples, samples per symbol is 8, sample rate (fs) is 200 kHz and center frequency is chosen as 902MHzfor digital signal modulation types and 100 MHz foe analog signal modulation types. Figure 2 represents the real and imaginary parts of different modulation classes considered for simulation.
Spectrograms of modulated signals are generated with the help of short-time Fourier transform (STFT).
Figure 3 represents the spectrograms of different modulation classes at 20 dB SNR. Each frame in the channel passes through different channel conditions like AWGN and Rician multipath fading.
CNN model makes use of 1024 samples for training. Based on training, it accurately finds out the type of modulation. CNN model is given training with different training rates and different timings. Parameters chosen for training are batch size, epochs and learning rate. The learning rate and epochs are varied based on training.
4 Simulation Results
QPSK, BPSK, 8PSK, CPFSK, GFSK, 16QAM and 64QAM signals are considered for training to check the performance of CNN at non-ideal channel conditions with SNR = (0,10,20 and 30) dB. CNN model is trained with different training rates at (80, 70, 60 and 50) and different testing times at (10, 20, 30 and 40). The confusion matrices of proposed CNN in terms of true class and prediction class at 20 dB SNR are shown in Fig. 4. For 80% training, performance reached is 93.14%. The performance levels at 70, 60 and 50% are 92.71, 92.71 and 92.43%.
Table 1 summarizes the accuracy of each modulation class at different training rates and SNR values. And Table 2 summarizes the test accuracies of CNN model for AMR with different SNR and different training rates. Table 3 shows the performance comparison of the proposed approach with other existing approaches.
From the simulations, it is clear that event at 50% training rate, the proposed classifier performance better in classification of modulations for SNR 10 dB and above.
5 Conclusion
This paper presented CNN model approach for AMR of digitally modulated signals. From the simulation results, it is proven that CNN works in non-ideal channel conditions. It achieved an accuracy of 94% at 30 dB SNR. The proposed classifier outperformed all existing approaches. In future, this model can be applied to different signal classification problems in all adverse conditions.
References
Almohamad TA, Salleh MFM, Mahmud MN, Karaş İR, Shah NSM, Al-Gailani SA (2021) Dual-Determination of modulation types and signal-to-noise ratios using 2D-ASIQH features for next generation of wireless communication systems. IEEE Access 9:25843–25857. https://doi.org/10.1109/ACCESS.2021.3057242
Venkata Subbarao M, Samundiswary P (2020) Performance analysis of modulation recognition in multipath fading channels using pattern recognition classifiers. Wireless Pers Commun 115:129–151. https://doi.org/10.1007/s11277-020-07564-z
Wan T, Jiang K, Tang Y, Xiong Y, Tang B (2020) Automatic LPI radar signal sensing method using visibility graphs. IEEE Access 8:159650–159660. https://doi.org/10.1109/ACCESS.2020.3020336
Wei S, Qu Q, Wang M, Wu Y, Shi J (2020) Automatic Modulation Recognition for Radar Signals via Multi-Branch ACSE Networks. IEEE Access 8:94923–94935. https://doi.org/10.1109/ACCESS.2020.2995203
Al-Qatab BA, Mustafa MB (2021) Classification of dysarthric speech according to the severity of impairment: an analysis of acoustic features. IEEE Access 9:18183–18194. https://doi.org/10.1109/ACCESS.2021.3053335
Subbarao MV, Samundiswary P (2021) Automatic modulation classification using cumulants and ensemble classifiers. In: Kalya S, Kulkarni M, Shivaprakasha KS (eds) Advances in VLSI, signal processing, power electronics, IoT, Communication and embedded systems. Lecture notes in electrical engineering, vol 752. Springer, Singapore. https://doi.org/10.1007/978-981-16-0443-0_9
Subbarao MV, Samundiswary P (2018) Spectrum sensing in cognitive radio networks using time–frequency analysis and modulation recognition. In: Anguera J, Satapathy S, Bhateja V, Sunitha K (eds) Microelectronics, electromagnetics and telecommunications. Lecture notes in electrical engineering, vol 471. Springer, Singapore. https://doi.org/10.1007/978-981-10-7329-8_85
Zhang T, Shuai C, Zhou Y (2020) Deep learning for robust automatic modulation recognition method for iot applications. IEEE Access 8:117689–117697. https://doi.org/10.1109/ACCESS.2020.2981130
Chen S, Zhang Y, He Z, Nie J, Zhang W (2020) A Novel attention cooperative framework for automatic modulation recognition. IEEE Access 8:15673–15686. https://doi.org/10.1109/ACCESS.2020.2966777
Zhu D, Mathews VJ, Detienne DH (2018) A likelihood-based algorithm for blind identification of qam and psk signals. IEEE Trans Wireless Commun 17(5):3417–3430
Ramkumar B (2009) ‘Automatic modulation classification for cognitive radios using cyclic feature detection.’ IEEE Circuits Syst Mag 9(2):27–45
Kumar S, Bohara VA, Darak SJ (Mar 2017) ‘‘Automatic modulation classification by exploiting cyclostationary features in wavelet domain,’’ in Proc 23rd Nat Conf Commun (NCC), pp 1–6
Subbarao MV, Samundiswary P (2018) Automatic modulation recognition in cognitive radio receivers using multi-order cumulants and decision trees. Int J Recent Technol Eng (IJRTE) 7(4):61–69
Swami A, Sadler BM (2000) Hierarchical digital modulation classification using cumulants. IEEE Trans Commun 48(3):429–461
Subbarao MV, Samundiswary P (2019) K nearest neighbors based automatic modulation classifier for next generation adaptive radio systems. Int J Security Appl 13:41–50
Xie L, Wan Q (2018) Automatic modulation recognition for phase shift keying signals with compressive measurements. IEEE Wireless Commun. Lett. 7(2):194–197
O’Shea TJ, Corgan J, Charles Clancy T (2016) “Convolutional Radio Modulation Recognition Networks”
Zeng Y, Zhang M, Han F, Gong Y, Zhang J (2019) Spectrum analysis and convolutional neural network for automatic modulation recognition. IEEE Wireless Commun Lett 8(3):929–932
Bu K, He Y, Jing X, Han J (2020) Adversarial transfer learning for deep learning based automatic modulation classification. IEEE Signal Process Lett 27:880–884
Wang Y, Yang J, Liu M, Gui G (2020) LightAMC: lightweight automatic modulation classification via deep learning and compressive sensing. IEEE Trans Veh Technol 69(3):3491–3495
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436
Diro AA, Chilamkurti N (2018) Distributed attack detection scheme using deep learning approach for Internet of Things. Future Gener Comput Syst 82:761–768
Lopez-Martin M, Carro B, Sanchez-Esguevillas A, Lloret J (2017) Conditional variational autoencoder for prediction and feature recovery applied to intrusion detection in IoT. Sensors 17(9):1967
Samuel N, Diskin T, Wiesel A (Jul 2017) ‘‘Deep MIMO detection,’’ in Proc IEEE 18th Int Workshop Signal Process Adv Wireless Commun (SPAWC), pp 1–5
Yan X, Long F, Wang J, Fu N, Ou W, Liu B (May 2017) ‘‘Signal detection of MIMO-OFDM system based on auto encoder and extreme learning machine.’’ In Proc Int Joint Conf Neural Netw (IJCNN), pp 1602–1606
Wijaya MA, Fukawa K, Suzuki H (Sep 2015) ‘‘Intercell-interference cancellation and neural network transmit power optimization for MIMO channels,’’ in Proc. IEEE 82nd Veh Technol Conf (VTC-Fall), pp 1–5
Naparstek O, Cohen K (Dec 2017) ‘‘Deep multi-user reinforcement learning for dynamic spectrum access in multichannel wireless networks,’’ in Proc. GLOBECOM—IEEE Global Commun Conf, pp 1–7
Wijaya MA, Fukawa K, Suzuki H (2016) ‘Neural network based transmit power control and interference cancellation for MIMO small cell networks.’ IEICE Trans Commun E99.B(5):1157–1169
D. Neumann, T. Wiese, and W. Utschick, ‘‘Deep channel estimation,’’ in Proc. WSA 21th Int. ITG Workshop Smart Antennas, 2017, pp. 1–6.
Huang H, Yang J, Huang H, Song Y, Gui G (2018) Deep learning for super-resolution channel estimation and DOA estimation based massive MIMO system. IEEE Trans Veh Technol 67(9):8549–8560
Simonyan K, Zisserman A (2015) ‘‘Deep convolutional networks for largescale image recognition,’’ in Proc ICLR, pp 1–14
Huang G, Liu Z, Van Der Maaten L, Weinberger KQ (Jul 2017) ‘‘Densely connected convolutional networks,’’ in Proc IEEE Conf Comput Vis Pattern Recognit (CVPR), pp 4700–4708
He K, Zhang X, Ren S, Sun J (Jun 2016) ‘‘Deep residual learning for image recognition,’’ in Proc IEEE Conf Comput Vis Pattern Recognit (CVPR), pp 770–778
Lin R, Ren W, Sun X, Yang Z, Fu K (2020) A hybrid neural network for fast automatic modulation classification. IEEE Access 8:130314–130322. https://doi.org/10.1109/ACCESS.2020.3009471
O’Shea TJ, Corgan J, Clancy TC (2016) ‘‘Convolutional radio modulation recognition networks,’’ in Proc Int Conf Eng Appl Neural Netw, Cham, Switzerland, Springer, pp 213–226
Wang Y, Liu M, Yang J, Gui G (2019) Data-driven deep learning for automatic modulation recognition in cognitive radios. IEEE Trans Veh Technol 68(4):4074–4077
Karpathy A, Fei-Fei L (Apr 2017) ‘‘Deep visual-semantic alignments for generating image descriptions,’’ IEEE Trans. Pattern Anal. Mach. Intell 39(4):664–676 Ettefagh Y, Moghaddam MH, Eghbalian S (2017) An adaptive neural network approach for automatic modulation recognition. 2017 51st Annual Conf Inf Sci Syst 1–5
Mossad OS, ElNainay M, Torki M (2019) Deep convolutional neural network with multi-task learning scheme for modulations recognition. 2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC), pp 1644–1649
Rajendran S, Meert W, Giustiniano D, Lenders V, Pollin S (2018) Deep learning models for wireless signal classification with distributed low-cost spectrum sensors. IEEE Trans Cognitive Commun Netw 4:433–445
Zhang M, Zeng Y, Han Z, Gong Y (2018) Automatic modulation recognition using deep learning architectures. 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp 1–5
Hu S, Pei Y, Liang PP, Liang Y (2018) Robust modulation classification under uncertain noise condition using recurrent neural network. IEEE Trans Vehicular Technol 69:564–577
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Rao, N.V., Krishna, B.T. (2022). Performance Analysis of Automatic Modulation Recognition Using Convolutional Neural Network. In: Chowdary, P.S.R., Anguera, J., Satapathy, S.C., Bhateja, V. (eds) Evolution in Signal Processing and Telecommunication Networks. Lecture Notes in Electrical Engineering, vol 839. Springer, Singapore. https://doi.org/10.1007/978-981-16-8554-5_42
Download citation
DOI: https://doi.org/10.1007/978-981-16-8554-5_42
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-16-8553-8
Online ISBN: 978-981-16-8554-5
eBook Packages: EngineeringEngineering (R0)