Deep learning based modulation classification for 5G and beyond wireless systems
- PDF / 2,399,406 Bytes
- 14 Pages / 595.224 x 790.955 pts Page_size
- 71 Downloads / 187 Views
Deep learning based modulation classification for 5G and beyond wireless systems J. Christopher Clement1
· N. Indira1 · P. Vijayakumar2 · R. Nandakumar3
Received: 2 May 2020 / Accepted: 11 September 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The 5G and beyond wireless networks will be more dynamic and heterogeneous, which needs to work on multistrand waveforms. One of the most significant challenges in such a dynamic network, especially non cooperated cases, is the identification of particular modulation type, which the transmitter uses at the given time to decode the data successfully. This research proposes a modulation classification algorithm using the combination architectures of modified convolutional neural network. The proposed deep learning architecture is developed by combining the convolutional neural network, dense network, and long short-term memory network (LSTM), which is named as convolutional LSTM dense neural network (CLDNN). Moreover, the mean cumulative sum metric (MCS) is introduced in the pooling layer for improved classification accuracy. Dimensionality reduction through Principal Component Analysis is also applied to minimize the training time, so that the proposed architecture can be adopted for its practical usage. The simulation results prove that the presented CLDNN outperforms an ordinary CNN, while taking less training time. Keywords Convolutional neural network · Dense network · LSTM · Modulation classification
1 Introduction Automatic modulation classification is used at the receiver to classify the modulation type of the signal that was This article is part of the Topical Collection: Special Issue on P2P Computing for Beyond 5G Network and Internet-of-Everything Guest Editors: Prakasam P, Ajayan John, Shohel Sayeed J. Christopher Clement
[email protected] N. Indira [email protected] P. Vijayakumar [email protected] R. Nandakumar [email protected] 1
School of Electronics Engineering, Vellore Institute of Technology, Vellore, India
2
Department of Electronics and Communication Engineering, SRM Institute of Science and Technology, Chennai, India
3
Department of Electronics and Communication Engineering, K.S.R. Institute for Engineering and Technology, Tamil Nadu, India
transmitted. Typical modulation classification requires expert signal processing algorithms that perform noise reduction and estimation of signal parameters, namely, carrier frequency and signal power. In general, the classification algorithms can be of • • •
Likelihood based (LB) Feature based (FB) Artificial Neural Networks based
In LB [22, 25, 26] and FB [6, 9, 27] techniques, the decision threshold is chosen manually, while in the Artificial Neural Network based techniques [15, 16, 19], the threshold is determined adaptively and automatically. In other words, LB algorithms compare the likelihood ratio of each possible hypothesis against a threshold, which is derived from the probability density function of the observed wave. The
Data Loading...
