5G cascaded channel estimation using convolutional neural networks

Coutinho, Fabio D. L.; Silva, Hugerles S.; Georgieva, Pétia; Oliveira, Arnaldo S. R.

doi:10.1016/j.dsp.2022.103483

Cited by 9 publications

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The findings show that the DL CSE based on ReLU DNNs may closely mimic the MMSE CSE with numerous training samples. Coutinho et al [21] suggested employing CNNs without forward error correction codes to solve the issue of cascaded channel state estimation in 5G and future communication systems. The findings reveal that the proposed CNN-based CSE approaches perfect (theoretical) channel estimation levels in terms of bit error rate (BER) values and beats LS practical estimation in terms of mean squared error (MSE).…”

Section: Introductionmentioning

confidence: 99%

Machine Learning-Based Channel State Estimators for 5G Wireless Communication Systems

Ali¹,

Alraddady²,

Al-Thunaibat³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

For a 5G wireless communication system, a convolutional deep neural network (CNN) is employed to synthesize a robust channel state estimator (CSE). The proposed CSE extracts channel information from transmit-and-receive pairs through offline training to estimate the channel state information. Also, it utilizes pilots to offer more helpful information about the communication channel. The proposed CNN-CSE performance is compared with previously published results for Bidirectional/long short-term memory (BiLSTM/LSTM) NNs-based CSEs. The CNN-CSE achieves outstanding performance using sufficient pilots only and loses its functionality at limited pilots compared with BiLSTM and LSTM-based estimators. Using three different loss function-based classification layers and the Adam optimization algorithm, a comparative study was conducted to assess the performance of the presented DNNs-based CSEs. The BiLSTM-CSE outperforms LSTM, CNN, conventional least squares (LS), and minimum mean square error (MMSE) CSEs. In addition, the computational and learning time complexities for DNN-CSEs are provided. These estimators are promising for 5G and future communication systems because they can analyze large amounts of data, discover statistical dependencies, learn correlations between features, and generalize the gotten knowledge. KEYWORDSDLNNs; channel state estimator; 5G and beyond communication systems; robust loss functions estimation challenging to implement [3]. Deep learning neural networks-based wireless communication applications recently received a lot of attention, such as coding and decoding, automatic signals classification, MIMO detection, and channel estimation [2,5-11]. As a result, several deep learning algorithms, such as recurrent neural networks (RNNs) (e.g., LSTM and BiLSTM NNs), convolutional neural networks (CNNs), and hybrid CNN and RNN structures, have utilized to estimate the channel state in 5G wireless communication networks.In [12], the channel response feature vectors and channel estimation were extracted using CNN and BiLSTM deep learning-based estimators, respectively. The main target was to enhance the estimate performance at the downlink, in communication environments concerning high-speed mobility. In [13], an online-trained CSE that integrates CNN and LSTM have developed. In addition, the authors have developed a way for combining offline and online learning for 5G systems. In [5], for OFDM systems which deal with frequency selective channels, a feedforward DLNN-based joint CSE was presented. When uncertain imperfections are taken into account, the suggested approach outperforms the classic MMSE estimate technique. In [14], Feedforward DNNs were used to develop an online CSE for doubly selective channels. The suggested CSE was found to be better than the classic LMMSE one in all circumstances studied. In [8], 1D-CNN-based CSE have introduced. Using different modulation methods, the authors compared the proposed estimator performance against FFNN, MMSE, and LS CSEs in terms of MSE and BER. 1D-CNN outs...

show abstract

Section: Introductionmentioning

confidence: 99%