Deep Learning Aided 5G Channel Estimation

Abstract

Wireless communications involves the transfer of voice and data without a cable or wires. It uses orthogonal frequency-division multiplexing, also known as a multicarrier transmission technique. In multiple input multiple output (MIMO) wireless communication (4G and 5G technologies), channel estimation is crucial. Multiple antennas are used at both the transmit and receive sides of a MIMO system to increase spectral efficiency and reliability. In 5G channel estimation is performed to improve the accuracy of the received signal. Least-squares estimation is a cheap method with relatively large channel estimation errors, but it is supported in this work by using a new channel estimation method that leverages deep learning. LS (least squares) and MMSE (minimum mean squared error) are two popular traditional approaches to channel estimation, but deep learning provides much more accurate results than previous channel estimation methods.

Introduction

I. INTRODUCTION

Fifth generation (5G) wireless technology was developed to accommodate the exponential increase in wireless data traffic and communication reliability. To overcome frequency selective fading in multipath propagation environments, OFDM (Orthogonal Frequency Division Multiplexing) technology is a natural success in existing networks. As a result, this method improves spectral efficiency compared to single-carrier approaches. Pilot symbols known to the transmitter and receiver are typically used for channel estimation. Depending on different deployment scenarios, 5G system pilot symbols can have different structures. Least-squares (LS) estimation is one of the traditional techniques for channel estimation with minimal computational effort, as it does not require prior knowledge of the statistical channel information.

However, in the context of many applications, this estimation method gives relatively poor results. As an alternative, a minimum mean squared error (MMSE) estimation approach was developed to reduce the average channel estimation error. However, the MMSE estimation approach is significantly more computationally complex as it requires channel statistical data, especially the mean and covariance matrices. DNN (deep neural network) models with two different architectures are used for frequency selective fading channel estimation in 5G MIMO OFDM systems. The effectiveness of the proposed deep learning-assisted channel estimation is then evaluated using two different receiver velocity-based scenarios. The performance of DNN-based channel estimation is compared with that of conventional LS and linear MMSE (LMMSE) in terms of mean squared error (MSE) and bit error rate (BER) versus signal-to-noise ratio (SNR) criteria. Estimate.

II. METHODOLOGY

Adaptive systems, called neural networks, also known as artificial neural networks, learn using nodes, or neurons, interconnected in a layered structure similar to the human brain. Neural networks can be trained to recognize patterns, classify data, and predict future events by learning from data. Neural networks abstract their inputs in layers. Similar to how the human brain recognizes patterns in sounds and images, it can be trained using different instances. Its behavior is determined by the connections between components and the weight or strength of these connections.

As long as there is no inter-carrier interference, each sub-carrier is represented as an independent channel and orthogonality between sub-carriers is maintained. Orthogonality allows each subcarrier component of the signal to be represented as the Hadamard product of the channel frequency response at the transmitted signal and the subcarrier as

Conclusion

When training the proposed DNN-based channel estimation method, the relevant full channel and least-squares channel estimation are used. By applying a QPSK modulation scheme, the performance of a given estimate is compared with traditional LS and LMMSE estimates in terms of channel estimation error and BER as a function of SNR level. With a correct understanding of the channel characteristics, we find that the proposed DNN-based estimation has good advantages in terms of minimizing the channel estimation error.

References

[1] T. Van Chien and E. Bjornson, “Massive MIMO communications,” in 5G Mobile Communications. Springer, 2017, pp. 77-116. [2] E.Bjornson, N.Jalden, M.Bengtsson, and B. Ottersten, “Optimality” properties distributed strategies, and measurement-based evaluation of coordinated multicell OFDMA transmission,” IEEE Trans. Signal Process., vol. 59, no. 12, pp. 6086–6101, 2011. [3] E. Dahlman, S. Parkvall, and J. Skold, 5G NR: The next generation wireless access technology. Academic Press, 2018. [4] S. Kay, Fundamentals of Statistical Signal Processing: Estimation Theory. Prentice Hall, 1993. [5] J. Zhu, O. Omomukuyo, R. Venkatesan, C. Li and O. A. Dobre, \"Rf-pilot phase noise compensation for long-haul coherent optical OFDM systems,\" 2015 IEEE 14th Canadian Workshop on Information Theory (CWIT), St. John\'s, NL, Canada,2015,pp. 127-130, doi: 10.1109/CWIT.2015.7255168. [6] T.JIACHIN Least-Squares Channel Estimation for Mobile OFDM Communication on Time Varying Frequency-Selective Fading Channels[A]. IEEE 2008,57(6) : 3538-3550. [7] COLERI S, ERGEN M, Puri A Channel estimation techniques based on pilot arrangement in OFDM systems[J]. IEEE Trans.on Broadcasting, 2008,48 (3) :223 229. [7] [8] J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Network, vol. 61, pp. 85–117, Jan. 2015. [9] J.-M. Kang, C.-J. Chun, and I.-M. Kim, “Deep-learning-based channel estimation for wireless energy transfer,” IEEE Communications Letters, vol. 22, no. 11, pp. 2310–2313, 2018. [8] Yi Gong and K.B Lataief, “Low rank channel estimation for space-time coded wide band OFDM systems”, IEEE Vehicular Technology Conference, vol. 2, pp. 772-776, 2001. 53 [9] D. Neumann, T. Wiese, and W. Utschick, “Learning the MMSE channel estimator,” IEEE Trans. Signal Process., vol. 66, no. 11, pp. 2905–2917, 2018. [10] T. O’Shea and J. Hoydis, “An introduction to deep learning for the physical layer,” IEEE Transactions on Cognitive Communications and Networking, vol. 3, no. 4, pp. 563–575, 2017. [11] ] E. Dahlman, S. Parkvall, and J. Skold, 5G NR: The next generation wireless access technology. Academic Press, 2018. [12] Jaipreet Kaur1, Hardeep Kaur2, Manit Sandhu “OFDM: Comparison with Multicarrier Techniques and Applications”, Volume 3, Issue 4 ,2016. [13] K. Mei, J. Liu, X. Zhang, and J. Wei, “Machine learning based channel estimation: A computational approach for universal channel conditions,” arXiv preprint arXiv:1911.03886, 2019.

Copyright

Copyright © 2023 Dr. M. VijayaLakshmi , B. Shashitha , B. Alekhya , Clementina Rithika , CH. Bhavana . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.