Channel Estimation and Signal Detection in OFDM Systems using Deep Learning
Manoj Kumar Ojha
Manoj Kumar Ojha, Department of Electronics Engineering, Sanskriti University, (Uttar Pradesh), India.
Manuscript received on 05 October 2019 | Revised Manuscript received on 19 October 2019 | Manuscript Published on 26 December 2019 | PP: 176-179 | Volume-8 Issue-12S October 2019 | Retrieval Number: L105210812S19/2019©BEIESP | DOI: 10.35940/ijitee.L1052.10812S19
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This article presents “channel estimation and signal detection in OFDM systems by using deep learning”. OFDM stands for “Orthogonal Frequency Division Multiplexing”. This paper exploits end to end handling of wireless OFDM channels by deep learning. It is different from the existing OFDM receivers as it estimates the channel state information (CSI) explicitly and then estimated CSI is used to recover the transmitted symbols, thee proposed approach of deep learning implicitly estimates CSI and the transmitted symbols are recovered directly. The online transmitted data is directly recovered by the offline training a deep learning model using simulation based channel statistics generated data for addressing channel distortion. The performance comparable to “minimum mean square error” (MSME) estimator with transmitted symbols is detected by using deep learning based channel distortion. Using fewer number of pilots, omitting cyclic prefix and in the existence of nonlinear clipping noise, the approach of deep learning is more robust as compared to traditional methods.
Keywords: OFDM, MSME, Channel State Information, DNN Model.
Scope of the Article: Deep Learning