Citation: Kang, X.-F.; Liu, Z.-H.; Yao,
M. Deep Learning for Joint Pilot
Design and Channel Estimation in
MIMO-OFDM Systems. Sensors 2022,
22, 4188. https://doi.org/10.3390/
s22114188
Academic Editors:
Panagiotis Sarigiannidis,
Thomas Lagkas,
Alexandros-Apostolos Boulogeorgos,
Vasileios Argyriou and
Pantelis Angelidis
Received: 16 April 2022
Accepted: 28 May 2022
Published: 31 May 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Deep Learning for Joint Pilot Design and Channel Estimation
in MIMO-OFDM Systems
Xiao-Fei Kang * , Zi-Hui Liu and Meng Yao
Affiliation College of Communication and Information Engineering, Xi’an University of Science and Technology,
Xi’an 710054, China; zh_liu0619@163.com (Z.-H.L.); shirley980912@126.com (M.Y.)
* Correspondence: kangxiaofei@sina.com; Tel.: +86-13572211700
Abstract:
In MIMO-OFDM systems, pilot design and estimation algorithm jointly determine the
reliability and effectiveness of pilot-based channel estimation methods. In order to improve the
channel estimation accuracy with less pilot overhead, a deep learning scheme for joint pilot design
and channel estimation is proposed. This new hybrid network structure is named CAGAN, which is
composed of a concrete autoencoder (concrete AE) and a conditional generative adversarial network
(cGAN). We first use concrete AE to find and select the most informative position in the time-
frequency grid to achieve pilot optimization design and then input the optimized pilots to cGAN to
complete channel estimation. Simulation experiments show that the CAGAN scheme outperforms
the traditional LS and MMSE estimation methods with fewer pilots, and has good robustness to
environmental noise.
Keywords:
autoencoder; channel estimation; conditional generative adversarial network; MIMO-
OFDM; pilot design
1. Introduction
MIMO-OFDM technology can effectively utilize resources in three dimensions of
time, frequency, and space to greatly improve the spectral efficiency, power efficiency,
and transmission rate of the system. It has become the core technology of broadband
wireless communication systems. Acquiring accurate channel state information (CSI)
through channel estimation is a prerequisite for realizing the huge potential of MIMO-
OFDM technology, and it is also an important basis for realizing precoding, resource
allocation, signal detection, indoor positioning, physical layer security, and so on [1].
Depending on whether pilot signals are needed, channel estimation can be classified
as blind channel estimation and pilot-based channel estimation. Blind channel estimation
does not require pilot signals, which perform channel estimation through the second-order
or high-order statistical information of the received signal. In [
2
], a blind channel estimation
algorithm is proposed that used the statistical information on the average power of the
received signal to convert the average power of the received signal into a quadratic equation
including the channel gain. The user terminal can estimate the massive MIMO downlink
channel gain without any downlink pilot resources, but the algorithm is only suitable
for time division duplex (TDD) systems and it does not take advantage of the channel
sparse characteristics and the estimation accuracy performed poorly. A blind estimation
algorithm based on expectation maximization (EM) is proposed for massive MIMO systems
in [
3
], which utilized the sparse characteristics of the channel in the angular domain to
improve the channel estimation accuracy but requires a large amount of computation. Blind
channel estimation has the advantages of less prior information and high spectral efficiency,
but its application is seriously limited to low channel estimation accuracy, high complexity,
and poor real-time performance. The pilot-based channel estimation method is to insert
pilot symbols into the transmitted signals, and the receiver implements channel estimation
Sensors 2022, 22, 4188. https://doi.org/10.3390/s22114188 https://www.mdpi.com/journal/sensors
