Title: Deep learning-based channel estimation in MIMO system for pilot decontamination

Authors: Gondhi Navabharat Reddy; C.V. Ravi Kumar

Addresses: SENSE, Embedded Technology, School of Electronics Engineering, Vellore Institute of Technology, Vellore – 632014, India ' SENSE, Embedded Technology, School of Electronics Engineering, Vellore Institute of Technology, Vellore – 632014, India

Abstract: The next-generation wireless communication system makes use of the upcoming technology known as massive multiple-input multiple-output (MIMO). The imprecise channel estimation leads to enormous error in communication, and this cause pilot contamination (PC), which is a major concern in the massive MIMO system. An efficient invasive shuffled shepherd optimisation (ISSO)-deep maxout-based channel estimate approach is proposed, which chooses the optimal channel in the massive MIMO system to avoid interference among nearby cells, reducing contamination of pilot sequences. Shuffled shepherd optimisation (SSO) and improved invasive weed optimisation (IWO) are combined to develop the proposed ISSO method. The deep maxout network is trained using the proposed ISSO technique, and the weight factors are computed using the loss function. The proposed channel estimation methodology produced bit error rate (BER) and mean square error (MSE) results for the Rayleigh channel of 0.0009 and 0.00070 and for the Rician channel of 0.0009 and 0.0007.

Keywords: channel estimation; pilot contamination; PC; deep learning; massive MIMO; multiple-input multiple-output; shuffled shepherd optimisation; SSO; invasive shuffled shepherd optimisation; ISSO; invasive weed optimisation; IWO.

DOI: 10.1504/IJAHUC.2023.134777

International Journal of Ad Hoc and Ubiquitous Computing, 2023 Vol.44 No.3, pp.148 - 166

Received: 11 Aug 2022
Accepted: 29 Mar 2023
Published online: 09 Nov 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article