A responsive compressive sensing based channel estimation algorithm using curve fitting and machine learning. (27th May 2022)

• Record Type: Journal Article
• Authors: Munshi, Ami; Unnikrishnan, Srija
• Is Part Of: International journal of systems, control and communications. Volume 13:Number 3(2022)
• Page Start: 241
• Page End: 252
• Publication Date: 2022-05-27
• Subjects: channel estimation -- bit error rate -- BER -- signal to noise ratio -- SNR -- MIMO -- orthogonal frequency division multiplexing -- OFDM -- compressive sensing -- sparsity -- machine learning -- random forest -- peak to average power ratio -- PAPR -- least square channel estimation; Computer networks -- Periodicals; Intelligent control systems -- Periodicals; Systems engineering -- Periodicals; 004.605
• Languages: English
• ISSNs: 1755-9340

Abstract

It is observed that compressive sensing based channel estimation in orthogonal frequency division multiplexing (OFDM) system with more number of subcarriers can achieve a good reconstruction of transmitted signal at the receiver side even if the channel is very noisy. However, with increase in the number of subcarriers, peak to average power ratio (PAPR) also increases. In this paper, we propose a responsive compressive sensing based channel estimation algorithm which will estimate the minimum signal to noise ratio (SNR) of the channel when the pilot signal is transmitted based on parameters such as the number of subcarriers and the total number of channel coefficients needed to attain negligible bit error rate (BER). Once the minimum channel SNR is estimated, the algorithm will put forward the optimum number of subcarriers to be employed in the MIMO-OFDM system to optimally reconstruct the transmitted data at the receiver side.