Performance of channel estimation schemes in the presence of gaussian mixture model
Visualitza/Obre
10.15676/ijeei.2021.13.3.10
Inclou dades d'ús des de 2022
Cita com:
hdl:2117/365851
Tipus de documentArticle
Data publicació2021-09-01
Condicions d'accésAccés obert
Llevat que s'hi indiqui el contrari, els
continguts d'aquesta obra estan subjectes a la llicència de Creative Commons
:
Reconeixement-SenseObraDerivada 4.0 Internacional
Abstract
Channel estimation (CE) plays a crucial role in establishing a wireless link, specifically at the receiver node. Most of the receivers that estimate the channel is in the presence of AWGN. However, these schemes perform expressively worse when the impulsive noise is added in AWGN which is introduced by manmade sources (pressure cooker, motorbike, electric supply) as well as natural noises (earthquakes and thundering). The major contribution of this research is to analyze the channel estimation schemes in the Gaussian mixture model (GMM) environment. The performance of channel estimation schemes has been compared in terms of mean square error (MSE) and bit error rate (BER). Four channel estimation schemes e.g., MMSE, DFT, correlation- based methods like Gauss-Seidel (GS) and Successive Over- Relaxation (SOR), are studied and analyzed. The study reveals that the correlation scheme based on the method of SOR is more effective as compared to the methods of DFT, MMSE and GS because of faster convergence rate along with the minimum number of iteration. SOR shows sustainable results up to the probability of an impulsive element of 5 Percent.
CitacióBai, S. [et al.]. Performance of channel estimation schemes in the presence of gaussian mixture model. "International Journal on Electrical Engineering and Informatics", 1 Setembre 2021, vol. 13, núm. 3, p. 653-665.
ISSN2085-6830
Versió de l'editorhttp://ijeei.org/docs-5026438976162dbd50b3ef.pdf
Col·leccions
Fitxers | Descripció | Mida | Format | Visualitza |
---|---|---|---|---|
docs-5026438976162dbd50b3ef.pdf | Article | 1,503Mb | Visualitza/Obre |