The Real Implementation of ANFIS Channel Equalizer on the System of Software-Defined Radio

Radek Martinek,Jan Zidek
DOI: https://doi.org/10.1080/03772063.2014.914698
IF: 1.8768
2014-03-04
IETE Journal of Research
Abstract:This paper deals with the usage of a combination of fuzzy system and artificial techniques, which are called adaptive neuro fuzzy inference system (ANFIS), in order to minimize the distance between the signal and SNR transmitting channel noise and then reduce the error rate of bit error rate (BER) transmission. The authors are focusing on a real implementation of ANFIS channel equalizer on software-defined radio (SDR) system working on PCI eXtensions for instrumentation (PXI) platform. This sophisticated modular measuring system consists of a vector signal generator RF VSG NI PXI-5670 and a vector signal analyzer RF VSA NI PXI-5661. Experimental results suggest that researched ANFIS equalizer embodies better BER values in comparison to commercially most common equalizers of the least mean square algorithm group. Moreover, the conducted experiments show that the usage of the SDR conception is very suitable for testing new principles in channel equalization field.
telecommunications,engineering, electrical & electronic
What problem does this paper attempt to address?