Variable step-size blind equalization optimization algorithm for Faster-than-Nyquist coherent FSO-MIMO systems
SUN Qinxue
CAO Minghua
ZHANG Xia
HUANG Min
ZHANG Lingyun
BAO Shan
ZHU Kai
Abstract：The utilization of Faster-than-Nyquist(FTN)technology in Multiple-Input Multiple-Output(MIMO)Free Space Optical(FSO)communication systems introduces inter-symbol interference and sig-nal distortion caused by atmospheric turbulence effects,necessitating the optimization of equalization algo-rithms to ensure communication system reliability.By incorporating a decision feedback structure into the Constant Modulus Algorithm(CMA),the equalizer exhibits enhanced resistance against ISI.Based on ob-served variations in the error vector function of the CMA equalizer,a variable step size decision feedback structured blind equalization optimization algorithm is designed by following an Exponentiated Weibull(EW)distribution model for determining the step size coefficient.Simulation results demonstrate that in-creasing the number of receiving antennas compared to transmitting antennas yields higher gains in Bit Er-ror Rate(BER)and Error Vector Magnitude(EVM),along with improved phase noise recovery.Under FTN acceleration parameter of 0.9 and moderate turbulence conditions,employing a 2×4 antenna configu-ration with the proposed algorithm enhances BER performance by 36.39%and EVM performance by 4.96%.By controlling the acceleration parameter within a certain range,effective recovery of FTN ISI is a-chieved using the proposed algorithm.Compared to fixed step size algorithms,although exhibiting in-creased computational complexity,our proposed algorithm demonstrates improved BER and EVM per-formance.Moreover,it demonstrates superior convergence capabilities for equalizers.
Keywords：free space optical communicationFaster-than-NyquistMultiple-Input Multiple-Outputblind equalization algorithm
Publication Date：2025-08-01
Online Publishing Date：2026-08-28(First online date of this platform, not the publication date of the document)
Pages：8( 527-534 )