Enhanced Angle-of-Arrival And Polarization Parameter Estimation Using Localized Hybrid Dual-polarized Arrays

Compared to commercial wireless communication systems working in the frequency band below6 GHz,which have the problem of spectrum resource constraint,millimeter wave(mm Wave)has great available bandwidth in the operating band of 30 GHz to 300 GHz.This advantage enables mm Wave communication to be a promising development direction for future communications.A large number of small antennas can be integrated in a limited space by using the array gain to compensate for the millimeter wave transmission loss when combined with large scale multiple-input multiple-output(MIMO)technology,which overcomes the shortage of communication spectrum resources.Due to the limited scattering and small diffraction characteristics of millimeter waves,lineof-sight(LOS)propagation dominates the mm Wave channel.Therefore,the estimation of the AoA and polarization parameters of mm Wave signals becomes very significant and forms the basis of relevant signal processing techniques in mm Wave systems.Focusing on dual-polarization antenna arrays under mm Wave large scale MIMO systems in this paper,an enhanced AoA and polarization estimation algorithm using the localized hybrid dualpolarization array in polarized millimeter-wave signals is proposed.Meanwhile,the performance metrics such as estimation accuracy are theoretically analyzed,such as the mean squared error(MSE),mean squared error low bound(MSELB)of the AoA information,MSE of the polarization parameter and signal-to-noise ratio(SNR)of the output of combined signals.The main contributions are as follows.(1)An enhanced AoA estimation algorithm for localized dual-polarized arrays using a specific phase shifter design is proposed.It is found that the cross correlations between adjacent subarrays all have the same sign,except for the signal at the maximum amplitude,which has a different sign.The maximum power of the received signal is acquired by calibrating the sign of the cross correlation between adjacent subarrays,and then the AoA information is extracted by coherent superposition,which effectively improves the calibration of the differential signal and thus the estimation accuracy of the phase shift between adjacent subarrays.The initial AoA estimation is obtained using the estimated phase shift,and then in turn the phase shift estimation is updated,thus improving the accuracy of the AoA estimation.In addition,in order to evaluate the estimation performance,the MSELBs for the AoA information are derived.(2)Based on the AoA acquired as above,a method based on the estimation of the cross correlation to power ratio is further proposed to improve the accuracy of the polarization parameter estimation.The method exploits two axis dipoles to receive the incoming signal and calculates the polarization parameter by the ratio of the cross correlation between adjacent subarrays to the total received power of the signal.By analyzing the average SNR of the polarization parameter,it can be demonstrated that the average SNR of the incoming signal collected using two axis dipoles is larger than that using only one axis dipole,leading to a higher estimation accuracy.Moreover,unlike the method proposed in [57],the polarization estimation method proposed in this paper calculates the cross correlation between adjacent subarrays and summation before the digital beamforming instead of after the digital beamforming.Both theoretical analysis and simulation results show that calculating the ratio of the cross correlation between adjacent subarrays to the signal power before the digital beamforming has a higher average SNR and thus a higher accuracy of the polarization parameter estimation.