Potential Optimum Performance of Nonlinearly Distorted MIMO-SVD Systems | IEEE Journals & Magazine | IEEE Xplore

Potential Optimum Performance of Nonlinearly Distorted MIMO-SVD Systems


Abstract:

Multiple-input, multiple-output (MIMO) systems are known to provide spatial multiplexing capabilities. For this reason, they are very important in 5G and will certainly h...Show More

Abstract:

Multiple-input, multiple-output (MIMO) systems are known to provide spatial multiplexing capabilities. For this reason, they are very important in 5G and will certainly have a key role in beyond 5G (B5G) systems. However, the precoding operation employed by many MIMO transmitters to compensate for the channel effects and/or separate users increases the envelope fluctuations of the transmitted signals, giving rise to a trade-off between linearity and energy efficiency. Commonly, nonlinear distortion is considered additional noise that severely affects performance. However, since the nonlinear distortion is a function of the transmitted signals, studies have demonstrated the potential of taking it as useful information for detection purposes. In this work, we followed this vision and consider the maximum likelihood (ML) detection to exploit the information of the nonlinear distortion term. We consider MIMO singular value decomposition (MIMO-SVD) systems impaired by bandpass nonlinearities and we derive the potential ML performance in independent and identically distributed (iid) Rayleigh fading channels through a theoretical study of the distribution of the normalized squared Euclidean distance (SED) between nonlinearly distorted quadrature amplitude modulation (QAM) signals. It is demonstrated that by using the nonlinear distortion as useful information, the ML detection of nonlinear MIMO-SVD not only can avoid the degradation associated with the conventional detection of nonlinear MIMO-SVD, but can even present performance gains over linear MIMO-SVD.
Published in: IEEE Transactions on Vehicular Technology ( Volume: 72, Issue: 5, May 2023)
Page(s): 6142 - 6153
Date of Publication: 22 December 2022

ISSN Information:

Funding Agency:


References

References is not available for this document.