Low Complexity CSI Feedback Technique for FDD Massive MIMO Systems

This paper introduces an innovative and highly efficient approach for channel compression and feedback in frequency division duplex (FDD) massive multiple-input-multiple-output (m-MIMO) systems. Conventional channel state information (CSI) feedback often leads to excessive overhead because of the substantial number of channel coefficients inherent in m-MIMO setups. To address this, we propose a technique that leverages the Fast Fourier Transform (FFT) across the time-domain FDD downlink channel, performing channel matrix compression in the frequency domain. By selecting a fixed number of coefficients for feedback, we can attain a more efficient process of channel reconstruction at the transmitter, facilitated by the inherent efficiency of the FFT computation. This paper's fundamental contribution revolves around identifying potent channel coefficients - those with the highest channel gains. We achieve this through two distinct approaches: one that maintains row orthogonality and another that does not have this constraint. Simulation results affirm that the orthogonal approach significantly reduces errors. Notably, the computational complexity of our approach is less than 1% of that of the existing techniques. This substantial reduction in complexity underscores the potential of our method as a promising feedback solution for m-MIMO communication systems.