Digital Baseband Self-Interference Cancellation Using Fractionally-Spaced Finite-Impulse-Response Structure for Full-Duplex MIMO

In this work, a fractionally spaced transversal finite impulse response (FS-FIR) structure is proposed for digital baseband (BB) self-interference cancellation (SIC) in full-duplex (FD) multiple-input multiple-output (MIMO) wireless communications systems. We develop three algorithms based on Minimum-Mean-Square-Error (MMSE), Least-Square (LS), and Maximum-Likelihood (ML) criteria, to estimate the self-interference channel impulse response and optimize the FS-FIR-SIC coefficients. Performance of the proposed FS-FIR-SIC is evaluated under different oversampling rates and FIR spans. Illustrative results reveal that the FS-FIR structure greatly improves the SIC performance with an oversampling rate larger than 1 by avoiding possible aliasing due to sampling at the symbol rate. Increasing FIR filter span can further enhance the SIC performance, although the improvement is reduced at a large FIR span. Meanwhile, the ML-FS-FIR-SIC outperforms the LS-FS-FIR-SIC and MMSE-FS-FIR-SIC, especially in the presence of the intended received signal.