Waveform covariance matrix design using Fourier series coefficients

Abstract

Multiple-input multiple-output (MIMO) radars may outperform other radar systems such as phased array radars, in terms of higher resolution, better detection probability in the presence of interferences, better parameter identifiability and more flexibility in beampattern design. Waveform covariance matrix design, because of its role in the beampattern synthesis process, is one of the most important problems in MIMO radar systems. In this study, the authors have proposed a closed-form solution based on Fourier series coefficients to design a covariance matrix. The resulting covariance matrix fulfils the practical constraints, i.e. positive semi-definiteness and the uniform elemental power constraint. It also provides performance similar to that of iterative methods, while requires lower computation time and provides better mean square error with respect to other existing closed-form methods. Eigenvalue decomposition is also utilised to convert the possible resulted pseudo-covariance matrices (pseudo-CM), which are not guaranteed to be positive semidefinite, into a covariance matrix. Simulation results show the performance of the proposed method.