DOA estimation with extended sparse and parametric approach in multi-carrier MIMO HFSWR

Abstract

To obtain a higher angle resolution of multiple-in multiple-out high-frequency surface wave radar (MIMO HFSWR) for direction of arrival (DOA) estimation with a limited number of antenna sensors, multiple working frequencies are proposed to enlarge the aperture of the virtual array of the MIMO HFSWR. The scenario that all targets have the identical reflection at all working frequencies is studied, which permits the abstraction of a virtual received data vector by using all frequencies. This virtual data vector can be taken as the measurement of a virtual non-uniform linear array (VNLA) with a single reference working frequency. To extend the sparse and parametric approach (SPA) to the VNLA for DOA estimation, the manifold separation technique is utilised to decompose the array steering vector of the VNLA into two different parts, one is a characteristic matrix that is related to the array itself. The other is a Vandermonde vector that contains the DOAs of the targets. Then the authors use the Vandermonde structure to develop a SPA-liked method for the DOA estimation. Simulation results are provided to confirm the validity of the proposed method.

Keywords

• array signal processing
• direction-of-arrival estimation
• matrix algebra
• mimo radar
• radar antennas
• linear antenna arrays
• vectors
• antenna sensors
• multiple working frequencies
• virtual array
• virtual received data vector
• nonuniform linear array
• vnla
• single reference working frequency
• doa estimation
• array steering vector
• vandermonde vector
• extended sparse approach
• multicarrier mimo hfswr
• high-frequency surface wave radar
• direction of arrival estimation
• angle resolution
• multiple-in multiple-out high-frequency surface wave radar
• sparse and parametric approach