Reactively Loaded Array Pattern Synthesis as a Quadratically Constrained Quadratic Program

TitleReactively Loaded Array Pattern Synthesis as a Quadratically Constrained Quadratic Program
Publication TypeJournal Article
Year of Publication2015
AuthorsCórcoles, J.
JournalAntennas and Propagation, IEEE Transactions on
Date PublishedNov
KeywordsAntenna arrays, Antenna pattern synthesis, antenna radiation patterns, Arrays, dipole antenna arrays, dipole array, Gain, Loaded antennas, optimization, Optimization methods, Ports (Computers), QCQP, quadratic programming, quadratically constrained quadratic program, reactively loaded array pattern synthesis, reconfigurable antennas, S-matrix theory, scattering matrix, semidefinite (SD) programming, semidefinite programming, semidefinite relaxation, sidelobe level, sidelobe supression
AbstractThe pattern synthesis problem of maximizing the gain at a scanning direction subject to a maximum sidelobe level for a reactively loaded antenna array is formulated as a quadratically constrained quadratic program (QCQP). This formulation is derived from the scattering matrix and the active element patterns of the array considering both fed and loaded antennas as ports. Since the resulting QCQP is nonconvex, the loads are computed from the solution of the semidefinite relaxation of the problem by forcing the feasibility of the constraints in a meaningful physical sense, i.e., by explicitly imposing passive or purely reactive loads. Numerical results of dipole arrays with various configurations are presented and compared to previous works.