A new AOA estimation algorithm of wideband chirp signals

A novel time-frequency approach to the direction finding of a wideband chirp source is proposed. Our technique to construct virtual sensors employs the characteristic that the direction vector of a wideband chirp signal is an arithmetic progression. By the time-frequency selectivity of the Wigner-Ville distribution, a new data model is constructed to estimate the 2D AOA of each signal one by one. A Radon transformation is used to suppress noise and the cross-terms of signals. The computation cost of the method is low because no focusing and interpolating is needed, even no singular value decomposition is needed. Simulation results illustrate the performance of the method.