DOA and Range Estimation of Multiple Sources under the Wideband Assumption

In this paper, two novel channel parameter estimation algorithms are proposed under the "wideband assumption" (WBA), where a wavefront varies significantly when traversing through the sensors of an array. The first is a covariance-based algorithm that utilizes the cross-covariance matrix between two subvectors of the received signal vector and its singular value decomposition to reconstruct the parameter-dependent signal subspace. The second employs the rotation of the array reference point to transform the WBA problem to its "narrowband assumption" (NBA) counterpart so that estimation techniques under the NBA are readily applicable. Through computer simulation studies, the two proposed approaches are shown to successfully estimate the channel parameters under the WBA with outstanding accuracy in terms of the root mean squared error.