Direction finding in the presence of fully correlated specular multipath

A novel nonlinear estimation technique for the problem of low-angle direction finding of a single source in the presence of fully correlated specular multipath is presented. The formulation used is that of maximum likelihood. The solution is general in that the array geometry, element types, and number of snapshots (which may be as little as one) are arbitrary, and the complex reflection coefficient of the reflecting surface is unknown. Both the symmetric and nonsymmetric scenarios are addressed. For the case where only a single snapshot of data is available, the estimation scheme is shown to reduce to the ML solution for the less-than-fully correlated scenario. Computer simulation verifies the utility of the proposed estimation technique