DOA estimation of coherent and incoherent targets based on monostatic co-prime MIMO array

For the traditional DOA estimation of coherent and incoherent targets based on monostatic (massive) uniform dense array, the number of resolvable targets is limited to the number of physical sensors. To exceed the limitation, it is desirable to have sparse transmitting and receiving arrays. In this study, we consider the monostatic co-prime MIMO array with N sparse transmitters and 2M-1 sparse receivers in the DOA estimation of mixed coherent and incoherent targets. The sparse co-prime MIMO array with O( M + N ) physical sensors generates a non-redundant and uniform sub-coarray with O(MN) contiguous sensors in the sum co-array. Based on the defined wide-sense or narrow-sense sum co-array equivalence, we can obtain different configurations of virtual MIMO arrays with O(MN) contiguous virtual sensors, and then construct the corresponding virtual data matrices, which provides different tradeoffs between the number of resolvable targets and the maximum number of mutually coherent targets that can be resolved. On the basis of the virtual data matrix and the conventional DOA estimation approaches such as MUSIC, O(MN) mixed coherent and incoherent targets can be resolved only with O( M + N ) physical sensors, namely the number of resolvable targets exceeds the limitation of the number of physical sensors. Furthermore, the application of two additional operation frequencies extends the contiguous sub-coarray accompanied with the improvement of degree-of-freedom for more resolvable coherent and incoherent targets. Finally, simulation results demonstrate the effectiveness of the proposed DOA estimation method.