The DOA Estimation Of Spatial Stepped-frequency MIMO Array

Spatial stepped-frequency Multiple input multiple output(MIMO) array is proposed due to the excellent performance of MIMO array and the simplicity of stepped-frequency signal. Wideband signal is synthesized by combining many single signals at the receiving end, which could enhance the anti-interference ability and improve the resolution. Taking the direction of arrival(DOA) estimation as the main clue, two new DOA estimation methods—snapshot-split-iterative power spectrum reconstruction algorithm and iterative power spectrum reconstruction algorithm suited for the spatial stepped-frequency MIMO array--are proposed. The simulation results show the efficiency of the proposed algorithms.Firstly, DOA estimation algorithm such as Capon method, MUSIC(Multiple Signal Classification) algorithm and ESPRIT(Estimation of Signal Parameters via Rotation Invariance Techniques) algorithm are introduced. A modified snapshot-split-iterative power spectrum reconstruction algorithm is put forward. The power peaks of this algorithm is sharper than that of Capon’s, the ratio of signal-to-noise(SNR) is 3dB higher than that of Capon algorithm.Secondly, the MIMO array model is introduced. For single base MIMO array, the virtual array extension aperture and the maximal number of distinguishable target is studied. By extending the model of receiving array to MIMO array, the basic algorithms of the MIMO array are described. Simulation results show that the aperture of MIMO array is expanded and the estimation accuracy is higher than that of receiving array.Thirdly, the spatial stepped-frequency MIMO array model is established. Focusing algorithm of wideband signal DOA estimation is extended to this situation. Although it can estimate the DOA, its performance is not good. So, an iterative power spectrum reconstruction algorithm is proposed. Simulations show that the latter algorithm has better performance at low SNR and small snapshots. The SNR of iterative power spectrum reconstruction algorithm is 3-4dB higher than that of the focusing algorithm.