DOA Estimation Based On New Nested Array

The array aperture and degree of freedom(DOF)important factors that affect the signal processing accuracy.Increasing the effective array aperture and the DOF is able to improve the accuracy of the spatial signal processing.Usually,many classic direction-of-arrival(DOA)estimation algorithms use common array structures such as uniform linear array(ULA)and uniform circular array(UCA).However,these structures suffer from the limitation of array aperture.Therefore,how to expand the array aperture and increase the DOF by designing new array structures has been attracted much attention.In this thesis,a new array structure is studied.By making use of the virtual array generated by this new array structure,the DOA estimation method based on Khatri-Rao(KR)subspace is also studied.In this thesis,a new nested array structure is proposed.It can effectively expand the array aperture and improve the DOF,and thus improve the accuracy of parameter estimation and the maximum number of signal sources that can be detected.Specifically,the structure consists of a uniform linear array and a non-uniform linear array,and all the sensors are in the same line.The whole array is composed of L(L = M + N)physical sensors,i.e.,a ULA composed of M sensors with inter-element spacing d,and a non-uniform linear array(NULA)composed of N sensors with inter-element spacing {2M,2M +1,,2M +N-2}d.The spacing between ULA and NULA is Md.The array structure has a continuous virtual array element with a length of 8M-1 and a range of [-(4M-1),4M-1].Through the design of multiple sets of quasi-stationary signal simulation experiments,the effectiveness of the array structure in expanding the array aperture is fully verified.Besides,in order to reduce the computational complexity of the algorithm,this thesis uses the proposed array structure to achieve low complexity DOA estimation under quasi-stationary signal conditions.This method is based on the principle of ESPRIT method and uses the nature of Khatri-Rao product to obtain KR subspace.In the subarray division,because the virtual array of the proposed new nested array is not completely continuous,this thesis provides two ways of subarray division.By solving the phase difference between the two subarrays,the estimation angle is obtained,and the underdetermined DOA estimation under quasi-stationary signal is realized.Since the spectral search method is avoided to estimate the angle,the computational complexity of the estimation process is greatly reduced.